{ "schemaVersion": "deck-intelligence.v1", "id": "019dd923-5e88-73ef-bd5c-f341d4394195", "slug": "46f66c49fd159048", "title": "2018 Air Street Capital The State of AI Report 2018", "subtitle": "Air Street Capital", "author": "Air Street Capital", "pageCount": 156, "kind": "consulting-deck", "orientation": "landscape", "aspectRatio": 1.777, "document": { "id": "019dd923-5e88-73ef-bd5c-f341d4394195", "slug": "46f66c49fd159048", "title": "2018 Air Street Capital The State of AI Report 2018", "rawTitle": "2018 Air Street Capital The State of AI Report 2018", "authorId": "AirStreetCapital", "authorName": "Air Street Capital", "authorDisplay": "Air Street Capital", "kind": { "slug": "consulting-deck", "label": "Consulting deck" }, "sourceType": { "slug": "vc_research", "label": "VC research" }, "fileName": "2018-Air-Street-Capital-The-State-of-AI-Report-2018.pdf", "sourceUrl": "https://stateof.ai", "sourceDate": "2018-10-01 00:00:00", "presentationDate": null, "ingestedAt": "2026-04-25 11:55:48+00", "status": "ready", "pageCount": 156, "orientation": "landscape", "aspectRatio": 1.777, "targetCompany": null, "targetTicker": null, "metadata": { "local_root": "/tmp/legacy-images/46f66c49fd159048", "origin_corpus": "consulting" } }, "coverage": { "pageCount": 156, "pagesIndexed": 156, "pagesWithImages": 156, "pagesWithMetadata": 156, "imageCoveragePct": 100, "metadataCoveragePct": 100, "componentCount": 944, "chartCount": 92, "metricSlides": 0, "toolMatches": 256, "documentToolMatches": 0, "frameworkMatches": 16, "arcMatches": 2, "beatMatches": 8, "loopMatches": 20, "patternMatches": 0, "storymakersMatch": true, "scoreAvailable": true, "reviewAvailable": true, "activistThesisAvailable": false, "pitchdeckMetadataAvailable": false, "pitchdeckProfileAvailable": false }, "blocks": [], "matches": { "arcs": [ { "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "evidence": "Six-section state-of-industry report walks Facts (Research/Talent pp.6-69) -> Implications (Industry/Politics pp.70-150) -> Action (Predictions pp.151-152).", "isPrimary": true, "confidence": 78, "extraction": { "at": "2026-04-29 13:07:28.977035+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "evidence": "Each section peels back one layer of AI: surface tech -> who builds it -> where deployed -> societal core -> implications.", "isPrimary": false, "confidence": 55, "extraction": { "at": "2026-04-29 13:07:28.977035+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "beats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 156, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "from": 151, "name": "The Action (Now What)", "beatId": "019dd95a-0682-776c-8e34-b4730e3b2f0d", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Action (Now What)", "slug": "triple-take-the-action-now-what", "status": "active", "canonId": "019dd9b8-0a5e-735a-bf53-3a1ba629b7aa", "version": 1, "description": null }, "evidence": "Section 5 '8 predictions for the next 12 months' and conclusion.", "position": 3, "isPrimaryArc": true, "parentBeatType": { "name": "Resolution", "slug": "resolution", "status": "active", "canonId": "019dd9b8-045d-703b-8dd7-e92aba3ba91b", "version": 1, "description": "How the deck ends. Required in 19/20 arcs. The exception (Sparkline) ends with new bliss which is functionally a resolution." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 9, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "from": 6, "name": "Surface Observation", "beatId": "019dd95a-0682-776c-8e34-b9affb23a9a4", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Surface Observation", "slug": "onion-surface-observation", "status": "active", "canonId": "019dd9b8-0b39-73c0-b415-6c4cf34cc343", "version": 1, "description": null }, "evidence": "Transfer learning headline result on skin cancer.", "position": 1, "isPrimaryArc": false, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 55 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 }, { "to": 156, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null, "categoryName": "Analysis & Information", "categorySlug": "analysis-information" }, "from": 151, "name": "Implications", "beatId": "019dd95a-0682-776c-8e34-cb622c4afe44", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Implications", "slug": "onion-implications", "status": "active", "canonId": "019dd9b8-0dda-7730-893d-deaac7e9c295", "version": 1, "description": null }, "evidence": "Predictions for next 12 months.", "position": 5, "isPrimaryArc": false, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 9, "from": 7, "loop": { "name": "03_aha_moment", "slug": "03-aha-moment", "status": "active", "bestFor": "Data-heavy sections, research findings, analytical arguments", "canonId": "019dd956-66f4-7668-ab25-cec9c996c0bc", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Problem/Question -> What the Data Shows -> The Insight", "description": "Establish a tension, present data, then deliver the insight that changes everything", "familyLabel": null, "categoryName": "Narrative", "categorySlug": "narrative" }, "matchId": "019dd95a-07fe-70ce-8d3c-46cbef5e2b62", "evidence": "Defines transfer learning then reveals model beats 21 dermatologists with ROC charts.", "position": 1, "objective": "Show transfer learning enables superhuman skin-cancer detection", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 35, "from": 30, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null, "categoryName": "Analysis", "categorySlug": "analysis" }, "matchId": "019dd95a-07fe-70ce-8d3c-514142d7981d", "evidence": "Big picture (CV detects nouns) -> focus on verbs/common sense -> specific case (multi-viewpoint scenes).", "position": 4, "objective": "Zoom from object detection to full scene understanding", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 39, "from": 36, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-55bec48734a5", "evidence": "AlphaZero, OpenAI Dota2, world-model RL — three parallel evidence pieces of RL progress.", "position": 5, "objective": "Show RL agents now beat human experts across domains", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 43, "from": 40, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null, "categoryName": "Analysis", "categorySlug": "analysis" }, "matchId": "019dd95a-07fe-70ce-8d3c-5930175dc2d9", "evidence": "Big picture (fairness) -> specific examples (Turkish gender, racial) -> 5-type allocation bias table.", "position": 6, "objective": "Diagnose ML bias by isolating its sources", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 50, "from": 44, "loop": { "name": "10_iceberg", "slug": "10-iceberg", "status": "active", "bestFor": "Consulting, complex problem solving, organizational change", "canonId": "019dd956-69f4-738c-852a-643798da0899", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause", "description": "Reveal that the visible problem is merely a symptom of a deeper root cause", "familyLabel": null, "categoryName": "Analysis", "categorySlug": "analysis" }, "matchId": "019dd95a-07fe-70ce-8d3c-5ed4c9b352e3", "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 55, "from": 51, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-619c71c15583", "evidence": "Architecture iteration -> AutoML automating engineers -> federated learning -> consensus mechanism.", "position": 8, "objective": "Catalogue frontier research trends (AutoML, federated)", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 67, "from": 65, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null, "categoryName": "Urgency", "categorySlug": "urgency" }, "matchId": "019dd95a-07fe-70ce-8d3c-689ec650b673", "evidence": "Salaries climb -> $300-500k base anecdotes -> $120M Google self-driving boss as tipping point.", "position": 10, "objective": "Quantify escalating talent compensation arms race", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 73, "from": 71, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-6ee015b0dd0a", "evidence": "Patent filings concentrated -> R&D spend leadership -> cloud APIs -> TensorFlow framework war.", "position": 11, "objective": "Show GAFAMBAT capture AI infrastructure layer", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 80, "from": 74, "loop": { "name": "14_day_in_life", "slug": "14-day-in-life", "status": "active", "bestFor": "Product demos, UX reviews, investor pitches focusing on pain points", "canonId": "019dd956-6b65-707f-9b14-89beaaab81fa", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The User's Struggle -> The Intervention (Your Solution) -> The Resolved State", "description": "Humanize abstract data by walking through a single user's struggle and resolution", "familyLabel": null, "categoryName": "Narrative", "categorySlug": "narrative" }, "matchId": "019dd95a-07fe-70ce-8d3c-735f28c87032", "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 85, "from": 81, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null, "categoryName": "Analysis", "categorySlug": "analysis" }, "matchId": "019dd95a-07fe-70ce-8d3c-7442641e422c", "evidence": "China surveillance -> Project Maven -> Cambridge Analytica -> privacy preservation tooling.", "position": 13, "objective": "Zoom into surveillance/privacy as the dark side of AI", "confidence": 68, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 129, "from": 123, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null, "categoryName": "Logical Reasoning", "categorySlug": "logical-reasoning" }, "matchId": "019dd95a-07fe-70ce-8d3c-823fbc64276b", "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null, "categoryName": "Urgency", "categorySlug": "urgency" }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 143, "from": 139, "loop": { "name": "15_why_now", "slug": "15-why-now", "status": "active", "bestFor": "Sales pitches, fundraising, requesting immediate budget approval", "canonId": "019dd956-6c00-70cd-bd42-6bacc4af726f", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Context (Trends) -> The Trigger Event -> The Window of Opportunity", "description": "Create temporal urgency by proving that the window of opportunity is opening or closing", "familyLabel": null, "categoryName": "Urgency", "categorySlug": "urgency" }, "matchId": "019dd95a-07fe-70ce-8d3c-8b5e319977cd", "evidence": "Five contiguous slides with $150B China, €1.5B France, $2B Korea, €20B EU, £603M UK strategies.", "position": 18, "objective": "Argue national AI strategies create urgent geopolitical window", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 149, "from": 144, "loop": { "name": "04_tale_two_worlds", "slug": "04-tale-two-worlds", "status": "active", "bestFor": "Competitive analysis, benchmarking, case for change", "canonId": "019dd956-674f-750b-b15d-f082e2717f4e", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Current State -> Desired State / Benchmark -> The Gap & Implication", "description": "Show the gap between two states to drive urgency or highlight opportunity", "familyLabel": null, "categoryName": "Comparison", "categorySlug": "comparison" }, "matchId": "019dd95a-07fe-70ce-8d3c-8fdb0c2bbdc3", "evidence": "Side-by-side US/China table on every dimension -> CFIUS scrutiny -> China import gap -> chip industry size.", "position": 19, "objective": "Compare US vs China AI capability head to head", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } }, { "to": 152, "from": 152, "loop": { "name": "35_scenario_fork", "slug": "35-scenario-fork", "status": "active", "bestFor": "Strategic planning, risk analysis, board presentations", "canonId": "019dd956-7430-71ea-ac7d-1630063fb962", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Decision Point -> Scenario A (If X) -> Scenario B (If Y) -> Scenario C (If Z) -> Recommendation", "description": "Present multiple futures based on different assumptions or decisions", "familyLabel": null, "categoryName": "Consulting Framework", "categorySlug": "consulting-framework" }, "matchId": "019dd95a-07fe-70ce-8d3c-90b4058f80f5", "evidence": "Numbered list of 8 distinct 'if X happens' predictions for next 12 months.", "position": 20, "objective": "Project eight forward scenarios for the year ahead", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "tools": [ { "tool": { "name": "Loss Aversion", "slug": "loss-aversion", "status": "active", "bestFor": null, "canonId": "019dd956-9761-738f-8fd4-558a63ea744a", "version": 1, "bodyDocId": null, "description": "Losses are felt 2x more strongly than equivalent gains - frame stakes as potential loss", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "block", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-728aae2ca6b4", "evidence": "Frames cost as constraint that risks excluding players.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 55, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 20, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "MECE Principle", "slug": "mece-principle", "status": "active", "bestFor": null, "canonId": "019dd956-7c60-71a8-8758-c79d914489c9", "version": 1, "bodyDocId": null, "description": "Mutually Exclusive, Collectively Exhaustive - ensuring complete, non-overlapping structure", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "block", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-e5bd648d5a3a", "evidence": "5-column matrix presents bias categories as MECE table.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Curiosity Gap", "slug": "curiosity-gap", "status": "active", "bestFor": null, "canonId": "019dd956-9703-74ef-8b85-96dfd8525025", "version": 1, "bodyDocId": null, "description": "People seek missing information - create open questions that demand resolution", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "block", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-eea570fe4a8e", "evidence": "Frames problem ('understanding them is hard') to set up later reveal.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 44, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Pareto Principle (80/20)", "slug": "pareto-principle-80-20", "status": "active", "bestFor": null, "canonId": "019dd956-8cc3-75ba-a35d-d6575b56c45d", "version": 1, "bodyDocId": null, "description": "80% of effects come from 20% of causes - focus on high-impact items", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "block", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-480ae1011fc7", "evidence": "Top-11 institutions ranked — concentration story.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 64, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Problem-Agitate-Solve (PAS)", "slug": "problem-agitate-solve-pas", "status": "active", "bestFor": null, "canonId": "019dd956-869f-7564-8d42-de54ed29ab90", "version": 1, "bodyDocId": null, "description": "Identify problem, agitate the pain, present the solution", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "block", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-8411d885a360", "evidence": "Why-now problem framing opens a P-A-S sequence for healthcare.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 74, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Curiosity Gap", "slug": "curiosity-gap", "status": "active", "bestFor": null, "canonId": "019dd956-9703-74ef-8b85-96dfd8525025", "version": 1, "bodyDocId": null, "description": "People seek missing information - create open questions that demand resolution", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "block", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-27de9843b78b", "evidence": "Open question opens analytical loop.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 130, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "MECE Principle", "slug": "mece-principle", "status": "active", "bestFor": null, "canonId": "019dd956-7c60-71a8-8758-c79d914489c9", "version": 1, "bodyDocId": null, "description": "Mutually Exclusive, Collectively Exhaustive - ensuring complete, non-overlapping structure", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "block", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-58f86578b64c", "evidence": "Driver categories (Hardware/Data/Research/Commercial) structure the table.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 144, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Chunking", "slug": "chunking", "status": "active", "bestFor": null, "canonId": "019dd956-abab-72ce-b64c-2924cc83cf39", "version": 1, "bodyDocId": null, "description": "Information grouped into chunks is remembered better - structure content in digestible units", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-2d7f3a6f9208", "evidence": "Six numbered sections chunk a 156-page report.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 2, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "loop", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-3bc9311dc5a5", "evidence": "First slide of context->conflict->insight aha-moment loop.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 7, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-4c5539c9e63f", "evidence": "First slide of pattern_hunter loop assembling hardware evidence.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 10, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Contrast Pairs", "slug": "contrast-pairs", "status": "active", "bestFor": null, "canonId": "019dd956-9b77-70d3-88b1-f34e26f60860", "version": 1, "bodyDocId": null, "description": "Creating tension through opposites: Before↔After, Problem↔Solution, Pain↔Relief", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "loop", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-787c2dff0c52", "evidence": "TPUv2 vs V100 explicit contrast pair.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 21, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-28dffa327633", "evidence": "First slide of talent-concentration pattern_hunter loop.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 57, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-a82f38546a5e", "evidence": "First slide of vertical-applications pattern_hunter loop.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 86, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-d49c07a21d14", "evidence": "Opens public-attitudes pattern_hunter loop.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 110, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Contrast Pairs", "slug": "contrast-pairs", "status": "active", "bestFor": null, "canonId": "019dd956-9b77-70d3-88b1-f34e26f60860", "version": 1, "bodyDocId": null, "description": "Creating tension through opposites: Before↔After, Problem↔Solution, Pain↔Relief", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "loop", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-f59fbb53f70b", "evidence": "Optimism-then-fear tension across consecutive slides.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 117, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-3e0e39c4a499", "evidence": "First slide of why-now AI nationalism evidence stack.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 139, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Contrast Pairs", "slug": "contrast-pairs", "status": "active", "bestFor": null, "canonId": "019dd956-9b77-70d3-88b1-f34e26f60860", "version": 1, "bodyDocId": null, "description": "Creating tension through opposites: Before↔After, Problem↔Solution, Pain↔Relief", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "loop", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-549506a89ada", "evidence": "Side-by-side US/China table — explicit contrast pair.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 144, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Chunking", "slug": "chunking", "status": "active", "bestFor": null, "canonId": "019dd956-abab-72ce-b64c-2924cc83cf39", "version": 1, "bodyDocId": null, "description": "Information grouped into chunks is remembered better - structure content in digestible units", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-762d8a53b101", "evidence": "8 numbered predictions chunked into discrete bullets.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 152, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "9108ca24-faba-40ed-b640-1cb49b73e962", "evidence": "list/bullet: Research: Technology breakthroughs and their capabilities.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-17 03:20:34.849092+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 2, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Visual Hierarchy", "slug": "visual-hierarchy", "status": "active", "bestFor": null, "canonId": "019dd956-b2ca-706b-937f-e52441086fd7", "version": 1, "bodyDocId": null, "description": "Order of perception through Size > Color > Position > Weight > Space", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-2a6cbc518380", "evidence": "Section navigation pipe-bar header repeats across deck.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 2, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-31ff604ffdc2", "evidence": "Definitions slide grounds abstract terms before use.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 4, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "be89dcce-c44e-4d1f-aee0-f8d80895b9d2", "evidence": "list/bullet: Artificial Intelligence (AI): A broad discipline...", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-17 03:20:34.945311+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 4, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "71d87fed-6c1f-4dd3-98b0-17b648c98554", "evidence": "Section 1: Research and technical breakthroughs", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:30.887403+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 6, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-351f5df0f588", "evidence": "Callout 'transferred' frames knowledge as portable cargo.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 7, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "1c4e8e44-b887-4a8d-8a0b-d51f385162fc", "evidence": "list/bullet: Why does transfer learning matter? Re-using previously acquired knowledge reduces the amount of data a model needs in order to learn a new task.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:30.988165+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 8, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-3efd50fe452d", "evidence": "Title states 'From predicting everyday objects on ImageNet to detecting skin cancer'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 9, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Annotation", "slug": "annotation", "status": "active", "bestFor": null, "canonId": "019dd956-b5b9-7668-bf0e-3af99823ec5c", "version": 1, "bodyDocId": null, "description": "Adding explanatory labels to highlight key data points", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-45d2416a6b79", "evidence": "Callout points to red dermatologist points below blue model curve.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 9, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Sinatra Test", "slug": "sinatra-test", "status": "active", "bestFor": null, "canonId": "019dd956-be4a-775f-8038-d4e7e58dfc5a", "version": 1, "bodyDocId": null, "description": "If you can make it there, you can make it anywhere - using flagship example", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-42b8fabe2e0f", "evidence": "Single flagship example (skin cancer) carries the transfer-learning argument.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 9, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "dd662028-4ab2-4381-a97c-28b981f5ca63", "evidence": "title/action-title: The role of semiconductors in driving AI performance", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:31.090485+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 10, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-482fa470ac47", "evidence": "Title 'AI hardware as the new frontier' states position not topic.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 10, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-5078758dbbcd", "evidence": "Headline 'The hardware war: More GPUs allows for faster training'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 11, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "6a394c96-7517-489c-9aa0-abb752d9c208", "evidence": "title/action-title: The hardware war: More GPUs allows for faster training, as well as bigger (more powerful) models.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:31.156759+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 11, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-574b318fb7e0", "evidence": "Title 'AI hardware is especially helpful for deep learning'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 12, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-58519a1fcec3", "evidence": "Title quantifies '300,000x' compute growth driving Go breakthrough.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 14, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Weber-Fechner Law", "slug": "weber-fechner-law", "status": "active", "bestFor": null, "canonId": "019dd956-cf72-75bd-b89a-48bc52f65964", "version": 1, "bodyDocId": null, "description": "Perceived change is proportional to initial stimulus - bigger changes needed to notice in larger contexts", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-5cfd5dc0d4ef", "evidence": "300,000x figure designed to register magnitude shock.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 55, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 14, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-62d54ed1da68", "evidence": "Title 'No wonder GPUs have grown immensely in popularity'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 15, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-6771c90ffa06", "evidence": "Growth metric '+15X' implies trend chart.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 15, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "befa04c5-9ac5-4708-82b2-11a89f69b58f", "evidence": "The slide contains a table with data, which is a clear instantiation of the data-ink ratio tool.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:31.794421+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 17, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "2ecef736-b4a3-4648-9206-97b137cd05a1", "evidence": "The slide contains a scatter chart, which is a clear instantiation of the chart selection guide tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:31.872341+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 18, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-689d46240d40", "evidence": "Title states '242x' performance from new architectures.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 19, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chartjunk Elimination", "slug": "chartjunk-elimination", "status": "active", "bestFor": null, "canonId": "019dd956-b49e-7453-ab01-b60f338eeda1", "version": 1, "bodyDocId": null, "description": "Remove decorative elements that don't convey data: 3D effects, shadows, unnecessary gridlines", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "6ed954bf-912b-4202-8510-c71359ca85e3", "evidence": "The slide contains a bar chart with clear and simple visuals, which is a clear instantiation of the chartjunk elimination tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:31.952625+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 19, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-6e9e06ffca00", "evidence": "Title 'GPUs and novel silicon are costly to rent... limited by financial resources'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 20, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "f01d68df-b1bf-4ecf-ac17-70e6a3b88f5e", "evidence": "The slide contains a bar chart with a clear narrative, which could be an instantiation of the data story arc tool.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:32.033938+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 20, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-7742d096b03a", "evidence": "Title makes head-to-head TPUv2 vs V100 economic claim.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 21, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Waterfall chart", "slug": "waterfall-chart", "status": "active", "bestFor": null, "canonId": "e2799da3-725f-44a9-a093-570f1f388e64", "version": 1, "bodyDocId": null, "description": null, "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "4f1655bb-a7b7-4fd6-bc42-0795aac493b3", "evidence": "The slide contains a line chart and a bar chart, which could be an instantiation of the waterfall chart tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:32.113519+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 21, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "ee9d9b02-12c9-4528-91da-63b9f80cdb2d", "evidence": "The slide contains a list with bullet points, which is a clear instantiation of the list presentation tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:32.193304+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 22, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Slide visual craft", "slug": "visual-craft", "status": "active", "bestFor": null, "canonId": "019df269-2d56-7739-ab65-cc2e091a109b", "version": 1, "bodyDocId": "019df269-2d32-736a-ae59-0250c5540c9a", "description": "What separates a 5/5-visual-craft deck from a 3/5 one.", "familyLabel": "slide-recipe", "categoryName": "Slide", "categorySlug": "slide" }, "agent": "designer", "layer": "slide", "agents": [ "designer" ], "matchId": "b3092564-982b-4650-9c0b-ac6d154c5e76", "evidence": "The slide contains an infographic with clear and simple visuals, which is a clear instantiation of the visual craft tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:32.272862+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 23, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-7c87341a72c5", "evidence": "Title 'NVIDIA's datacenter business breaks $2B run-rate, growing >100% YoY'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 24, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "799a0830-2a20-4e97-9300-63475d82b601", "evidence": "The slide contains a bar chart, which is a clear instantiation of the chart selection guide tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:32.352489+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 24, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-833999375475", "evidence": "Title 'NVIDIA's enterprise value has 10x in 3 years'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 25, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-87d804342658", "evidence": "Title quantifies Intel datacenter 30% revenue share.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 26, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-895f07012035", "evidence": "Title 'Many companies are developing custom AI chips'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 27, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Small Multiples", "slug": "small-multiples", "status": "active", "bestFor": null, "canonId": "019dd956-b32d-7586-b368-baa1dd388ed4", "version": 1, "bodyDocId": null, "description": "Repeating similar charts for comparison across categories or time", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-8cd9df00b36a", "evidence": "Market landscape implies many parallel logos/cards.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 27, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-909994feab22", "evidence": "Title 'Large technology companies are hedging their hardware suppliers'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 28, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-97c9f2224b12", "evidence": "Title quantifies $76bn cloud capex by 2020.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 29, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "AIDA Model", "slug": "aida-model", "status": "active", "bestFor": null, "canonId": "019dd956-86fc-773c-813f-d416714b8413", "version": 1, "bodyDocId": null, "description": "Attention, Interest, Desire, Action - classic persuasion sequence", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "bb74dc5d-4d24-4b98-879f-b07178a9a0f9", "evidence": "The slide presents a clear and concise message, grabbing the audience's attention with a bold title and highlighting a key statistic, which is a common structure used in the AIDA model.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.473228+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 29, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-9aa29bd9b41c", "evidence": "Title 'Traditional computer vision describes scenes by detecting objects (nouns)'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 30, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-9e8bc4f23920", "evidence": "Nouns/verbs metaphor frames CV vs scene understanding.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 30, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-a05ddd60664e", "evidence": "Title 'detecting objects in images is not enough to produce real scene understanding'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 31, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "2228f7da-f69c-43ec-8ec7-801aa19269cb", "evidence": "The slide clearly articulates a problem, providing context and highlighting the limitations of current solutions, which is a key component of the problem statement canvas.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:38.568648+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 31, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-a5bca6c0c634", "evidence": "Title 'True scene understanding requires understanding actions (verbs) and common sense'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 32, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-a9c29f584dc8", "evidence": "Verbs vs nouns metaphor extended.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 32, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Three-Act Structure", "slug": "three-act-structure", "status": "active", "bestFor": null, "canonId": "019dd956-8812-7796-9105-e7bc0e405681", "version": 1, "bodyDocId": null, "description": "Setup, Confrontation, Resolution - classic dramatic structure", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "441f482b-4600-4ddd-b37e-1e6ff2b3a324", "evidence": "The slide presents a clear progression of ideas, introducing a problem, describing a solution, and highlighting its potential impact, which follows a three-act structure.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:38.629451+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 32, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-af9d57e86a31", "evidence": "Title describes dataset-building approach.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 33, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-b0062aa4438d", "evidence": "Title 'Deep learning models can actually understand the verbs as well as the nouns'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 34, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-b6c66082ba7c", "evidence": "Title 'Machines can also understand visual scenes by learning from multiple viewpoints'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 35, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-b80e848e0ccc", "evidence": "Title 'AlphaZero showed that a deep RL system can learn from scratch to beat Go champions'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 37, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Sinatra Test", "slug": "sinatra-test", "status": "active", "bestFor": null, "canonId": "019dd956-be4a-775f-8038-d4e7e58dfc5a", "version": 1, "bodyDocId": null, "description": "If you can make it there, you can make it anywhere - using flagship example", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-be170044e6f0", "evidence": "Flagship case (AlphaZero, 40 days) carries the RL claim.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 37, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-c39aa2ae7b9a", "evidence": "Title 'OpenAI's multi-agent RL system learns to play Dota2'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 38, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-c41cb2294f10", "evidence": "Title 'RL agents can also build their own world models'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 39, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-ca8bf72fb965", "evidence": "Title posed as question 'How do we ensure our models are not biased?'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 40, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-ccbd7bd7560a", "evidence": "Title 'An example of biased machine learning systems: Stereotyping'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 41, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Anchoring Bias", "slug": "anchoring-bias", "status": "active", "bestFor": null, "canonId": "019dd956-9254-73de-8f1b-c7250f8f4c9c", "version": 1, "bodyDocId": null, "description": "First piece of information becomes reference point for all subsequent judgments", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect", "Storyteller" ], "matchId": "d705c7de-1ddd-4aa4-8fa4-06c41bc36559", "evidence": "Turkish is a gender-neutral language, yet Google Translate swaps the gender of the pronouns when translating from English to Turkish and back to English", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.547572+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 41, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-d3fddd63b37b", "evidence": "Specific Turkish-Google Translate example.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 41, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Singularity Effect", "slug": "singularity-effect", "status": "active", "bestFor": null, "canonId": "019dd956-dcd3-7287-94d8-5a49496c2ddb", "version": 1, "bodyDocId": null, "description": "People care more about individuals than statistics - use specific examples", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-d68d2afdc37e", "evidence": "Single concrete bias example over abstract stats.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 41, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-d87139e35c6e", "evidence": "Title 'Another example of biased machine learning systems: Racial bias'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 42, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Anchoring Bias", "slug": "anchoring-bias", "status": "active", "bestFor": null, "canonId": "019dd956-9254-73de-8f1b-c7250f8f4c9c", "version": 1, "bodyDocId": null, "description": "First piece of information becomes reference point for all subsequent judgments", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect", "Storyteller" ], "matchId": "a9e06996-6bef-4523-b486-ec40da386df7", "evidence": "When trained on datasets that do not appropriately reflect diversity of skin color, computer vision systems exhibit offensive racial bias", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.614502+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 42, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Emotional Appeal", "slug": "emotional-appeal", "status": "active", "bestFor": null, "canonId": "019dd956-bde9-7668-8be7-0097678fec59", "version": 1, "bodyDocId": null, "description": "Connecting to audience feelings rather than just logic", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-dc0843dbf1ed", "evidence": "Racial-bias example designed for moral charge.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 42, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "2x2 matrix", "slug": "matrix-2x2", "status": "active", "bestFor": null, "canonId": "019dd9e1-50ab-72d1-8afb-3ab0a38e32a8", "version": 1, "bodyDocId": null, "description": "BCG matrix, importance/urgency, any axes-and-quadrants frame.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "71c8db6e-2e62-4890-9fbf-8c23ab36054b", "evidence": "Matrix of 7 AI bias examples mapped against 5 bias categories", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:36.681189+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-e27c214e7f2d", "evidence": "Title 'ML models have 5 types of allocation bias that stem from training data'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-e9d49e4c82c6", "evidence": "Callout 'black boxes' metaphor for opaque models.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 44, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-f307bd88bbaf", "evidence": "Title 'Explainability helps validate that ML models perform well for the right reasons'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 45, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Visual Anchors", "slug": "visual-anchors", "status": "active", "bestFor": null, "canonId": "019dd956-c264-7357-a791-89fa5f02a400", "version": 1, "bodyDocId": null, "description": "Design elements that guide users' eyes through the content intentionally", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "a60d6033-b99e-40e4-a048-9da3377ae645", "evidence": "Figure 4: Explaining an image classification prediction made by Google’s Inception neural network", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:36.785603+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 45, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-f4bad5980e99", "evidence": "Title 'Next step: Justifying decisions in plain language'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 46, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Waterfall chart", "slug": "waterfall-chart", "status": "active", "bestFor": null, "canonId": "e2799da3-725f-44a9-a093-570f1f388e64", "version": 1, "bodyDocId": null, "description": null, "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "1403d9ce-3e85-4c5c-b98c-df7956e22af0", "evidence": "Lollipop chart showing feature importance for cervical cancer prediction", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:36.885106+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 47, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-f84ec4d232d2", "evidence": "Title 'Imperceptible changes to data can alter a deep learning model's prediction' with 99.3% stat.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 48, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Unexpected Pattern", "slug": "unexpected-pattern", "status": "active", "bestFor": null, "canonId": "019dd956-bc0f-71af-8d32-3c1a2da38c40", "version": 1, "bodyDocId": null, "description": "Breaking a pattern to create surprise and attention", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-fd335a7f063a", "evidence": "Adversarial example breaks expectation that small change = small effect.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 48, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-01602bc9539b", "evidence": "Same title; toaster patch example.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 49, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-07a15920615d", "evidence": "Toaster patch is concrete, vivid example.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 49, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-0ba566bd457e", "evidence": "Title 'Adversarial attacks present serious safety challenges in the real world'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 50, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Emotional Appeal", "slug": "emotional-appeal", "status": "active", "bestFor": null, "canonId": "019dd956-bde9-7668-8be7-0097678fec59", "version": 1, "bodyDocId": null, "description": "Connecting to audience feelings rather than just logic", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-0c27bc0c0126", "evidence": "Pedestrian-at-zebra-crossing made invisible — visceral.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 50, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-132694e4f723", "evidence": "Title with '80%' stat on accuracy reductions.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 52, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chartjunk Elimination", "slug": "chartjunk-elimination", "status": "active", "bestFor": null, "canonId": "019dd956-b49e-7453-ab01-b60f338eeda1", "version": 1, "bodyDocId": null, "description": "Remove decorative elements that don't convey data: 3D effects, shadows, unnecessary gridlines", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "d54e1dc4-78c3-4172-af83-ed386f0a8b71", "evidence": "Bar chart showing Top-1 accuracy of various models from 2013 to 2016", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.083383+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 52, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-16354f7fb19c", "evidence": "Title 'AI to automate away AI engineers'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 53, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Unexpected Pattern", "slug": "unexpected-pattern", "status": "active", "bestFor": null, "canonId": "019dd956-bc0f-71af-8d32-3c1a2da38c40", "version": 1, "bodyDocId": null, "description": "Breaking a pattern to create surprise and attention", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-1bf7068e6cbc", "evidence": "Recursive twist — AI replacing AI engineers — surprises.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 53, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Waterfall chart", "slug": "waterfall-chart", "status": "active", "bestFor": null, "canonId": "e2799da3-725f-44a9-a093-570f1f388e64", "version": 1, "bodyDocId": null, "description": null, "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "38105218-80ce-4f5c-9713-9171e92b6162", "evidence": "chart/scatter: Accuracy vs. # Mult-Add operations", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:33.279949+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 53, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-1c66fe5f8199", "evidence": "Title 'Distributed federated learning to decentralise data acquisition'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 54, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-22b21615e6d2", "evidence": "Title repeats federated framing; process diagram.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 55, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Process flow", "slug": "process-flow", "status": "active", "bestFor": null, "canonId": "019dd9e1-4dca-76bc-8135-401836a1c4c0", "version": 1, "bodyDocId": null, "description": "Workflow, steps, sequential diagram.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "4712ee03-225f-433d-8595-d348d006c361", "evidence": "diagram/process: Federated learning process flow", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:33.381408+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 55, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "f1dd0563-6012-4bb8-8362-c92362705192", "evidence": "type: section_divider", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:33.447831+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 56, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-26078fa4b136", "evidence": "Title 'Element AI estimates 22,000 PhD educated AI researchers'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 57, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-2dc3a646c539", "evidence": "Title '5,000 high-level researchers worldwide'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 58, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-3070433042ce", "evidence": "Title 'America remains the hub for talent exchange'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 59, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-3521b6617347", "evidence": "Title 'Google is widely acknowledged as the leading employer of AI talent'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 60, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-3843985ba57b", "evidence": "Title quantifies '6.3% of 2017 ICML papers had a Google/DeepMind author'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 61, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-3ce963013302", "evidence": "Title 'Percentage of ICML papers with Google/DeepMind author doubles'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 62, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-4211c677f508", "evidence": "Title 'Google lead in contribution to 2017 NIPS papers'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 63, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-445ac238214f", "evidence": "Title 'Google & DeepMind dominate NIPS authorship'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 64, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "cccf6def-a3b7-45fb-9806-032281b8f953", "evidence": "chart/bar-stacked-100pct: Salary / Compensation Changes in the Past 3 Years for Data Scientists and Machine Learning Engineers", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:57:08.80941+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 65, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-4f2e66e5c135", "evidence": "Title 'Anecdotally salaries continue to grow'; $300-500k quote.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 66, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "AIDA Model", "slug": "aida-model", "status": "active", "bestFor": null, "canonId": "019dd956-86fc-773c-813f-d416714b8413", "version": 1, "bodyDocId": null, "description": "Attention, Interest, Desire, Action - classic persuasion sequence", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "14050086-8286-46e2-bbd2-946ee577317e", "evidence": "quote: Typical A.I. specialists, including both Ph.D.s fresh out of school and people with less education and just a few years of experience, can be paid from $300,000 to $500,000 a year or more in salary and company sto", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:57:08.88717+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 66, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-50e8ece3290f", "evidence": "Specific salary band rather than abstract 'high pay'.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 66, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-557714f6ece3", "evidence": "Title 'Compensation can be astronomical and relationships litigious'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 67, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Monroe's Motivated Sequence", "slug": "monroes-motivated-sequence", "status": "active", "bestFor": null, "canonId": "019dd956-8758-700f-83cf-d260a54ad24f", "version": 1, "bodyDocId": null, "description": "Five-step persuasion: Attention, Need, Satisfaction, Visualization, Action", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "f15a095d-a4e0-4e94-8b3a-87bb0872b445", "evidence": "paragraph/paragraph: Google paid its self-driving car boss $120 million — and then he left for Uber.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:57:08.965632+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 67, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Story Moments", "slug": "story-moments", "status": "active", "bestFor": null, "canonId": "019dd956-ae4e-768d-a18b-108e6cda511e", "version": 1, "bodyDocId": null, "description": "Design key emotional beats: The Shock, The Vision, The Proof, The Choice, The Call", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-5debd1d5d586", "evidence": "Single anecdote functions as 'the shock' moment.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 67, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Von Restorff Effect", "slug": "von-restorff-effect", "status": "active", "bestFor": null, "canonId": "019dd956-beab-73d8-9f62-30d0332e926e", "version": 1, "bodyDocId": null, "description": "Items that stand out are noticed and remembered - make key points visually distinct", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8ecb-5bcaaef47223", "evidence": "$120M figure stands out, designed to shock.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 67, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-605693573b66", "evidence": "Topic-with-stat title '17%' on diversity.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 68, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "2410d9be-5f15-4f41-89df-a525adb529c4", "evidence": "list/bullet: Key research labs are not yet making their workforce diversity statistics public.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:57:09.045763+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 68, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "b85f5e8d-bc34-45e2-aac6-684aad5d19af", "evidence": "chart/bar-vertical: % of NIPS conference registrations from women", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:57:09.121678+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 69, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "da6b3cbc-9b5d-4596-b850-3b7f42e7f99f", "evidence": "paragraph/paragraph: Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:57:09.198897+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 70, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-67b724cb8b9d", "evidence": "Title 'AI takes the world stage: GAFAMBAT are in the ring together'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 71, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8ecb-6d6ab973373f", "evidence": "Stacked bar (patents) + horizontal bar (R&D spend).", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 71, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-6bfef7291ac7", "evidence": "Boxing 'in the ring' metaphor.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 71, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Visual Hierarchy", "slug": "visual-hierarchy", "status": "active", "bestFor": null, "canonId": "019dd956-b2ca-706b-937f-e52441086fd7", "version": 1, "bodyDocId": null, "description": "Order of perception through Size > Color > Position > Weight > Space", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8ecb-73b3aebb4967", "evidence": "Tech vs non-tech color-coded for emphasis.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 71, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-740ea173cacc", "evidence": "Title 'Big cloud providers... building blocks of intelligence via API'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 72, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Canvas framework", "slug": "canvas-framework", "status": "active", "bestFor": null, "canonId": "019dd9e1-5005-762c-82f8-265b08436ca1", "version": 1, "bodyDocId": null, "description": "Named canvas (business model canvas, value prop canvas, etc.).", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "4660af07-cfc2-4a30-848c-e634dc0b063a", "evidence": "list/bullet: Google is investing heavily to expose ML services through their cloud ecosystem", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:57:09.32286+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 72, "whyItWorks": null, "antipattern": null, "cardinality": "optional", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-79bc24e8f5b2", "evidence": "Title 'Google's TensorFlow is winning the ML framework war'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 73, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-7d9024dc625b", "evidence": "'Framework war' metaphor.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 73, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-804988813afe", "evidence": "Title 'Why now? Today's drug development process is too slow and expensive'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 74, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "72bf1512-1be1-46f8-a976-5a084d17ca6b", "evidence": "callout: Why now? Today's drug development process is too slow and expensive", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:57:09.440965+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 74, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "42dd50ba-b558-4481-8f96-53067bfd7a2e", "evidence": "list/bullet: Repurpose existing drugs: Discover how existing drugs can be repurposed for new conditions... Selected examples: [Logos]", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:57:09.52683+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 75, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-885ca802b5ea", "evidence": "Title 'Healthcare systems worldwide are costly and overburdened'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 76, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-8f11fc303e1d", "evidence": "Title 'Breast cancer as a case study: Not enough doctors, diagnosing is hard'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 77, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "AIDA Model", "slug": "aida-model", "status": "active", "bestFor": null, "canonId": "019dd956-86fc-773c-813f-d416714b8413", "version": 1, "bodyDocId": null, "description": "Attention, Interest, Desire, Action - classic persuasion sequence", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "7cc7e898-c2fc-4dd8-9c50-e419ecdd557e", "evidence": "The slide presents a problem (not enough doctors), agitation (diagnosing is hard and care is expensive), and a potential solution or call to action (early detection leads to higher 5-year survival rates).", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:34.922061+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 77, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Sinatra Test", "slug": "sinatra-test", "status": "active", "bestFor": null, "canonId": "019dd956-be4a-775f-8038-d4e7e58dfc5a", "version": 1, "bodyDocId": null, "description": "If you can make it there, you can make it anywhere - using flagship example", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-9225dd0e2e32", "evidence": "Breast cancer flagship case proves healthcare AI value.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 77, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "2x2 matrix", "slug": "matrix-2x2", "status": "active", "bestFor": null, "canonId": "019dd9e1-50ab-72d1-8afb-3ab0a38e32a8", "version": 1, "bodyDocId": null, "description": "BCG matrix, importance/urgency, any axes-and-quadrants frame.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "5e6a4188-bfef-4a0d-ab6e-58966cf28462", "evidence": "Although not explicitly a 2x2 matrix, the logo grid could be argued to represent a simple matrix structure.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:35.043077+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 78, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "cf55e093-a729-4ff6-b7c1-cf7c0ab3f7dd", "evidence": "The slide presents information in a list format with bullet points, making it a list presentation.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.00183+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 78, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Narrative foundations (from the Storymakers methodology)", "slug": "narrative-foundations", "status": "active", "bestFor": null, "canonId": "019df269-258b-71bd-ad14-53598df04d2b", "version": 1, "bodyDocId": "019df269-2567-736e-8b9c-4ec77611334a", "description": "This skill focuses on the *adversarial* genre — activist campaigns, short", "familyLabel": "narrative-block", "categoryName": "Narrative & Transformation", "categorySlug": "narrative-transformation" }, "agent": "storyteller", "layer": "slide", "agents": [ "storyteller" ], "matchId": "28dc91c9-409d-44fb-b49c-39bb7e68d954", "evidence": "The slide has a structured approach with Introduction, Research, Talent, Industry, Politics, Predictions, and Conclusion sections.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:35.124053+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 79, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-97b35c34d4cb", "evidence": "Title 'Expect to see more activity as companies move products through trials'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 80, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "020daf1f-efda-4040-a9f8-e8bd4caf93ee", "evidence": "The slide uses a horizontal bar chart to show company counts by year, which is an appropriate chart type for categorical data.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.201868+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 80, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-98a520264498", "evidence": "Title 'Population-level surveillance is taking off in China' with 400M figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 81, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Storytelling Effect", "slug": "storytelling-effect", "status": "active", "bestFor": null, "canonId": "019dd956-981b-74f9-9096-5c8df03dab5a", "version": 1, "bodyDocId": null, "description": "Stories are remembered up to 22x better than facts alone", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "78136482-56b8-45a2-a804-6bbfe2ec01ff", "evidence": "The slide tells a story about the growth of CCTV surveillance in China and its implications.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.282017+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 81, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Problem-Agitate-Solve (PAS)", "slug": "problem-agitate-solve-pas", "status": "active", "bestFor": null, "canonId": "019dd956-869f-7564-8d42-de54ed29ab90", "version": 1, "bodyDocId": null, "description": "Identify problem, agitate the pain, present the solution", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "00b6b4ac-f5ed-4b1c-bf08-d5eb1407209f", "evidence": "The slide presents a problem (Project Maven backlash), agitates it (Google's involvement), and hints at a solution (Google's new AI principles).", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.362031+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 82, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Emotional Appeal", "slug": "emotional-appeal", "status": "active", "bestFor": null, "canonId": "019dd956-bde9-7668-8be7-0097678fec59", "version": 1, "bodyDocId": null, "description": "Connecting to audience feelings rather than just logic", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "2aba70c9-dcd8-4626-b304-81d1b51ce901", "evidence": "The slide uses a quote and an image to evoke an emotional response to the Cambridge Analytica scandal.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:35.44079+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 83, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-9ef711004660", "evidence": "Concrete count 2,600,968,280 in title indicates breach scale.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 84, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chartjunk Elimination", "slug": "chartjunk-elimination", "status": "active", "bestFor": null, "canonId": "019dd956-b49e-7453-ab01-b60f338eeda1", "version": 1, "bodyDocId": null, "description": "Remove decorative elements that don't convey data: 3D effects, shadows, unnecessary gridlines", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "46daa846-80a7-47e4-9a5f-e15aea63b94d", "evidence": "The slide uses a simple and clear bar chart to present data on trust in government and social media sites.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.518968+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 84, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-a2c08559ad58", "evidence": "Exact integer '2,600,968,280' chosen over rounded version.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 84, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "2ffd5f9b-d92a-408d-b128-88705b793abb", "evidence": "The slide presents information in a list format with bullet points, making it a list presentation.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.596288+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 85, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-a64a500c77a4", "evidence": "Title 'Why now? Satellite data is decreasing in cost and increasing in resolution'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 86, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Before-After-Bridge", "slug": "before-after-bridge", "status": "active", "bestFor": null, "canonId": "019dd956-8604-7188-883f-42370288ad05", "version": 1, "bodyDocId": null, "description": "Narrative structure: Paint the before, show the after, explain the bridge", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "5ff146c7-6603-410a-bca8-85948b92aca3", "evidence": "The slide presents a before (high cost, low resolution) and after (decreasing cost, increasing resolution) scenario.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.673821+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 86, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "82ef5b33-954f-4892-9203-d1f9a092a312", "evidence": "The slide presents information in a list format with bullet points, making it a list presentation.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.753082+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 87, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Narrative foundations (from the Storymakers methodology)", "slug": "narrative-foundations", "status": "active", "bestFor": null, "canonId": "019df269-258b-71bd-ad14-53598df04d2b", "version": 1, "bodyDocId": "019df269-2567-736e-8b9c-4ec77611334a", "description": "This skill focuses on the *adversarial* genre — activist campaigns, short", "familyLabel": "narrative-block", "categoryName": "Narrative & Transformation", "categorySlug": "narrative-transformation" }, "agent": "storyteller", "layer": "slide", "agents": [ "storyteller" ], "matchId": "cd5275ee-21bc-42c5-9967-c34f0fdc6500", "evidence": "The slide has a structured approach with Introduction, Research, Talent, Industry, Politics, Predictions, and Conclusion sections.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:35.834454+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 88, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-ada4a571e7d8", "evidence": "Topic title 'Cybersecurity' with $11.7M cost stat.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 89, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "352e1766-28c9-45c4-87a8-7f47cd14488d", "evidence": "Cloud computing, mobile devices, and more interconnected supply chains means the attack surface for cyber attacks is expanding. At the same time there is a growing shortage of cybersecurity", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.254257+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 89, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Storytelling: audience calibration", "slug": "audience", "status": "active", "bestFor": null, "canonId": "019df269-23ea-723a-87c0-af1d78d702d2", "version": 1, "bodyDocId": "019df269-23c5-72ec-8b94-78fc1da7b656", "description": "The same thesis, written for four different audiences, produces four", "familyLabel": "narrative-block", "categoryName": "Narrative & Transformation", "categorySlug": "narrative-transformation" }, "agent": "storyteller", "layer": "slide", "agents": [ "storyteller" ], "matchId": "0faf014d-94fc-49f0-8af4-9d7195050ba3", "evidence": "Why now? Cloud computing, mobile devices, and more interconnected supply chains means the attack surface for cyber attacks is expanding.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.220822+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 89, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "d8b08b24-dd68-48a2-82fe-3c9473939eac", "evidence": "Network and endpoint security: Supervised learning is used to detect malicious activity on an organisation’s network based on data from past attacks.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.323252+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 90, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-b2a5dbb1945f", "evidence": "Topic 'Warehouse automation' with 120,000 figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 91, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Before-After-Bridge", "slug": "before-after-bridge", "status": "active", "bestFor": null, "canonId": "019dd956-8604-7188-883f-42370288ad05", "version": 1, "bodyDocId": null, "description": "Narrative structure: Paint the before, show the after, explain the bridge", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "45e90108-6056-4a85-886f-c284d027d9b1", "evidence": "Why now? eCommerce growth decreases order size for item picking in warehouses and increases customer expectations around the speed of fulfilment.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.391278+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 91, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "2b09dfec-b6c0-4a16-b5e9-ee68bf7a6997", "evidence": "Warehouse management systems: Using machine learning within warehouse management software to optimise inventory, order picking and queues to minimise waste.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.458999+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 92, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-b4fc28da99bf", "evidence": "Callout drives why-now: sensor cost down, vision better, robots cheaper than labour.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 94, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "98bacd4d-71eb-449c-9592-b77dd0bce44e", "evidence": "Index of average robot prices and labour compensation in U.S. manufacturing", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.52678+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 94, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "9943f72f-f19e-4a15-8110-30eb7fbafc28", "evidence": "list/bullet: Cleaning: Self-driving cleaning robots for industrial spaces.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-17 03:20:35.043489+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 95, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-bb2599bcb4f3", "evidence": "Callout '70% more food calories needed by 2050'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 97, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "9893e71e-aee2-4549-b04e-245cbbc5ab79", "evidence": "The world population is expected to grow from 7.6 billion to 9.6 billion by 2050. We need to produce 70% more food calories to feed the world's population by then.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.593063+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 97, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-bf6cc421d2de", "evidence": "Title 'Fewer farm workers... higher wages... but more automation leads to stable labor cost share'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 98, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chartjunk Elimination", "slug": "chartjunk-elimination", "status": "active", "bestFor": null, "canonId": "019dd956-b49e-7453-ab01-b60f338eeda1", "version": 1, "bodyDocId": null, "description": "Remove decorative elements that don't convey data: 3D effects, shadows, unnecessary gridlines", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "450d2da4-c7aa-429c-ad7b-4821873c4138", "evidence": "Labor costs as a share of total gross cash income for selected farm specializations, 2000-16", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.660088+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 98, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "4c6c8567-514d-4aee-9690-12dae8cc0d40", "evidence": "Vertically-integrated farming: Compact, self-contained greenhouses for growing crops closer to the point of consumption.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.726501+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 99, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "8742df94-a17e-41ef-b0a1-bc52d2b4a7ae", "evidence": "components: list/bullet: Crop picking robots: Build robots capable of mapping and navigating through crop fields while identifying and carefully picking ripe fruit automatically.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-17 03:20:48.623151+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 100, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "7e2e35e9-af2e-4824-b1a9-113a8e0e3dc4", "evidence": "The slide presents a series of big numbers and metrics related to road safety, which is a typical structure for a problem statement.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.778221+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 101, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Storytelling Effect", "slug": "storytelling-effect", "status": "active", "bestFor": null, "canonId": "019dd956-981b-74f9-9096-5c8df03dab5a", "version": 1, "bodyDocId": null, "description": "Stories are remembered up to 22x better than facts alone", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "d3579080-887a-4fcb-a338-8770b3f346aa", "evidence": "The use of multiple big numbers and metrics creates a narrative around the problem of road safety.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.811698+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 101, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "1ba85ea4-1eca-4d61-af66-9e735195e3b9", "evidence": "The slide presents information in a list format with bullet points.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.876817+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 102, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Slide visual craft", "slug": "visual-craft", "status": "active", "bestFor": null, "canonId": "019df269-2d56-7739-ab65-cc2e091a109b", "version": 1, "bodyDocId": "019df269-2d32-736a-ae59-0250c5540c9a", "description": "What separates a 5/5-visual-craft deck from a 3/5 one.", "familyLabel": "slide-recipe", "categoryName": "Slide", "categorySlug": "slide" }, "agent": "designer", "layer": "slide", "agents": [ "designer" ], "matchId": "b78c6fd2-6bcc-4dad-8e49-aaa3a6fce2e4", "evidence": "The slide uses images and logos to visually present information.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.945092+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 103, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-c2521190d621", "evidence": "Topic 'Finance' with 7x stat.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 104, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "3bdaa9cf-9c31-4d05-8371-d769bc943ff0", "evidence": "The slide includes various charts, such as line and bar charts.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.013838+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 104, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "332b462c-09d5-4236-9834-5d4d7f04a977", "evidence": "The slide presents information in a list format with bullet points.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.080904+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 105, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-c748ca9fd3b9", "evidence": "Callout '2/3 day is unstructured or unpredictable' frames automation gap.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 106, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "61cfaeb9-488a-4cff-89d2-816360663dc1", "evidence": "The slide includes various charts, such as bar and line charts.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.181991+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 106, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "7f9694e2-e18b-4cdf-a949-c72cf8a686c7", "evidence": "The slide presents a problem statement around enterprise automation.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.149179+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 106, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "a24354e6-7cac-4718-b88d-52db0e133339", "evidence": "The slide presents information in a list format with bullet points.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.248672+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 107, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-c86368581379", "evidence": "Title 'Material science' with callout '167x' R&D speedup.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 108, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Von Restorff Effect", "slug": "von-restorff-effect", "status": "active", "bestFor": null, "canonId": "019dd956-beab-73d8-9f62-30d0332e926e", "version": 1, "bodyDocId": null, "description": "Items that stand out are noticed and remembered - make key points visually distinct", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8ecb-ce15efacf4fb", "evidence": "167x stat designed to stand out.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 108, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "106cb033-4f43-4930-885e-e40dee666f00", "evidence": "The slide clearly divides the presentation into sections.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:37.352664+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 109, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-d100588b1ba0", "evidence": "Sets up two specific surveys to be referenced.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 110, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "7ad4c0a9-1065-488b-b45a-ca39bb5bc4f0", "evidence": "The slide presents information in a list format with bullet points.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.420707+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 110, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-da378bc4ee1a", "evidence": "Title 'Growing awareness of automation impacting jobs' with 77%.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 111, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "f97b401b-1778-49e3-804f-42d10ed42981", "evidence": "The slide includes various charts, such as bar charts.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.486975+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 111, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-dc7a67b21a1c", "evidence": "Title 'Young, part-time, hispanic and lower-income Americans report most impact'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 112, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Storytelling Effect", "slug": "storytelling-effect", "status": "active", "bestFor": null, "canonId": "019dd956-981b-74f9-9096-5c8df03dab5a", "version": 1, "bodyDocId": null, "description": "Stories are remembered up to 22x better than facts alone", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "194d817b-1387-4f36-8492-e4a9b8d98e17", "evidence": "The slide tells a story around the data presented.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.554181+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 112, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-e25bfde09521", "evidence": "Title 'Rising concerns about automation increasing inequality'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 113, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "a1e36292-f66f-47fe-be42-ed05e4020c83", "evidence": "chart/bar-stacked: Diverging bar chart showing survey results for negative and positive outcomes of automation.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.413343+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 113, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "2x2 matrix", "slug": "matrix-2x2", "status": "active", "bestFor": null, "canonId": "019dd9e1-50ab-72d1-8afb-3ab0a38e32a8", "version": 1, "bodyDocId": null, "description": "BCG matrix, importance/urgency, any axes-and-quadrants frame.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "31c9de04-45b5-4554-be83-c10f955c40b6", "evidence": "chart/bar-grouped: Grouped bar chart showing percentages for 'Impacted by automation' vs 'Not impacted' across four categories.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.492121+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 114, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-e4275df09fbd", "evidence": "Title 'Those whose job has been impacted favor more radical policies'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 114, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-ea35fbb09931", "evidence": "Callout 'Overall optimism around AI' with 41% stat.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 115, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "6cbe9b34-07bb-4899-a3da-798722230479", "evidence": "chart/pie: How optimistic are you about artificial intelligence?", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.609979+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 115, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "f57f7134-e461-45d2-bf57-9a840002ae47", "evidence": "list/bullet: very optimistic, somewhat optimistic, not very optimistic, don't know or no answer", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.570194+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 115, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-ef108df14f19", "evidence": "Callout '...expectation that AI will make my life easier'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 116, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "1abf1dfb-56a6-487d-8e4f-8b537924a952", "evidence": "chart/pie: Do you expect artificial intelligence to make your day-to-day life: easier (34%), harder (13%), don't know or no answer (53%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.687958+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 116, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-f2991456b5d3", "evidence": "Title '...but expectation that AI will reduce privacy'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 117, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "a37aacfe-9ae2-460e-892c-41223ea498b8", "evidence": "chart/pie: Do you expect artificial intelligence to: increase personal privacy (5%), have no effect on personal privacy (12%), reduce personal privacy (49%), don't know or no answer (34%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.767055+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 117, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "a265b959-90a7-4cf5-bf5d-67b50e72dcea", "evidence": "chart/pie: Do you expect artificial intelligence to: create jobs (12%), have no effect on jobs (13%), reduce jobs (38%), don't know or no answer (37%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.845885+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 118, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "836ae640-b274-46de-a7f8-8eb20c825400", "evidence": "chart/pie: Do you think artificial intelligence represents a: threat to human beings (32%), no threat to human beings (24%), don't know or no answer (44%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.923349+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 119, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "f0b9bd96-0271-48d7-bcfb-0c6b459544a7", "evidence": "chart/pie: Do you think government officials should: regulate artificial intelligence (42%), not regulate artificial intelligence (17%), don't know or no answer (41%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.003641+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 120, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-fab416adcdbe", "evidence": "Callout 'Americans believe US is currently the world leader'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 121, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "712f6f6c-9859-4b99-88db-8661a54a07c3", "evidence": "chart/bar-horizontal: Which one is the leading country when it comes to artificial intelligence?", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.085378+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 121, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-fdbf6e341b86", "evidence": "Title '...China will close the gap over the next ten years'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 122, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "97be04bd-57d0-4207-8077-2c348e642053", "evidence": "chart/bar-horizontal: In 10 years, which one will be the leading country when it comes to artificial intelligence?", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.163249+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 122, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-01a189082ec7", "evidence": "Title 'Despite increased automation the US unemployment rate is at a 17 year low'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 123, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "ba0d0cdb-835e-4443-8135-58fea9ec00c3", "evidence": "chart/line: Unemployment Rate", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.24311+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 123, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Unexpected Pattern", "slug": "unexpected-pattern", "status": "active", "bestFor": null, "canonId": "019dd956-bc0f-71af-8d32-3c1a2da38c40", "version": 1, "bodyDocId": null, "description": "Breaking a pattern to create surprise and attention", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-06352be76c2f", "evidence": "Counter-intuitive pairing of automation rise + low unemployment.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 123, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-0afbe66e9cb6", "evidence": "Callout 'Routine jobs have stagnated'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 124, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "21ce9b37-ffc1-4f92-af93-0e2d7a7beb95", "evidence": "chart/line: Jobs: Routine Vs. Nonroutine, Cognitive Vs. Manual", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.322896+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 124, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-0db031fedbd8", "evidence": "Callout 'Wages have lagged the increase in jobs'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 125, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "5539f9f1-80ad-480b-9281-476893842f35", "evidence": "chart/line: Line chart showing Unemployment rate (light blue) and Wage growth (black) from 2008 to 2018.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:42.917039+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 125, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-13d91fab0a78", "evidence": "Callout 'marked change in how long unemployment lasts for'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 126, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-1790edc3ce78", "evidence": "Callout 'Labour productivity and hourly compensation have diverged'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 127, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "9c0a94d3-c480-4e2a-ad91-aa074b8c0692", "evidence": "chart/line: Real Labor Productivity and Hourly Compensation, 1947–2017.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.009831+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 127, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-18294dab0b06", "evidence": "Callout 'Labour's share of income has been declining steadily'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 128, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "de6879c5-cd34-4285-9e84-700062ddb85e", "evidence": "chart/line: Line chart of US labor share of income (1973-2017).", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.069858+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 128, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-1ed224a59dbf", "evidence": "Callout 'Workers are experiencing greater income volatility'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 129, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "d5cc1dbd-4efb-42b7-bc14-60cfa0aae27f", "evidence": "chart/scatter: Average monthly change in income vs Annual Income", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.133113+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 129, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-20732ba33df9", "evidence": "Question title 'How much of this is due to automation?'", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 130, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "ae209c37-bae3-4438-bfe5-fb5b7be0013f", "evidence": "There are two poles of thought on how machine learning will affect the labour market.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:43.193929+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 130, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "d3cf1188-84ae-48ea-b614-900a0f345425", "evidence": "chart/bar-horizontal: Dual bar chart comparing job growth (left) and median wage (right) by occupation.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.254642+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 131, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-2a2855e78437", "evidence": "Title 'It is also still early, only 2 million industrial robots in the world'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 132, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "eb35aeac-5d14-4b0c-b983-bd1a0f3bd388", "evidence": "chart/bar-vertical: Estimated worldwide operational stock of industrial robots 2015-2016 and forecast for 2017*-2020*", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.315085+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 132, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-2eb49b036ec4", "evidence": "Title 'There are fewer robots in U.S. factories compared to other advanced economies'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 133, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Slide recipe: the peer-gap chart", "slug": "peer-gap-chart-recipe", "status": "active", "bestFor": null, "canonId": "019df269-2b04-75bc-ad85-e0179264bdd7", "version": 1, "bodyDocId": "019df269-2ae1-771d-9cd1-193a2e25f21c", "description": "The single most-copied slide in activism. This is the exact layout, used", "familyLabel": "slide-recipe", "categoryName": "Slide", "categorySlug": "slide" }, "agent": "designer", "layer": "slide", "agents": [ "designer" ], "matchId": "d481df83-bd91-4c66-944d-f51df0d7ab62", "evidence": "chart/bar-horizontal: Robots per 10000 manufacturing employees", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:43.375047+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 133, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "87576d9a-eec1-4094-8387-8a753dedd938", "evidence": "chart/line: Line chart showing exponential growth in the computer subsector compared to flat growth in other manufacturing sectors.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.434951+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 134, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "26147f2d-bba1-41d9-9817-ba71c0e25157", "evidence": "chart/bar-grouped: Change in employees and robots at AMZN and AMZN-related retail", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.494859+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 135, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-313effc0b6a5", "evidence": "Title 'Basic Income trials roll out'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 137, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "b2a38234-4191-4e53-a415-27fee0ba17ee", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.983879+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 137, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-34b742690b57", "evidence": "Title 'Universal Basic Services proposed by UCL researchers' with £42.16Bn.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 138, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-38a83ee7ef14", "evidence": "Title 'AI Nationalism: flurry of National AI strategies announced' with $150B China.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 139, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "1b3854e3-fccf-4bd8-96a0-a8c39c260340", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.083649+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 139, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-40cadf752666", "evidence": "Title repeats AI Nationalism with €1.5B France figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 140, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-45a83a93428b", "evidence": "$2B Korea, callout 'global top 4 by 2022'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 141, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "2eab976a-16b0-499f-9380-2152fb7b8013", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.184084+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 141, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-4a6a305b0f82", "evidence": "€20B EU AI investment figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 142, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "b994cc10-2e9e-461e-adff-8e23af92bdb3", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.251146+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 142, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-4fdee9adf969", "evidence": "£603M UK strategy figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 143, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "3e9cc291-6a72-44a2-92ea-999cf1141c1e", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.316262+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 143, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-53f908119754", "evidence": "Title 'For now, US leads China on almost every measure other than data'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 144, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-5f5053365ad2", "evidence": "Title 'America increasingly using CFIUS to scrutinise foreign acquisitions'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 145, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-63ae8fd05bb4", "evidence": "Title 'China's annual imports of semiconductors have risen to $260 billion'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 147, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-66bf07ca8af2", "evidence": "Title 'China's semiconductor industry is small compared to that of the U.S.'", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 148, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "980b92b3-ea06-45f0-8a45-ae88357d2ea5", "evidence": "chart/bar-vertical: Cross-border semiconductor M&As by Chinese companies (in billions of dollars)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.124149+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 149, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-6bd1ec0b98ad", "evidence": "Title 'Fake views: Generating synthetic video is getting cheaper, easier and more realistic'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 150, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Emotional Appeal", "slug": "emotional-appeal", "status": "active", "bestFor": null, "canonId": "019dd956-bde9-7668-8be7-0097678fec59", "version": 1, "bodyDocId": null, "description": "Connecting to audience feelings rather than just logic", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-6cd5adb00ef3", "evidence": "Disinformation/propaganda framing.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 150, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "a4270a2d-6749-457c-b444-cab81c5af940", "evidence": "Section 5: Predictions", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:38.24304+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 151, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-737871a8f4f5", "evidence": "Title '8 predictions for the next 12 months'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 152, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Closing Techniques", "slug": "closing-techniques", "status": "active", "bestFor": null, "canonId": "019dd956-af13-774b-a4fd-7db0a3a88d51", "version": 1, "bodyDocId": null, "description": "Strong endings: Callback, challenge, vision, clear CTA, memorable line", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-7917deca52c5", "evidence": "Forward-looking close to deck.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 152, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "bbd0278b-42fa-4cd3-ba91-6789e363871d", "evidence": "list/numbered: 1. A lab located in China makes a significant research breakthrough.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.323547+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 152, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "27fb5db9-f574-4ad3-bfa3-0b4448c07ae3", "evidence": "Section 6: Conclusion", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:38.401609+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 153, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Closing Techniques", "slug": "closing-techniques", "status": "active", "bestFor": null, "canonId": "019dd956-af13-774b-a4fd-7db0a3a88d51", "version": 1, "bodyDocId": null, "description": "Strong endings: Callback, challenge, vision, clear CTA, memorable line", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-7fb92e7ae0ce", "evidence": "Thanks slide invites feedback — explicit CTA.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 154, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Storytelling: the closing ask", "slug": "closing-ask", "status": "active", "bestFor": null, "canonId": "019df269-24e0-70ab-8dd9-983832c1c665", "version": 1, "bodyDocId": "019df269-24bc-7554-b88d-245127606b8d", "description": "The single most important slide in the deck — after the cover. The final", "familyLabel": "narrative-block", "categoryName": "Narrative & Transformation", "categorySlug": "narrative-transformation" }, "agent": "storyteller", "layer": "slide", "agents": [ "storyteller" ], "matchId": "09768c91-5896-492b-8b2e-eec9533a8129", "evidence": "callout: We would appreciate any and all feedback on how we could improve this report further.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:38.483467+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 154, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Slide recipe: Cover", "slug": "cover-slide", "status": "active", "bestFor": null, "canonId": "019df269-2a08-767d-9128-e7431e6fe8a3", "version": 1, "bodyDocId": "019df269-29e5-7738-b5fa-6892e5dc8e65", "description": "The first ten seconds. Build it deliberately.", "familyLabel": "slide-recipe", "categoryName": "Slide", "categorySlug": "slide" }, "agent": "designer", "layer": "slide", "agents": [ "designer" ], "matchId": "06b05a93-8683-408e-9a39-df7027d63caa", "evidence": "State of AI", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:38.602577+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 156, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [ { "matchId": "edf29fe1-20fd-4386-8312-0bcb6de22219", "evidence": "The charts demonstrate the empirical relationship between data size, loss, and parameter count.", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:43:53.446+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 12, "frameworkName": "scaling-laws" }, { "matchId": "7a8a46ae-f108-4b15-aebe-a15ca8d09a13", "evidence": "The slide describes the process of memorizing data and then compressing it to retain only relevant information for prediction.", "framework": null, "confidence": 1, "extraction": { "at": "2026-05-02 17:43:53.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 13, "frameworkName": "Information Bottleneck Theory" }, { "matchId": "d21e7740-bdff-4de4-bb8d-090d61722bd3", "evidence": "The diagram depicts a self-reinforcing cycle of Bigger model -> Better performance -> More money -> More user data -> Bigger model.", "framework": null, "confidence": 1, "extraction": { "at": "2026-05-02 17:43:53.541+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 14, "frameworkName": "flywheel" }, { "matchId": "52a32db1-36bf-415a-a9a6-8bebd8ab1b96", "evidence": "The diagram depicts a cyclical process with feedback loops between taxonomy, crowd-acting, model training, and testing.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:43:58.847+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 33, "frameworkName": "flywheel" }, { "matchId": "c14f3493-0f12-432c-9d39-da62e5118bb9", "evidence": "Diagram showing Agent-Environment interaction via Observations, Rewards, and Actions", "framework": null, "confidence": 1, "extraction": { "at": "2026-05-02 17:43:57.489+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 36, "frameworkName": "Reinforcement Learning Loop" }, { "matchId": "76a3b04a-5ab0-4bee-a49a-8b29c22d5d22", "evidence": "Diagram shows VAE, MDN-RNN, and Controller (C) architecture", "framework": null, "confidence": 1, "extraction": { "at": "2026-05-02 17:43:57.888+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 39, "frameworkName": "World Models" }, { "matchId": "019dd95a-1ca5-70bb-bba0-69200c8b5702", "evidence": "Title declares 5-type taxonomy; columns: denigration/stereotype/recognition/under-representation/ex-nomination.", "framework": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:34.826634+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "frameworkName": "5 Types of Allocation Bias Matrix" }, { "matchId": "f5ba34a6-11e2-4604-a90c-a3c512b9069f", "evidence": "The slide uses a grid to categorize examples against defined bias types.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:43:59.577+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "frameworkName": "matrix-nxn" }, { "matchId": "08e6e216-27de-40a6-944e-f08ffa5929a6", "evidence": "Chronological presentation of model architectures from 2013 to 2016", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:43:59.258+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 51, "frameworkName": "timeline" }, { "matchId": "68bb0cfe-265f-4656-b66e-a8a100f5dfa3", "evidence": "The slide depicts a step-by-step technical workflow.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:43:59.907+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 55, "frameworkName": "process" }, { "matchId": "8e39347e-fb1f-40b5-9d61-a67c12e37c04", "evidence": "Visual representation of geographic talent flows", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:44:01.668+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 59, "frameworkName": "map-spatial" }, { "matchId": "9a0612bd-5f06-4635-ad0f-c3d034d27ee8", "evidence": "Uses reputable news sources to validate a market trend.", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:44:00.645+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 66, "frameworkName": "authority-citation" }, { "matchId": "1f7b570e-7da8-4f46-8451-1a468634375f", "evidence": "The slide depicts the sequential stages of drug development from preclinical to clinical approval.", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 74, "frameworkName": "process" }, { "matchId": "cb3bd79d-b718-4a68-8522-fa71e2c3132f", "evidence": "The slide uses a 'Why now?' structure to diagnose the drivers of a market trend.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:44:04.352+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 91, "frameworkName": "diagnostic-test" }, { "matchId": "c18cd0bb-0678-439b-8984-6d0e00470c11", "evidence": "The slide presents two opposing viewpoints ('Don't worry' vs 'Worry') to frame the debate.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:44:08.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 130, "frameworkName": "comparison_frame" }, { "matchId": "019dd95a-1ca5-70bb-bba0-6c779bc4c430", "evidence": "Side-by-side table across Hardware, Data, Research, Commercial drivers.", "framework": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:34.826634+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 144, "frameworkName": "US vs China AI Capability Benchmark" } ], "patterns": [] }, "storymakers": { "model": "claude-cli-sonnet-4.6", "backend": null, "rawJson": { "arc": { "id": 11, "slug": "triple-take", "beats": [ { "name": "The Facts (What)", "end_page": 69, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "start_page": 4 }, { "name": "The Implications (So What)", "end_page": 150, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "start_page": 70 }, { "name": "The Action (Now What)", "end_page": 156, "evidence": "Section 5 '8 predictions for the next 12 months' and conclusion.", "position": 3, "start_page": 151 } ], "evidence": "Six-section state-of-industry report walks Facts (Research/Talent pp.6-69) -> Implications (Industry/Politics pp.70-150) -> Action (Predictions pp.151-152).", "confidence": 78 }, "loops": [ { "id": 3, "slug": "03-aha-moment", "end_page": 9, "evidence": "Defines transfer learning then reveals model beats 21 dermatologists with ROC charts.", "position": 1, "objective": "Show transfer learning enables superhuman skin-cancer detection", "confidence": 80, "start_page": 7 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 21, "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "start_page": 10 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 29, "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "start_page": 22 }, { "id": 6, "slug": "06-zoom-in", "end_page": 35, "evidence": "Big picture (CV detects nouns) -> focus on verbs/common sense -> specific case (multi-viewpoint scenes).", "position": 4, "objective": "Zoom from object detection to full scene understanding", "confidence": 75, "start_page": 30 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 39, "evidence": "AlphaZero, OpenAI Dota2, world-model RL — three parallel evidence pieces of RL progress.", "position": 5, "objective": "Show RL agents now beat human experts across domains", "confidence": 78, "start_page": 36 }, { "id": 6, "slug": "06-zoom-in", "end_page": 43, "evidence": "Big picture (fairness) -> specific examples (Turkish gender, racial) -> 5-type allocation bias table.", "position": 6, "objective": "Diagnose ML bias by isolating its sources", "confidence": 75, "start_page": 40 }, { "id": 10, "slug": "10-iceberg", "end_page": 50, "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "start_page": 44 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 55, "evidence": "Architecture iteration -> AutoML automating engineers -> federated learning -> consensus mechanism.", "position": 8, "objective": "Catalogue frontier research trends (AutoML, federated)", "confidence": 70, "start_page": 51 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 64, "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "start_page": 57 }, { "id": 27, "slug": "27-cost-of-inaction", "end_page": 67, "evidence": "Salaries climb -> $300-500k base anecdotes -> $120M Google self-driving boss as tipping point.", "position": 10, "objective": "Quantify escalating talent compensation arms race", "confidence": 70, "start_page": 65 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 73, "evidence": "Patent filings concentrated -> R&D spend leadership -> cloud APIs -> TensorFlow framework war.", "position": 11, "objective": "Show GAFAMBAT capture AI infrastructure layer", "confidence": 75, "start_page": 71 }, { "id": 14, "slug": "14-day-in-life", "end_page": 80, "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "start_page": 74 }, { "id": 6, "slug": "06-zoom-in", "end_page": 85, "evidence": "China surveillance -> Project Maven -> Cambridge Analytica -> privacy preservation tooling.", "position": 13, "objective": "Zoom into surveillance/privacy as the dark side of AI", "confidence": 68, "start_page": 81 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 108, "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "start_page": 86 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 122, "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "start_page": 110 }, { "id": 2, "slug": "02-pattern-hunter", "end_page": 129, "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "start_page": 123 }, { "id": 27, "slug": "27-cost-of-inaction", "end_page": 138, "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "start_page": 130 }, { "id": 15, "slug": "15-why-now", "end_page": 143, "evidence": "Five contiguous slides with $150B China, €1.5B France, $2B Korea, €20B EU, £603M UK strategies.", "position": 18, "objective": "Argue national AI strategies create urgent geopolitical window", "confidence": 75, "start_page": 139 }, { "id": 4, "slug": "04-tale-two-worlds", "end_page": 149, "evidence": "Side-by-side US/China table on every dimension -> CFIUS scrutiny -> China import gap -> chip industry size.", "position": 19, "objective": "Compare US vs China AI capability head to head", "confidence": 80, "start_page": 144 }, { "id": 35, "slug": "35-scenario-fork", "end_page": 152, "evidence": "Numbered list of 8 distinct 'if X happens' predictions for next 12 months.", "position": 20, "objective": "Project eight forward scenarios for the year ahead", "confidence": 70, "start_page": 152 } ], "slide_tools": [ { "tools": [ { "id": 132, "slug": "visual-hierarchy", "layer": "Slide", "evidence": "Section navigation pipe-bar header repeats across deck.", "confidence": 70 }, { "id": 116, "slug": "chunking", "layer": "Loop", "evidence": "Six numbered sections chunk a 156-page report.", "confidence": 75 } ], "page_number": 2 }, { "tools": [ { "id": 154, "slug": "concrete-language", "layer": "Slide", "evidence": "Definitions slide grounds abstract terms before use.", "confidence": 65 } ], "page_number": 4 }, { "tools": [ { "id": 121, "slug": "metaphor-analogy", "layer": "Slide", "evidence": "Callout 'transferred' frames knowledge as portable cargo.", "confidence": 70 }, { "id": 80, "slug": "data-story-arc", "layer": "Loop", "evidence": "First slide of context->conflict->insight aha-moment loop.", "confidence": 65 } ], "page_number": 7 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title states 'From predicting everyday objects on ImageNet to detecting skin cancer'.", "confidence": 90 }, { "id": 157, "slug": "sinatra-test", "layer": "Slide", "evidence": "Single flagship example (skin cancer) carries the transfer-learning argument.", "confidence": 75 }, { "id": 138, "slug": "annotation", "layer": "Slide", "evidence": "Callout points to red dermatologist points below blue model curve.", "confidence": 70 } ], "page_number": 9 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'AI hardware as the new frontier' states position not topic.", "confidence": 88 }, { "id": 74, "slug": "inductive-reasoning", "layer": "Loop", "evidence": "First slide of pattern_hunter loop assembling hardware evidence.", "confidence": 70 } ], "page_number": 10 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Headline 'The hardware war: More GPUs allows for faster training'.", "confidence": 85 } ], "page_number": 11 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'AI hardware is especially helpful for deep learning'.", "confidence": 85 } ], "page_number": 12 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title quantifies '300,000x' compute growth driving Go breakthrough.", "confidence": 85 }, { "id": 195, "slug": "weber-fechner-law", "layer": "Slide", "evidence": "300,000x figure designed to register magnitude shock.", "confidence": 55 } ], "page_number": 14 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'No wonder GPUs have grown immensely in popularity'.", "confidence": 88 }, { "id": 129, "slug": "chart-selection-guide", "layer": "Slide", "evidence": "Growth metric '+15X' implies trend chart.", "confidence": 65 } ], "page_number": 15 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title states '242x' performance from new architectures.", "confidence": 85 } ], "page_number": 19 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'GPUs and novel silicon are costly to rent... limited by financial resources'.", "confidence": 88 }, { "id": 68, "slug": "loss-aversion", "layer": "Block", "evidence": "Frames cost as constraint that risks excluding players.", "confidence": 55 } ], "page_number": 20 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title makes head-to-head TPUv2 vs V100 economic claim.", "confidence": 88 }, { "id": 78, "slug": "contrast-pairs", "layer": "Loop", "evidence": "TPUv2 vs V100 explicit contrast pair.", "confidence": 60 } ], "page_number": 21 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'NVIDIA's datacenter business breaks $2B run-rate, growing >100% YoY'.", "confidence": 90 } ], "page_number": 24 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'NVIDIA's enterprise value has 10x in 3 years'.", "confidence": 90 } ], "page_number": 25 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title quantifies Intel datacenter 30% revenue share.", "confidence": 88 } ], "page_number": 26 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Many companies are developing custom AI chips'.", "confidence": 80 }, { "id": 133, "slug": "small-multiples", "layer": "Slide", "evidence": "Market landscape implies many parallel logos/cards.", "confidence": 60 } ], "page_number": 27 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Large technology companies are hedging their hardware suppliers'.", "confidence": 85 } ], "page_number": 28 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title quantifies $76bn cloud capex by 2020.", "confidence": 90 } ], "page_number": 29 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Traditional computer vision describes scenes by detecting objects (nouns)'.", "confidence": 85 }, { "id": 121, "slug": "metaphor-analogy", "layer": "Slide", "evidence": "Nouns/verbs metaphor frames CV vs scene understanding.", "confidence": 75 } ], "page_number": 30 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'detecting objects in images is not enough to produce real scene understanding'.", "confidence": 88 } ], "page_number": 31 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'True scene understanding requires understanding actions (verbs) and common sense'.", "confidence": 88 }, { "id": 121, "slug": "metaphor-analogy", "layer": "Slide", "evidence": "Verbs vs nouns metaphor extended.", "confidence": 80 } ], "page_number": 32 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title describes dataset-building approach.", "confidence": 80 } ], "page_number": 33 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Deep learning models can actually understand the verbs as well as the nouns'.", "confidence": 88 } ], "page_number": 34 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Machines can also understand visual scenes by learning from multiple viewpoints'.", "confidence": 85 } ], "page_number": 35 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'AlphaZero showed that a deep RL system can learn from scratch to beat Go champions'.", "confidence": 88 }, { "id": 157, "slug": "sinatra-test", "layer": "Slide", "evidence": "Flagship case (AlphaZero, 40 days) carries the RL claim.", "confidence": 75 } ], "page_number": 37 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'OpenAI's multi-agent RL system learns to play Dota2'.", "confidence": 85 } ], "page_number": 38 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'RL agents can also build their own world models'.", "confidence": 80 } ], "page_number": 39 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title posed as question 'How do we ensure our models are not biased?'.", "confidence": 85 } ], "page_number": 40 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'An example of biased machine learning systems: Stereotyping'.", "confidence": 85 }, { "id": 154, "slug": "concrete-language", "layer": "Slide", "evidence": "Specific Turkish-Google Translate example.", "confidence": 80 }, { "id": 224, "slug": "singularity-effect", "layer": "Slide", "evidence": "Single concrete bias example over abstract stats.", "confidence": 65 } ], "page_number": 41 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Another example of biased machine learning systems: Racial bias'.", "confidence": 85 }, { "id": 156, "slug": "emotional-appeal", "layer": "Slide", "evidence": "Racial-bias example designed for moral charge.", "confidence": 70 } ], "page_number": 42 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'ML models have 5 types of allocation bias that stem from training data'.", "confidence": 90 }, { "id": 2, "slug": "mece-principle", "layer": "Block", "evidence": "5-column matrix presents bias categories as MECE table.", "confidence": 70 } ], "page_number": 43 }, { "tools": [ { "id": 121, "slug": "metaphor-analogy", "layer": "Slide", "evidence": "Callout 'black boxes' metaphor for opaque models.", "confidence": 80 }, { "id": 67, "slug": "curiosity-gap", "layer": "Block", "evidence": "Frames problem ('understanding them is hard') to set up later reveal.", "confidence": 60 } ], "page_number": 44 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Explainability helps validate that ML models perform well for the right reasons'.", "confidence": 85 } ], "page_number": 45 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Next step: Justifying decisions in plain language'.", "confidence": 80 } ], "page_number": 46 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Imperceptible changes to data can alter a deep learning model's prediction' with 99.3% stat.", "confidence": 85 }, { "id": 152, "slug": "unexpected-pattern", "layer": "Slide", "evidence": "Adversarial example breaks expectation that small change = small effect.", "confidence": 70 } ], "page_number": 48 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Same title; toaster patch example.", "confidence": 80 }, { "id": 154, "slug": "concrete-language", "layer": "Slide", "evidence": "Toaster patch is concrete, vivid example.", "confidence": 75 } ], "page_number": 49 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Adversarial attacks present serious safety challenges in the real world'.", "confidence": 88 }, { "id": 156, "slug": "emotional-appeal", "layer": "Slide", "evidence": "Pedestrian-at-zebra-crossing made invisible — visceral.", "confidence": 70 } ], "page_number": 50 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title with '80%' stat on accuracy reductions.", "confidence": 80 } ], "page_number": 52 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'AI to automate away AI engineers'.", "confidence": 88 }, { "id": 152, "slug": "unexpected-pattern", "layer": "Slide", "evidence": "Recursive twist — AI replacing AI engineers — surprises.", "confidence": 70 } ], "page_number": 53 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Distributed federated learning to decentralise data acquisition'.", "confidence": 80 } ], "page_number": 54 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title repeats federated framing; process diagram.", "confidence": 75 } ], "page_number": 55 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Element AI estimates 22,000 PhD educated AI researchers'.", "confidence": 88 }, { "id": 74, "slug": "inductive-reasoning", "layer": "Loop", "evidence": "First slide of talent-concentration pattern_hunter loop.", "confidence": 70 } ], "page_number": 57 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title '5,000 high-level researchers worldwide'.", "confidence": 85 } ], "page_number": 58 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'America remains the hub for talent exchange'.", "confidence": 85 } ], "page_number": 59 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Google is widely acknowledged as the leading employer of AI talent'.", "confidence": 85 } ], "page_number": 60 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title quantifies '6.3% of 2017 ICML papers had a Google/DeepMind author'.", "confidence": 88 } ], "page_number": 61 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Percentage of ICML papers with Google/DeepMind author doubles'.", "confidence": 88 } ], "page_number": 62 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Google lead in contribution to 2017 NIPS papers'.", "confidence": 85 } ], "page_number": 63 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Google & DeepMind dominate NIPS authorship'.", "confidence": 88 }, { "id": 50, "slug": "pareto-principle-80-20", "layer": "Block", "evidence": "Top-11 institutions ranked — concentration story.", "confidence": 60 } ], "page_number": 64 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Anecdotally salaries continue to grow'; $300-500k quote.", "confidence": 80 }, { "id": 154, "slug": "concrete-language", "layer": "Slide", "evidence": "Specific salary band rather than abstract 'high pay'.", "confidence": 75 } ], "page_number": 66 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Compensation can be astronomical and relationships litigious'.", "confidence": 90 }, { "id": 158, "slug": "von-restorff-effect", "layer": "Slide", "evidence": "$120M figure stands out, designed to shock.", "confidence": 70 }, { "id": 122, "slug": "story-moments", "layer": "Slide", "evidence": "Single anecdote functions as 'the shock' moment.", "confidence": 60 } ], "page_number": 67 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Topic-with-stat title '17%' on diversity.", "confidence": 75 } ], "page_number": 68 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'AI takes the world stage: GAFAMBAT are in the ring together'.", "confidence": 88 }, { "id": 121, "slug": "metaphor-analogy", "layer": "Slide", "evidence": "Boxing 'in the ring' metaphor.", "confidence": 75 }, { "id": 129, "slug": "chart-selection-guide", "layer": "Slide", "evidence": "Stacked bar (patents) + horizontal bar (R&D spend).", "confidence": 80 }, { "id": 132, "slug": "visual-hierarchy", "layer": "Slide", "evidence": "Tech vs non-tech color-coded for emphasis.", "confidence": 70 } ], "page_number": 71 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Big cloud providers... building blocks of intelligence via API'.", "confidence": 80 } ], "page_number": 72 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Google's TensorFlow is winning the ML framework war'.", "confidence": 88 }, { "id": 121, "slug": "metaphor-analogy", "layer": "Slide", "evidence": "'Framework war' metaphor.", "confidence": 70 } ], "page_number": 73 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Why now? Today's drug development process is too slow and expensive'.", "confidence": 88 }, { "id": 27, "slug": "problem-agitate-solve-pas", "layer": "Block", "evidence": "Why-now problem framing opens a P-A-S sequence for healthcare.", "confidence": 60 } ], "page_number": 74 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Healthcare systems worldwide are costly and overburdened'.", "confidence": 85 } ], "page_number": 76 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Breast cancer as a case study: Not enough doctors, diagnosing is hard'.", "confidence": 88 }, { "id": 157, "slug": "sinatra-test", "layer": "Slide", "evidence": "Breast cancer flagship case proves healthcare AI value.", "confidence": 70 } ], "page_number": 77 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Expect to see more activity as companies move products through trials'.", "confidence": 80 } ], "page_number": 80 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Population-level surveillance is taking off in China' with 400M figure.", "confidence": 85 } ], "page_number": 81 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Concrete count 2,600,968,280 in title indicates breach scale.", "confidence": 75 }, { "id": 154, "slug": "concrete-language", "layer": "Slide", "evidence": "Exact integer '2,600,968,280' chosen over rounded version.", "confidence": 70 } ], "page_number": 84 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Why now? Satellite data is decreasing in cost and increasing in resolution'.", "confidence": 88 }, { "id": 74, "slug": "inductive-reasoning", "layer": "Loop", "evidence": "First slide of vertical-applications pattern_hunter loop.", "confidence": 65 } ], "page_number": 86 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Topic title 'Cybersecurity' with $11.7M cost stat.", "confidence": 70 } ], "page_number": 89 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Topic 'Warehouse automation' with 120,000 figure.", "confidence": 70 } ], "page_number": 91 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout drives why-now: sensor cost down, vision better, robots cheaper than labour.", "confidence": 75 } ], "page_number": 94 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout '70% more food calories needed by 2050'.", "confidence": 80 } ], "page_number": 97 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Fewer farm workers... higher wages... but more automation leads to stable labor cost share'.", "confidence": 85 } ], "page_number": 98 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Topic 'Finance' with 7x stat.", "confidence": 70 } ], "page_number": 104 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout '2/3 day is unstructured or unpredictable' frames automation gap.", "confidence": 80 } ], "page_number": 106 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Material science' with callout '167x' R&D speedup.", "confidence": 88 }, { "id": 158, "slug": "von-restorff-effect", "layer": "Slide", "evidence": "167x stat designed to stand out.", "confidence": 60 } ], "page_number": 108 }, { "tools": [ { "id": 154, "slug": "concrete-language", "layer": "Slide", "evidence": "Sets up two specific surveys to be referenced.", "confidence": 60 }, { "id": 74, "slug": "inductive-reasoning", "layer": "Loop", "evidence": "Opens public-attitudes pattern_hunter loop.", "confidence": 70 } ], "page_number": 110 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Growing awareness of automation impacting jobs' with 77%.", "confidence": 85 } ], "page_number": 111 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Young, part-time, hispanic and lower-income Americans report most impact'.", "confidence": 85 }, { "id": 34, "slug": "segmentation-split", "layer": "Loop", "evidence": "Slide segments by demographic to surface differential impact.", "confidence": 70 } ], "page_number": 112 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Rising concerns about automation increasing inequality'.", "confidence": 85 } ], "page_number": 113 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Those whose job has been impacted favor more radical policies'.", "confidence": 85 } ], "page_number": 114 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout 'Overall optimism around AI' with 41% stat.", "confidence": 75 } ], "page_number": 115 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout '...expectation that AI will make my life easier'.", "confidence": 75 } ], "page_number": 116 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title '...but expectation that AI will reduce privacy'.", "confidence": 80 }, { "id": 78, "slug": "contrast-pairs", "layer": "Loop", "evidence": "Optimism-then-fear tension across consecutive slides.", "confidence": 65 } ], "page_number": 117 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout 'Americans believe US is currently the world leader'.", "confidence": 80 } ], "page_number": 121 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title '...China will close the gap over the next ten years'.", "confidence": 85 } ], "page_number": 122 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Despite increased automation the US unemployment rate is at a 17 year low'.", "confidence": 85 }, { "id": 152, "slug": "unexpected-pattern", "layer": "Slide", "evidence": "Counter-intuitive pairing of automation rise + low unemployment.", "confidence": 65 } ], "page_number": 123 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout 'Routine jobs have stagnated'.", "confidence": 80 } ], "page_number": 124 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout 'Wages have lagged the increase in jobs'.", "confidence": 80 } ], "page_number": 125 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout 'marked change in how long unemployment lasts for'.", "confidence": 80 } ], "page_number": 126 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout 'Labour productivity and hourly compensation have diverged'.", "confidence": 80 } ], "page_number": 127 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout 'Labour's share of income has been declining steadily'.", "confidence": 80 } ], "page_number": 128 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Callout 'Workers are experiencing greater income volatility'.", "confidence": 80 } ], "page_number": 129 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Question title 'How much of this is due to automation?'", "confidence": 75 }, { "id": 67, "slug": "curiosity-gap", "layer": "Block", "evidence": "Open question opens analytical loop.", "confidence": 60 } ], "page_number": 130 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'It is also still early, only 2 million industrial robots in the world'.", "confidence": 85 } ], "page_number": 132 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'There are fewer robots in U.S. factories compared to other advanced economies'.", "confidence": 88 }, { "id": 39, "slug": "benchmark-gap", "layer": "Loop", "evidence": "Compares US to peer economies on robot density.", "confidence": 75 } ], "page_number": 133 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Basic Income trials roll out'.", "confidence": 80 } ], "page_number": 137 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Universal Basic Services proposed by UCL researchers' with £42.16Bn.", "confidence": 85 } ], "page_number": 138 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'AI Nationalism: flurry of National AI strategies announced' with $150B China.", "confidence": 88 }, { "id": 74, "slug": "inductive-reasoning", "layer": "Loop", "evidence": "First slide of why-now AI nationalism evidence stack.", "confidence": 65 } ], "page_number": 139 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title repeats AI Nationalism with €1.5B France figure.", "confidence": 80 } ], "page_number": 140 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "$2B Korea, callout 'global top 4 by 2022'.", "confidence": 80 } ], "page_number": 141 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "€20B EU AI investment figure.", "confidence": 80 } ], "page_number": 142 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "£603M UK strategy figure.", "confidence": 80 } ], "page_number": 143 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'For now, US leads China on almost every measure other than data'.", "confidence": 90 }, { "id": 78, "slug": "contrast-pairs", "layer": "Loop", "evidence": "Side-by-side US/China table — explicit contrast pair.", "confidence": 80 }, { "id": 2, "slug": "mece-principle", "layer": "Block", "evidence": "Driver categories (Hardware/Data/Research/Commercial) structure the table.", "confidence": 65 } ], "page_number": 144 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'America increasingly using CFIUS to scrutinise foreign acquisitions'.", "confidence": 88 } ], "page_number": 145 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'China's annual imports of semiconductors have risen to $260 billion'.", "confidence": 88 } ], "page_number": 147 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'China's semiconductor industry is small compared to that of the U.S.'", "confidence": 85 } ], "page_number": 148 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title 'Fake views: Generating synthetic video is getting cheaper, easier and more realistic'.", "confidence": 85 }, { "id": 156, "slug": "emotional-appeal", "layer": "Slide", "evidence": "Disinformation/propaganda framing.", "confidence": 65 } ], "page_number": 150 }, { "tools": [ { "id": 118, "slug": "action-titles", "layer": "Slide", "evidence": "Title '8 predictions for the next 12 months'.", "confidence": 80 }, { "id": 116, "slug": "chunking", "layer": "Loop", "evidence": "8 numbered predictions chunked into discrete bullets.", "confidence": 75 }, { "id": 124, "slug": "closing-techniques", "layer": "Slide", "evidence": "Forward-looking close to deck.", "confidence": 65 } ], "page_number": 152 }, { "tools": [ { "id": 124, "slug": "closing-techniques", "layer": "Slide", "evidence": "Thanks slide invites feedback — explicit CTA.", "confidence": 70 } ], "page_number": 154 } ], "slides_seen": 156, "deck_summary": "Air Street Capital's State of AI 2018 is a 156-page evidence-stacking report that loosely follows a Triple Take (Facts -> Implications -> Predictions) with strong Storymakers alignment on action titles and pattern-hunter loops, but is structured by topic rather than narrative arc — no single hero, no resolution, and limited use of contrast/before-after framing typical of consulting decks.", "secondary_arcs": [ { "id": 12, "slug": "onion", "beats": [ { "name": "Surface Observation", "end_page": 9, "evidence": "Transfer learning headline result on skin cancer.", "position": 1, "start_page": 6 }, { "name": "First Layer", "end_page": 55, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "start_page": 10 }, { "name": "Deeper Layer", "end_page": 108, "evidence": "Talent and industry deployment across verticals.", "position": 3, "start_page": 56 }, { "name": "Core Insight", "end_page": 150, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "start_page": 109 }, { "name": "Implications", "end_page": 156, "evidence": "Predictions for next 12 months.", "position": 5, "start_page": 151 } ], "evidence": "Each section peels back one layer of AI: surface tech -> who builds it -> where deployed -> societal core -> implications.", "confidence": 55 } ], "images_inspected": 6, "slide_frameworks": [ { "frameworks": [ { "name": "5 Types of Allocation Bias Matrix", "slug": "allocation-bias-matrix", "evidence": "Title declares 5-type taxonomy; columns: denigration/stereotype/recognition/under-representation/ex-nomination.", "confidence": 70 } ], "page_number": 43 }, { "frameworks": [ { "name": "US vs China AI Capability Benchmark", "slug": "us-china-ai-benchmark", "evidence": "Side-by-side table across Hardware, Data, Research, Commercial drivers.", "confidence": 75 } ], "page_number": 144 } ] }, "matchedAt": "2026-04-26 02:11:06+00", "slidesSeen": 156, "deckSummary": "Air Street Capital's State of AI 2018 is a 156-page evidence-stacking report that loosely follows a Triple Take (Facts -> Implications -> Predictions) with strong Storymakers alignment on action titles and pattern-hunter loops, but is structured by topic rather than narrative arc — no single hero, no resolution, and limited use of contrast/before-after framing typical of consulting decks.", "imagesInspected": 6, "extractionSeconds": 477.94073 }, "score": { "backend": "codex", "scoredAt": "2026-05-02 22:58:56.659+00", "subScores": { "scqa_arc": 43, "action_titles": 66, "mece_structure": 60, "closing_strength": 35, "evidence_quality": 84, "clarity_of_thesis": 62, "production_quality": 68, "visual_storytelling": 78 }, "totalScore": 62, "coveragePct": 100, "explanations": { "scqa_arc": "The flow establishes a broad AI situation and many complications, but it mostly reads as an analytical survey rather than a clear Situation-Complication-Question-Answer arc with one answer.", "action_titles": "The deck has many insight-bearing titles, such as slide 24, \"NVIDIA's datacenter business breaks $2B run-rate, is growing >100% year on year and accounts for almost 20% of their group revenue,\" but many slides remain topic labels.", "mece_structure": "The top-level sections on slides 6, 56, 70, 109, and 151 give a readable structure, but Research, Industry, and Politics overlap and the industry vertical tour is selective rather than exhaustive.", "closing_strength": "Slide 152 provides predictions, but slides 153-154 close with a divider and thanks rather than a clear recommendation, decision, or next-step agenda.", "evidence_quality": "Claims are generally backed by data, charts, cited surveys, market figures, and precedent examples, especially across slides 57-67, 71-80, and 111-149.", "clarity_of_thesis": "Slide 2 identifies AI as a technological force multiplier and lays out Research, Talent, Industry, and Politics, but the first 5 slides do not state a sharp declarative thesis.", "production_quality": "Production is solid and consistent for a long report, but only 98 of 156 slides have explicit action titles and the inventory flags many dense or overcrowded pages such as slides 9, 12, 31, 68, and 98.", "visual_storytelling": "The deck usually matches message to format, with charts, maps, tables, frameworks, case examples, and big metrics used effectively on slides such as 14, 57, 73, 98, and 144." }, "slidesAnalyzed": 156 }, "review": { "backend": null, "verdict": "This is a strong research compendium with useful evidence runs, but only a mid-tier Storymakers exemplar because it prioritizes coverage over thesis-led narrative.", "reviewedAt": "2026-05-01 22:01:04.475+00", "slidesSeen": 156, "suggestions": [ "Add an executive thesis slide immediately after the cover or agenda that states the 2018 State of AI answer in one sentence and previews the four tensions the deck will prove.", "Rewrite section dividers as answer-bearing pillars, for example 'Research progress is now gated by compute, robustness, and interpretability' instead of 'Section 1: Research and technical breakthroughs.'", "Compress the Industry section into a MECE opportunity matrix or ranked set of adoption archetypes, then use sector pages as evidence rather than a long topic tour.", "Replace the final 'Thanks!' close with a synthesis slide that turns the 8 predictions into implications for founders, investors, policymakers, and incumbents." ], "closingScore": 56, "openingScore": 38, "topStrengths": [ "Clear macro structure around Research, Talent, Industry, Politics, Predictions, and Conclusion (slides 6, 56, 70, 109, 151, 153) gives readers a stable map.", "Several evidence runs build cumulative logic, especially AI hardware from compute constraints to cost, vendors, and capex (slides 10-29).", "The Politics section creates real tension by moving from public attitudes (slides 110-122) to labor-market facts (slides 123-136), policy responses (slides 137-138), and AI nationalism/geopolitics (slides 139-150)." ], "topWeaknesses": [ "The deck does not lead with a governing thesis; the first five slides are cover, agenda, author bios, and definitions rather than an answer.", "The Industry chapter is a broad catalog of use cases and sectors (slides 74-108), with repeated 'Where and how is machine learning being used effectively?' prompts rather than a ranked argument.", "The close is underpowered: slide 152 lists predictions and slide 154 asks for feedback, but there is no memorable synthesis of what decision-makers should do next." ], "narrativeScore": 64, "pillarCritique": "The section dividers are broadly MECE at the chapter level: Research, Talent, Industry, Politics, Predictions, and Conclusion (slides 6, 56, 70, 109, 151, 153). Inside the pillars, especially Industry (slides 74-108), the structure becomes a sector tour rather than a mutually exclusive argument tree.", "closingCritique": "The last non-appendix sequence gives predictions on slide 152 and a feedback ask on slide 154, so there is some forward motion. It lacks a synthesized recommendation or next-step agenda, and the disclosure/closing-cover slides 155-156 dull the ending.", "openingCritique": "The first 5 slides establish scope, author credibility, and vocabulary, but they do not state the answer or stakes; slide 2 is an agenda and slides 4-5 are definitions. There is no real hook in the first 3 slides, so the reader must wait until section 1 evidence to infer the point.", "extractionSeconds": 85.34, "narrativeCritique": "The deck has a recognizable setup -> analysis -> resolution shape: front matter/definitions (slides 1-5), then research/talent/industry/politics evidence (slides 6-150), then predictions and a close (slides 151-154). The missing act is a crisp tension/thesis slide that converts the inventory into one governing argument before the long analysis.", "titleQualityScore": 68, "titleQualityCritique": "Title quality is mixed: many pages have real declarative titles, such as slide 18, 'Processor clock frequencies are not getting faster...' and slide 24 on NVIDIA's datacenter run-rate. But many others remain topic labels or repeated buckets, such as 'Definitions' on slides 4-5, 'AI hardware' across slides 15-29, and several sector pages in slides 75-107." }, "activistThesis": null, "pitchdeck": { "metadata": null, "profile": null }, "slides": [ { "page": 1, "type": "setup", "function": "front_matter", "imagePath": "https://imgproxy.kitesheet.com/05WXQVeEkLhKwfg9OwdFQJxnnPKtluJCfw4RK9kbQpU/rs:fit:1200:1200:0/q:82/f:webp/czM6Ly9raXRlc2hlZXQvY29ycHVzL3NsaWRlcy80NmY2NmM0OWZkMTU5MDQ4L3AwMDEuanBn", "rawType": "cover", "block": null, "metadata": { "slideType": "cover", "slideTypeCanon": { "name": "Cover", "slug": "cover", "status": "active", "canonId": "019de52d-05f2-729d-9f93-e0504f0a2410", "version": 1, "description": null }, "function": "front_matter", "functionCanon": { "name": "Front matter", "slug": "front_matter", "status": "active", "canonId": "019de52d-133d-73f0-9e26-201eda96b098", "version": 1, "description": null }, "density": "balanced", "densityScore": 26, "componentCount": 6, "textChars": 68, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:54.878+00", "seconds": 5.657, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/1", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-1", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.05, "w": 0.15, "x": 0.05, "y": 0.9 }, "kind": "paragraph", "text": "Nathan Benaich", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.878+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0ceca72d-7c38-4b1d-888e-2edcc0332da7", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.1, "x": 0.85, "y": 0.05 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.878+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "24a16cac-d30d-4c75-b5c6-986f8d046e1c", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.3, "x": 0.35, "y": 0.45 }, "kind": "paragraph", "text": "June 29, 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.878+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "253f7cd9-4451-4623-ab3b-5581190614fc", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.1, "x": 0.85, "y": 0.9 }, "kind": "paragraph", "text": "Ian Hogarth", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.878+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a66fde18-e8dc-4054-9a4c-2795e8ff8ceb", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.1, "x": 0.05, "y": 0.05 }, "kind": "paragraph", "text": "stateof.ai", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.878+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dac74cd2-e954-4414-80a7-b60e102a3893", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.3, "x": 0.35, "y": 0.35 }, "kind": "title", "text": "State of AI", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.878+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ceae1dfb-0d13-4b7d-9ebe-82064bdaa639", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [], "loops": [], "thumbSrc": "https://imgproxy.kitesheet.com/ODA9jEPm7dUorhSCE_3x0FAfIOmv3f2BNUR6VjOf_wU/rs:fit:480:480:0/q:78/f:webp/czM6Ly9raXRlc2hlZXQvY29ycHVzL3NsaWRlcy80NmY2NmM0OWZkMTU5MDQ4L3AwMDEuanBn", "imagePathAlt": "https://imgproxy.kitesheet.com/s5yhCU-N-DIDMdYfTuPZEzlsGX70yoVOwoA8KKMyYDY/rs:fit:1200:1200:0/q:82/f:webp/czM6Ly9raXRlc2hlZXQvY29ycHVzL3NsaWRlcy80NmY2NmM0OWZkMTU5MDQ4L3AwMDEucG5n", "thumbSrcAlt": "https://imgproxy.kitesheet.com/myLNLtGaWHmLOxDApsBpyZn78ILQtJhgpFx6djVcKiA/rs:fit:480:480:0/q:78/f:webp/czM6Ly9raXRlc2hlZXQvY29ycHVzL3NsaWRlcy80NmY2NmM0OWZkMTU5MDQ4L3AwMDEucG5n" }, { "page": 2, "type": "setup", "title": "The slide serves as a front-matter/introductory page for a consulting-style report.", "function": "summarize", "rawType": "executive_summary", "block": null, "metadata": { "slideType": "executive_summary", "slideTypeCanon": { "name": "Executive summary", "slug": "executive_summary", "status": "active", "canonId": "019de52d-06be-73ff-81b3-f99fe7f0025e", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "dense", "densityScore": 60, "componentCount": 6, "textChars": 1375, "nDataPoints": 0, "notes": "The slide serves as a front-matter/introductory page for a consulting-style report.", "elementsJson": [ "headline_text", "paragraph", "bullet_list" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:51.817+00", "seconds": 2.538, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/2", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-2", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "We consider the following key dimensions in our report: Research: Technology breakthroughs and their capabilities. Talent: Supply, demand and concentration of talent working in the field. Industry: Large platforms, financings and areas of application for AI-driven innovation today and tomorrow. Politics: Public opinion of AI, economic implications and the emerging geopolitics of AI.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-af24fb1e56eb", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.9, "x": 0.05, "y": 0.55 }, "kind": "list", "text": "Research: Technology breakthroughs and their capabilities. Talent: Supply, demand and concentration of talent working in the field. Industry: Large platforms, financings and areas of application for AI-driven innovation today and tomorrow. Politics: Public opinion of AI, economic implications and the emerging geopolitics of AI.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:51.817+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ecca49cd-77e1-437b-bb84-966c7c618b8d", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.7, "x": 0.05, "y": 0.85 }, "kind": "paragraph", "text": "Collaboratively produced in East London, UK by: Ian Hogarth @soundboy Nathan Benaich @nathanbenaich", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:51.817+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "34e049c4-6490-4913-ab8c-f1feb1105c39", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.35 }, "kind": "paragraph", "text": "In this report, we set out to capture a snapshot of the exponential progress in AI with a focus on developments in the past 12 months. Consider this report as a compilation of the most interesting things we’ve seen that seeks to trigger informed conversation about the state of AI and its implication for the future.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:51.817+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6629496a-57d7-47ff-8f2a-e0c9f677ab66", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.13 }, "kind": "paragraph", "text": "Artificial intelligence (AI) is a multidisciplinary field of science whose goal is to create intelligent machines. We believe that AI will be a force multiplier on technological progress in our increasingly digital, data-driven world.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:51.817+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b13daceb-44d2-4df0-a674-cd7d94cc56e6", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.05, "y": 0.03 }, "kind": "title", "text": "Introduction", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:51.817+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b35e6e08-724b-45a8-82d6-73be27e0afb8", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Chunking", "slug": "chunking", "status": "active", "bestFor": null, "canonId": "019dd956-abab-72ce-b64c-2924cc83cf39", "version": 1, "bodyDocId": null, "description": "Information grouped into chunks is remembered better - structure content in digestible units", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-2d7f3a6f9208", "evidence": "Six numbered sections chunk a 156-page report.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 2, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "9108ca24-faba-40ed-b640-1cb49b73e962", "evidence": "list/bullet: Research: Technology breakthroughs and their capabilities.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-17 03:20:34.849092+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 2, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Visual Hierarchy", "slug": "visual-hierarchy", "status": "active", "bestFor": null, "canonId": "019dd956-b2ca-706b-937f-e52441086fd7", "version": 1, "bodyDocId": null, "description": "Order of perception through Size > Color > Position > Weight > Space", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-2a6cbc518380", "evidence": "Section navigation pipe-bar header repeats across deck.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 2, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [], "loops": [], "locked": true }, { "page": 3, "type": "setup", "function": "other", "rawType": "team_bio", "block": null, "metadata": { "slideType": "team_bio", "slideTypeCanon": { "name": "Team bio", "slug": "team_bio", "status": "active", "canonId": "019de52d-0a7d-7394-ae5b-037478e71f1e", "version": 1, "description": null }, "function": "other", "functionCanon": { "name": "Other (fallback)", "slug": "other", "status": "active", "canonId": "019de52d-15e1-75bc-99dc-07732380e2ce", "version": 1, "description": null }, "density": "balanced", "densityScore": 49, "componentCount": 5, "textChars": 782, "nDataPoints": 2, "notes": null, "elementsJson": [ "headshot", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:52.022+00", "seconds": 2.598, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/3", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-3", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.3, "w": 0.2, "x": 0.22, "y": 0.21 }, "kind": "image", "text": "Nathan Benaich photo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Headshot", "slug": "headshot", "status": "active", "canonId": "019de52c-fd5f-765c-9282-38b54f61bbbf", "version": 1, "description": null }, "framework": null }, "subkind": "headshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.022+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "24affc66-8725-4081-a059-f4715d3f3223", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.2, "x": 0.61, "y": 0.21 }, "kind": "image", "text": "Ian Hogarth photo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Headshot", "slug": "headshot", "status": "active", "canonId": "019de52c-fd5f-765c-9282-38b54f61bbbf", "version": 1, "description": null }, "framework": null }, "subkind": "headshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.022+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "944cda0e-aa30-472a-8f61-9f8ee40dd051", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.35, "x": 0.13, "y": 0.6 }, "kind": "paragraph", "text": "Nathan Benaich: Nathan studied biology at Williams College and earned a PhD from Cambridge in computational and experimental cancer biology. He is an investor in machine learning-driven technology companies with his new firm, Air Street Capital, and as a Venture Partner at Point Nine Capital. He founded the RAAIS community and Foundation to advance progress in AI.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.022+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0dc69bc4-8a17-4f76-ab31-8825d86d30ac", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.35, "x": 0.52, "y": 0.6 }, "kind": "paragraph", "text": "Ian Hogarth: Ian studied engineering at Cambridge, specialising in machine learning. His Masters project was a computer vision system to classify breast cancer biopsy images. He was co-founder and CEO of Songkick, the concert service used by 17 million music fans every month. He is an angel investor in over 30 startups with a focus on applied machine learning.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.022+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9d35b5db-eac2-4c17-80fa-e1333e7ec4d4", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.2, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "About the authors", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.022+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e94423b7-f7bf-403f-a462-9ae7de6a0832", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [], "loops": [], "locked": true }, { "page": 4, "type": "appendix", "title": "Part of a larger report series, likely an introductory slide.", "function": "establish_context", "rawType": "other", "block": null, "metadata": { "slideType": "other", "slideTypeCanon": { "name": "Other (fallback)", "slug": "other", "status": "active", "canonId": "019de52d-0af5-71ed-83de-d51ae245fc43", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "balanced", "densityScore": 41, "componentCount": 3, "textChars": 1408, "nDataPoints": 0, "notes": "Part of a larger report series, likely an introductory slide.", "elementsJson": [ "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:52.65+00", "seconds": 3.19, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/4", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-4", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.7, "w": 0.9, "x": 0.05, "y": 0.2 }, "kind": "list", "text": "Artificial Intelligence (AI): A broad discipline with the goal of creating intelligent machines, as opposed to the natural intelligence that is demonstrated by humans and animals. It has become a somewhat catch all term that nonetheless captures the long term ambition of the field to build machines that emulate and then exceed the full range of human cognition.\n\nMachine learning (ML): A subset of AI that often uses statistical techniques to give machines the ability to \"learn\" from data without being explicitly given the instructions for how to do so. This process is known as \"training\" a \"model\" using a learning \"algorithm\" that progressively improves model performance on a specific task.\n\nReinforcement learning (RL): An area of ML that has received particular attention from the research community over the past decade. It is concerned with software agents that learn goal-oriented behavior by trial and error in an environment that provides rewards or penalties in response to the agent’s actions towards achieving that goal.\n\nDeep learning (DL): An area of ML that attempts to mimic the activity in layers of neurons in the brain to learn how to recognise complex patterns in data. The \"deep\" in deep learning refers to the large number of layers of neurons in contemporary ML models that help to learn rich representations of data to achieve better performance gains.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.65+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7d79c9ff-ffa2-445b-a929-411b9cc836d3", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.85, "y": 0.94 }, "kind": "source-note", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.65+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e58a5638-e7e1-412c-86f1-c0499ec2cef6", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Definitions", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.65+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "94e9c997-bdcc-4d07-a4dd-e38f49d17aae", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-31ff604ffdc2", "evidence": "Definitions slide grounds abstract terms before use.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 4, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "be89dcce-c44e-4d1f-aee0-f8d80895b9d2", "evidence": "list/bullet: Artificial Intelligence (AI): A broad discipline...", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-17 03:20:34.945311+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 4, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 } ], "loops": [], "locked": true }, { "page": 5, "type": "appendix", "title": "The slide serves as a foundational reference for technical terms.", "function": "establish_context", "rawType": "other", "block": null, "metadata": { "slideType": "other", "slideTypeCanon": { "name": "Other (fallback)", "slug": "other", "status": "active", "canonId": "019de52d-0af5-71ed-83de-d51ae245fc43", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "balanced", "densityScore": 33, "componentCount": 2, "textChars": 911, "nDataPoints": 0, "notes": "The slide serves as a foundational reference for technical terms.", "elementsJson": [ "headline_text", "bullet_list" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:54.553+00", "seconds": 5.072, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/5", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-5", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.65, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "list", "text": "Algorithm: An unambiguous specification of how to solve a particular problem.\nModel: Once a ML algorithm has been trained on data, the output of the process is known as the model. This can then be used to make predictions.\nSupervised learning: This is the most common kind of (commercial) ML algorithm today where the system is presented with labelled examples to explicitly learn from.\nUnsupervised learning: In contrast to supervised learning, the ML algorithm has to infer the inherent structure of the data that is not annotated with labels.\nTransfer learning: This is an area of research in ML that focuses on storing knowledge gained in one problem and applying it to a different or related problem, thereby reducing the need for additional training data and compute.\nGood old fashioned AI: A name given to an early symbolic AI paradigm that fell out of favour amongst researchers in the 1990s.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.553+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "722fcc00-8968-4757-a156-991f8b7b9d82", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Definitions", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.553+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2008ca7f-b6cc-42d3-aacc-d0221aef6ed8", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 } ], "loops": [], "locked": true }, { "page": 6, "type": "setup", "function": "transition", "rawType": "section_divider", "block": null, "metadata": { "slideType": "section_divider", "slideTypeCanon": { "name": "Section divider", "slug": "section_divider", "status": "active", "canonId": "019de52d-066a-7402-9554-6ae504999041", "version": 1, "description": null }, "function": "transition", "functionCanon": { "name": "Transition", "slug": "transition", "status": "active", "canonId": "019de52d-1365-7149-92aa-66df859e75fd", "version": 1, "description": null }, "density": "sparse", "densityScore": 15, "componentCount": 4, "textChars": 152, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:56.855+00", "seconds": 7.248, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/6", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-6", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.05, "w": 0.1, "x": 0.88, "y": 0.04 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.855+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4e1d4472-8eb7-41da-8d24-3e9e62e0acc7", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.12, "x": 0.85, "y": 0.92 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.855+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c6d0496f-7a94-47a2-a0f4-81a8abe6ba26", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.75, "x": 0.02, "y": 0.04 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.855+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e4091021-620f-415f-b0e9-e41a76f28561", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.7, "x": 0.15, "y": 0.45 }, "kind": "title", "text": "Section 1: Research and technical breakthroughs", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.855+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c656a387-edbc-4a29-8049-7077e6957978", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "71d87fed-6c1f-4dd3-98b0-17b648c98554", "evidence": "Section 1: Research and technical breakthroughs", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:30.887403+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 6, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 9, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 6, "name": "Surface Observation", "beatId": "019dd95a-0682-776c-8e34-b9affb23a9a4", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Surface Observation", "slug": "onion-surface-observation", "status": "active", "canonId": "019dd9b8-0b39-73c0-b415-6c4cf34cc343", "version": 1, "description": null }, "evidence": "Transfer learning headline result on skin cancer.", "position": 1, "isPrimaryArc": false, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 7, "type": "appendix", "title": "The slide uses a process flow diagram to illustrate the concept of transfer learning.", "function": "present_framework", "rawType": "other", "block": null, "metadata": { "slideType": "other", "slideTypeCanon": { "name": "Other (fallback)", "slug": "other", "status": "active", "canonId": "019de52d-0af5-71ed-83de-d51ae245fc43", "version": 1, "description": null }, "function": "present_framework", "functionCanon": { "name": "Present framework", "slug": "present_framework", "status": "active", "canonId": "019de52d-1403-7667-9100-d2ac6709f750", "version": 1, "description": null }, "density": "balanced", "densityScore": 37, "componentCount": 4, "textChars": 736, "nDataPoints": 0, "notes": "The slide uses a process flow diagram to illustrate the concept of transfer learning.", "elementsJson": [ "headline_text", "paragraph", "process_diagram" ], "confidence": 0.9, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:51.787+00", "seconds": 2.147, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/7", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-7", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Transfer learning posits that knowledge acquired by a trained machine learning model can be re-applied (or 'transferred') during the training process for a new task.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:07+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80b-a8b8722cb215", "parentComponentId": null }, { "bbox": { "h": 0.25, "w": 0.58, "x": 0.21, "y": 0.58 }, "kind": "diagram", "text": "Process flow from random neural network to trained neural network via ImageNet training and fine-tuning with 'Your data'.", "attrs": null, "canon": { "kind": { "name": "Diagram", "slug": "diagram", "status": "active", "canonId": "019de52c-f9f0-734a-81de-15382ff28c65", "version": 1, "description": "Schematic or relational diagram." }, "tool": null, "subkind": { "name": "Process flow", "slug": "process", "status": "active", "canonId": "019de52d-039c-718f-ad3a-efe9c4b3c838", "version": 1, "description": null }, "framework": null }, "subkind": "process", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:51.787+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "064c4f46-b5ae-4d8b-95a9-4e0797d60ab1", "parentComponentId": null }, { "bbox": { "h": 0.25, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "What is transfer learning and how does it relate to machine learning? Machine learning models are trained to solve a task by learning from examples. However, to solve a new and different task, a trained model needs to be retrained with new data specific to that task. Transfer learning posits that knowledge acquired by a trained machine learning model can be re-applied (or 'transferred') during the training process for a new task.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:51.787+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d7175e3b-f751-4576-b7ad-edbed86930b5", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.2, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Transfer Learning", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:51.787+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c141dd99-6833-460f-8684-e7c59c342037", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "loop", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-3bc9311dc5a5", "evidence": "First slide of context->conflict->insight aha-moment loop.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 7, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-351f5df0f588", "evidence": "Callout 'transferred' frames knowledge as portable cargo.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 7, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 9, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 6, "name": "Surface Observation", "beatId": "019dd95a-0682-776c-8e34-b9affb23a9a4", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Surface Observation", "slug": "onion-surface-observation", "status": "active", "canonId": "019dd9b8-0b39-73c0-b415-6c4cf34cc343", "version": 1, "description": null }, "evidence": "Transfer learning headline result on skin cancer.", "position": 1, "isPrimaryArc": false, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 9, "from": 7, "loop": { "name": "03_aha_moment", "slug": "03-aha-moment", "status": "active", "bestFor": "Data-heavy sections, research findings, analytical arguments", "canonId": "019dd956-66f4-7668-ab25-cec9c996c0bc", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Problem/Question -> What the Data Shows -> The Insight", "description": "Establish a tension, present data, then deliver the insight that changes everything", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-46cbef5e2b62", "evidence": "Defines transfer learning then reveals model beats 21 dermatologists with ROC charts.", "position": 1, "objective": "Show transfer learning enables superhuman skin-cancer detection", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 8, "type": "appendix", "title": "The slide uses a technical diagram to illustrate the concept of transfer learning in AI.", "function": "present_framework", "rawType": "other", "block": null, "metadata": { "slideType": "other", "slideTypeCanon": { "name": "Other (fallback)", "slug": "other", "status": "active", "canonId": "019de52d-0af5-71ed-83de-d51ae245fc43", "version": 1, "description": null }, "function": "present_framework", "functionCanon": { "name": "Present framework", "slug": "present_framework", "status": "active", "canonId": "019de52d-1403-7667-9100-d2ac6709f750", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 6, "textChars": 643, "nDataPoints": 2, "notes": "The slide uses a technical diagram to illustrate the concept of transfer learning in AI.", "elementsJson": [ "headline_text", "paragraph", "process_diagram" ], "confidence": 0.9, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:52.642+00", "seconds": 2.854, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/8", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-8", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Re-using previously acquired knowledge reduces the amount of data a model needs in order to learn a new task.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:07+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80b-bc2e7e967294", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.7, "x": 0.15, "y": 0.55 }, "kind": "diagram", "text": "Diagram showing pre-trained InceptionV3 network feeding into a skin cancer-specific classification layer.", "attrs": null, "canon": { "kind": { "name": "Diagram", "slug": "diagram", "status": "active", "canonId": "019de52c-f9f0-734a-81de-15382ff28c65", "version": 1, "description": "Schematic or relational diagram." }, "tool": null, "subkind": { "name": "Process flow", "slug": "process", "status": "active", "canonId": "019de52d-039c-718f-ad3a-efe9c4b3c838", "version": 1, "description": null }, "framework": null }, "subkind": "process", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.642+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7ed9b0b3-429f-47a3-8afd-8bdaaf005d5e", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.95, "x": 0.02, "y": 0.48 }, "kind": "list", "text": "Example: Repurposing Google's InceptionV3 image recognition network for skin cancer detection", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.642+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6c28e60c-a14b-4dc3-a590-adcd1ae90223", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "Why does transfer learning matter? Re-using previously acquired knowledge reduces the amount of data a model needs in order to learn a new task.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.642+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b2cb098b-bff0-411e-b219-4da451c366d8", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.95, "x": 0.02, "y": 0.35 }, "kind": "paragraph", "text": "A model pre-trained on many different problems will internalise an increasingly rich understanding of the world and is therefore considered a key step towards generalising AI.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.642+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e5681166-6919-41fa-b780-c4d9fa0aa794", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.2, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Transfer Learning", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.642+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "00451fb5-565b-452f-945d-6b29a104c797", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "1c4e8e44-b887-4a8d-8a0b-d51f385162fc", "evidence": "list/bullet: Why does transfer learning matter? Re-using previously acquired knowledge reduces the amount of data a model needs in order to learn a new task.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:30.988165+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 8, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 9, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 6, "name": "Surface Observation", "beatId": "019dd95a-0682-776c-8e34-b9affb23a9a4", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Surface Observation", "slug": "onion-surface-observation", "status": "active", "canonId": "019dd9b8-0b39-73c0-b415-6c4cf34cc343", "version": 1, "description": null }, "evidence": "Transfer learning headline result on skin cancer.", "position": 1, "isPrimaryArc": false, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 9, "from": 7, "loop": { "name": "03_aha_moment", "slug": "03-aha-moment", "status": "active", "bestFor": "Data-heavy sections, research findings, analytical arguments", "canonId": "019dd956-66f4-7668-ab25-cec9c996c0bc", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Problem/Question -> What the Data Shows -> The Insight", "description": "Establish a tension, present data, then deliver the insight that changes everything", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-46cbef5e2b62", "evidence": "Defines transfer learning then reveals model beats 21 dermatologists with ROC charts.", "position": 1, "objective": "Show transfer learning enables superhuman skin-cancer detection", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 9, "type": "evidence", "title": "The slide uses ROC curves to compare algorithm performance against human experts.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 70, "componentCount": 7, "textChars": 963, "nDataPoints": 6, "notes": "The slide uses ROC curves to compare algorithm performance against human experts.", "elementsJson": [ "headline_text", "bullet_list", "paragraph", "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:52.559+00", "seconds": 2.513, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/9", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-9", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "In the charts on the right, you'll see that the majority of red points (dermatologist) reside below the blue curve (sensitivity-specificity for the model).", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80b-d4a76ffd96e2", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.46, "x": 0.52, "y": 0.4 }, "kind": "chart", "text": "ROC curves comparing algorithm performance (blue line) to dermatologists (red dots).", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.559+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ce281eb6-ee7b-4384-9658-b7de22df2a34", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.48, "x": 0.02, "y": 0.41 }, "kind": "list", "text": "The model outperforms 21 Stanford dermatologists. Dermatologists: biopsy or treat the lesion? Model: what probability is the lesion dangerous?", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.559+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4e0153df-0243-4f71-86ae-34fc4e43d70e", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.48, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "Transfer learning enabled automatic state-of-the-art detection of dangerous skin lesions on human patients. The Google InceptionV3 network was first trained on ImageNet and then re-trained with 129,450 clinical images of 2,032 different skin diseases.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.559+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f095ff2e-8b72-4e4d-9063-55f88a62f698", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "AUC: 0.96", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80b-d9c338240d0a", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.48, "x": 0.02, "y": 0.61 }, "kind": "paragraph", "text": "In the charts on the right, you'll see that the majority of red points (dermatologist) reside below the blue curve (sensitivity-specificity for the model). This means the model achieves superior performance compared to dermatologists.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.559+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c7bfeb2e-e3c1-4cad-82bc-b4db76e8c129", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.96, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Transfer Learning: From predicting everyday objects on ImageNet to detecting skin cancer", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.559+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3dcc0f28-7651-48d4-962b-288d80254f2a", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-3efd50fe452d", "evidence": "Title states 'From predicting everyday objects on ImageNet to detecting skin cancer'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 9, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Annotation", "slug": "annotation", "status": "active", "bestFor": null, "canonId": "019dd956-b5b9-7668-bf0e-3af99823ec5c", "version": 1, "bodyDocId": null, "description": "Adding explanatory labels to highlight key data points", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-45d2416a6b79", "evidence": "Callout points to red dermatologist points below blue model curve.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 9, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Sinatra Test", "slug": "sinatra-test", "status": "active", "bestFor": null, "canonId": "019dd956-be4a-775f-8038-d4e7e58dfc5a", "version": 1, "bodyDocId": null, "description": "If you can make it there, you can make it anywhere - using flagship example", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-42b8fabe2e0f", "evidence": "Single flagship example (skin cancer) carries the transfer-learning argument.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 9, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 9, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 6, "name": "Surface Observation", "beatId": "019dd95a-0682-776c-8e34-b9affb23a9a4", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Surface Observation", "slug": "onion-surface-observation", "status": "active", "canonId": "019dd9b8-0b39-73c0-b415-6c4cf34cc343", "version": 1, "description": null }, "evidence": "Transfer learning headline result on skin cancer.", "position": 1, "isPrimaryArc": false, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 9, "from": 7, "loop": { "name": "03_aha_moment", "slug": "03-aha-moment", "status": "active", "bestFor": "Data-heavy sections, research findings, analytical arguments", "canonId": "019dd956-66f4-7668-ab25-cec9c996c0bc", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Problem/Question -> What the Data Shows -> The Insight", "description": "Establish a tension, present data, then deliver the insight that changes everything", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-46cbef5e2b62", "evidence": "Defines transfer learning then reveals model beats 21 dermatologists with ROC charts.", "position": 1, "objective": "Show transfer learning enables superhuman skin-cancer detection", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 10, "type": "analysis", "title": "The slide uses two bar charts to demonstrate the performance advantage of GPUs in AI training.", "function": "establish_context", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "dense", "densityScore": 67, "componentCount": 7, "textChars": 796, "nDataPoints": 6, "notes": "The slide uses two bar charts to demonstrate the performance advantage of GPUs in AI training.", "elementsJson": [ "headline_text", "paragraph", "bar_chart_vertical", "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:52.899+00", "seconds": 2.468, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/10", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-10", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Graphics processing units (GPUs) are today's workhorse chip for AI models largely because they offer immense computational parallelism over central processing units (CPUs).", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-93fd0650af3d", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.45, "x": 0.02, "y": 0.52 }, "kind": "chart", "text": "AI models run much faster on GPUs", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.899+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "82ee007f-5d4e-4c03-9b37-39ef193f3385", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.45, "x": 0.52, "y": 0.52 }, "kind": "chart", "text": "New GPU supercomputers eclipse older chips", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.899+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "960abe1f-2985-4b82-982d-42feffefc212", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Training time: 18 Mins", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-94c3b4d5a672", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.95, "x": 0.02, "y": 0.27 }, "kind": "paragraph", "text": "Semiconductor (or 'chip') performance is a key driver behind progress in AI research and applications. This is because AI models often require huge amounts of training data to properly learn a task (e.g. image recognition). Graphics processing units (GPUs) are today's workhorse chip for AI models largely because they offer immense computational parallelism over central processing units (CPUs). This means faster training and model iteration.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.899+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d00a760a-030f-4743-a08a-ab2cdc42576c", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "The role of semiconductors in driving AI performance", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.899+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8e1cd0e4-c76f-4d6d-b2b4-c6db3978ebfd", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.35, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI hardware as the new frontier", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.899+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f2e19d57-e826-42fc-9c00-d9f8d8b11ae5", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-4c5539c9e63f", "evidence": "First slide of pattern_hunter loop assembling hardware evidence.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 10, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-482fa470ac47", "evidence": "Title 'AI hardware as the new frontier' states position not topic.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 10, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "dd662028-4ab2-4381-a97c-28b981f5ca63", "evidence": "title/action-title: The role of semiconductors in driving AI performance", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:31.090485+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 10, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 11, "type": "analysis", "title": "The slide uses two charts to demonstrate the scaling efficiency of GPU clusters in AI model training.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 53, "componentCount": 7, "textChars": 350, "nDataPoints": 15, "notes": "The slide uses two charts to demonstrate the scaling efficiency of GPU clusters in AI model training.", "elementsJson": [ "headline_text", "bar_chart_vertical", "line_chart", "callout_box" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:52.969+00", "seconds": 2.398, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/11", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-11", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "The hardware war: More GPUs allows for faster training, as well as bigger (more powerful) models.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-75bd65449c4b", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.45, "x": 0.05, "y": 0.3 }, "kind": "chart", "text": "Training speedup over one GPU (x-fold)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.969+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0db9fe69-26af-40ee-9239-a8268898d5fe", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.45, "x": 0.52, "y": 0.3 }, "kind": "chart", "text": "32 GPUs = same accuracy, 25x faster", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.969+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f4bc794e-6cd2-4850-95dd-717e81666b05", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Training speedup: 25x", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-7b579b54ab15", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.95, "x": 0.02, "y": 0.92 }, "kind": "source-note", "text": "Baidu Google AI stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.969+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "829dc2e0-20e4-4dc6-8dd8-fb7d8bf16990", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.85, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "The hardware war: More GPUs allows for faster training, as well as bigger (more powerful) models.", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.969+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6d6c5094-80de-43fc-9ce1-50294ea6f4d5", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.35, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI hardware as the new frontier", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:52.969+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0c75a12f-3149-440c-aa5d-17b90c231661", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-5078758dbbcd", "evidence": "Headline 'The hardware war: More GPUs allows for faster training'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 11, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "6a394c96-7517-489c-9aa0-abb752d9c208", "evidence": "title/action-title: The hardware war: More GPUs allows for faster training, as well as bigger (more powerful) models.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:31.156759+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 11, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 12, "type": "analysis", "title": "The slide uses three distinct charts to demonstrate scaling laws: a conceptual framework, a loss-reduction chart, and a parameter-growth chart.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 62, "componentCount": 8, "textChars": 451, "nDataPoints": 3, "notes": "The slide uses three distinct charts to demonstrate scaling laws: a conceptual framework, a loss-reduction chart, and a parameter-growth chart.", "elementsJson": [ "headline_text", "action_title", "line_chart", "line_chart", "line_chart", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:53.446+00", "seconds": 2.774, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/12", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-12", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "AI model performance scales with dataset size and the # of model parameters, thus necessitating more compute", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-edfaf9e6e1bb", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.3, "x": 0.68, "y": 0.35 }, "kind": "chart", "text": "More data = bigger model", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.446+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0ab0ca58-3d86-4cea-8d30-fbcac8707d0f", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.3, "x": 0.04, "y": 0.35 }, "kind": "chart", "text": "A framework for scaling AI models", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.446+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2d35b3ec-85cf-4c5c-9a5a-8e427c5a7408", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.3, "x": 0.36, "y": 0.35 }, "kind": "chart", "text": "More data = fewer mistakes", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.446+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "74fe9cbd-5e1a-46f3-b9f2-b3f2ece5d928", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Generalization Error", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-f0ec91b70401", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.8, "x": 0.02, "y": 0.02 }, "kind": "other", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Other", "slug": "other", "status": "active", "canonId": "019de52c-fa90-743c-8bf6-20d1fbe82b6d", "version": 1, "description": "Fallback when no canonical kind applies." }, "tool": null, "subkind": { "name": "Unclassified", "slug": "unclassified", "status": "active", "canonId": "019de52d-05ca-77db-81fb-db7d4234bbc0", "version": 1, "description": null }, "framework": null }, "subkind": "unclassified", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.446+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5eb45f14-72d8-42ae-8686-d1d058218512", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.9, "x": 0.04, "y": 0.22 }, "kind": "paragraph", "text": "AI model performance scales with dataset size and the # of model parameters, thus necessitating more compute", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.446+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "51aa4da0-f75c-48dc-813d-af5cecc77687", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI hardware is especially helpful for deep learning", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.446+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "31360058-3b30-489c-8a4c-52c60b0d9e4b", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-574b318fb7e0", "evidence": "Title 'AI hardware is especially helpful for deep learning'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 12, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [ { "matchId": "edf29fe1-20fd-4386-8312-0bcb6de22219", "evidence": "The charts demonstrate the empirical relationship between data size, loss, and parameter count.", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:43:53.446+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 12, "frameworkName": "scaling-laws" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 13, "type": "analysis", "title": "The slide illustrates the Information Bottleneck theory, showing the evolution of layers (L1-L5) through phases A-E.", "function": "present_framework", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "present_framework", "functionCanon": { "name": "Present framework", "slug": "present_framework", "status": "active", "canonId": "019de52d-1403-7667-9100-d2ac6709f750", "version": 1, "description": null }, "density": "dense", "densityScore": 50, "componentCount": 5, "textChars": 1120, "nDataPoints": 0, "notes": "The slide illustrates the Information Bottleneck theory, showing the evolution of layers (L1-L5) through phases A-E.", "elementsJson": [ "headline_text", "line_chart", "numbered_list" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:53.893+00", "seconds": 3.215, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/13", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-13", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "First memorise the data, then forget what doesn't help the model make predictions.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-28ca6c40c530", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.45, "x": 0.05, "y": 0.32 }, "kind": "chart", "text": "Inside Deep Learning: New experiments reveal how deep neural networks evolve as they learn.", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "155f700b-5a59-4b6b-b4c2-4f81baaba9db", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.45, "x": 0.52, "y": 0.32 }, "kind": "list", "text": "A INITIAL STATE: Neurons in Layer 1 encode everything about the input data, including all information about its label. Neurons in the highest layers are in a nearly random state bearing little to no relationship to the data or its label.\nB FITTING PHASE: As deep learning begins, neurons in higher layers gain information about the input and get better at fitting labels to it.\nC PHASE CHANGE: The layers suddenly shift gears and start to \"forget\" information about the input.\nD COMPRESSION PHASE: Higher layers compress their representation of the input data, keeping what is most relevant to the output label. They get better at predicting the label.\nE FINAL STATE: The last layer achieves an optimal balance of accuracy and compression, retaining only what is needed to predict the label.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9bc7a207-89ce-4cf6-8bc1-eb225b836e46", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.9, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "The Information Theory: First memorise the data, then forget what doesn't help the model make predictions", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b59ffc7d-1a58-44e3-bcd6-e7273d63488a", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI hardware is especially helpful for deep learning", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e3108e8a-7cf2-4a47-9058-ced032b9019a", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [ { "matchId": "7a8a46ae-f108-4b15-aebe-a15ca8d09a13", "evidence": "The slide describes the process of memorizing data and then compressing it to retain only relevant information for prediction.", "framework": null, "confidence": 1, "extraction": { "at": "2026-05-02 17:43:53.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 13, "frameworkName": "Information Bottleneck Theory" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 14, "type": "analysis", "title": "The chart shows a log-scale progression of compute usage for various AI models from 2013 to 2018. The right side depicts a flywheel effect.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 57, "componentCount": 7, "textChars": 446, "nDataPoints": 12, "notes": "The chart shows a log-scale progression of compute usage for various AI models from 2013 to 2018. The right side depicts a flywheel effect.", "elementsJson": [ "headline_text", "scatter_plot", "process_diagram", "callout_box" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:53.541+00", "seconds": 2.752, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/14", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-14", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "More compute means new solutions to previously intractable problems, e.g. machines learning to play Go", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80b-f5fa9405c011", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.45, "x": 0.05, "y": 0.3 }, "kind": "chart", "text": "AlexNet to AlphaGo Zero: 300,000x more compute", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Scatter plot", "slug": "scatter", "status": "active", "canonId": "019de52d-007d-72ec-8689-c79fb78e21a2", "version": 1, "description": null }, "framework": null }, "subkind": "scatter", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.541+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "67fc1252-b86a-4877-9fc9-8daf3236e273", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.55, "y": 0.3 }, "kind": "diagram", "text": "A positive feedback loop drives AI competitiveness", "attrs": null, "canon": { "kind": { "name": "Diagram", "slug": "diagram", "status": "active", "canonId": "019de52c-f9f0-734a-81de-15382ff28c65", "version": 1, "description": "Schematic or relational diagram." }, "tool": null, "subkind": { "name": "Process flow", "slug": "process", "status": "active", "canonId": "019de52d-039c-718f-ad3a-efe9c4b3c838", "version": 1, "description": null }, "framework": null }, "subkind": "process", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.541+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "64bf81d0-9a03-45d2-bd42-d0a82d384f8f", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.55, "y": 0.3 }, "kind": "framework", "text": "A positive feedback loop drives AI competitiveness", "attrs": null, "canon": { "kind": { "name": "Framework", "slug": "framework", "status": "active", "canonId": "019de52c-fa18-77c6-b10a-ccb8277ac0fa", "version": 1, "description": "Named conceptual framework instance." }, "tool": null, "subkind": { "name": "Framework instance", "slug": "instance", "status": "active", "canonId": "019de52d-05a2-72af-be0b-7edc60708a03", "version": 1, "description": null }, "framework": null }, "subkind": "instance", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.541+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2d1d2247-7c2f-4115-ac3e-8385bc5d1447", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Petaflop/s-day (Training): 300,000x", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80b-f82ccdf9d7c0", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "More compute means new solutions to previously intractable problems, e.g. machines learning to play Go", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.541+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c0b3fe07-8777-46d1-9acb-18ad3493857c", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "AI hardware rate limits progress in today's deep learning era", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.541+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d55b7546-9d95-4b57-b39c-4ab506623734", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-58519a1fcec3", "evidence": "Title quantifies '300,000x' compute growth driving Go breakthrough.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 14, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Weber-Fechner Law", "slug": "weber-fechner-law", "status": "active", "bestFor": null, "canonId": "019dd956-cf72-75bd-b89a-48bc52f65964", "version": 1, "bodyDocId": null, "description": "Perceived change is proportional to initial stimulus - bigger changes needed to notice in larger contexts", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-5cfd5dc0d4ef", "evidence": "300,000x figure designed to register magnitude shock.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 55, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 14, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [ { "matchId": "d21e7740-bdff-4de4-bb8d-090d61722bd3", "evidence": "The diagram depicts a self-reinforcing cycle of Bigger model -> Better performance -> More money -> More user data -> Bigger model.", "framework": null, "confidence": 1, "extraction": { "at": "2026-05-02 17:43:53.541+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 14, "frameworkName": "flywheel" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 15, "type": "analysis", "title": "The slide uses growth multipliers (+15X and +5X) to highlight the rapid adoption of GPU-based AI development.", "function": "quantify_impact", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "quantify_impact", "functionCanon": { "name": "Quantify impact", "slug": "quantify_impact", "status": "active", "canonId": "019de52d-1630-720d-b57d-fa7f76bebd8d", "version": 1, "description": null }, "density": "dense", "densityScore": 50, "componentCount": 7, "textChars": 337, "nDataPoints": 4, "notes": "The slide uses growth multipliers (+15X and +5X) to highlight the rapid adoption of GPU-based AI development.", "elementsJson": [ "bar_chart_vertical", "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:53.3+00", "seconds": 2.464, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/15", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-15", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "No wonder that GPUs have grown immensely in popularity amongst developers", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-d260cdf2ad52", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.5, "y": 0.315 }, "kind": "chart", "text": "Exhibit 5: ...driving demand for software to support AI development", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.3+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "244e5166-b369-42a9-8371-1a4a51299fb8", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.05, "y": 0.315 }, "kind": "chart", "text": "Exhibit 4: The number of GPU developers has grown significantly...", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.3+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "eb26b6bc-62c4-4cf7-838a-784cf624276e", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Developer count / Downloads: +15X", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-d69f9df96538", "parentComponentId": null }, { "bbox": { "h": 0.02, "w": 0.1, "x": 0.05, "y": 0.86 }, "kind": "source-note", "text": "Source: Nvidia", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.3+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bbe9fb44-bd98-437c-a6c0-b6d11ecc3cfc", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.65, "x": 0.03, "y": 0.21 }, "kind": "title", "text": "No wonder that GPUs have grown immensely in popularity amongst developers", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.3+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "469fe6fd-f134-41ff-86f0-27b22574daa4", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:53.3+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3fd90b33-3729-4bcf-8510-0b58e52d9cbb", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-62d54ed1da68", "evidence": "Title 'No wonder GPUs have grown immensely in popularity'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 15, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-6771c90ffa06", "evidence": "Growth metric '+15X' implies trend chart.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 15, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 16, "type": "analysis", "title": "The slide illustrates that while GPUs are used for ML, a significant portion of their core area is dedicated to other tasks, yet they consume substantial power.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 41, "componentCount": 6, "textChars": 315, "nDataPoints": 3, "notes": "The slide illustrates that while GPUs are used for ML, a significant portion of their core area is dedicated to other tasks, yet they consume substantial power.", "elementsJson": [ "bar_chart_stacked", "paragraph" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:54.254+00", "seconds": 2.744, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/16", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-16", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "However, GPUs were built for graphics workloads and evolved for high performance computing and AI workloads", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-0ddd7cf5eef0", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.16, "x": 0.42, "y": 0.35 }, "kind": "chart", "text": "GPU core area and power usage breakdown", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked bar chart", "slug": "bar-stacked", "status": "active", "canonId": "019de52c-feec-74bf-9028-15b74f84f0a7", "version": 1, "description": null }, "framework": null }, "subkind": "bar-stacked", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.254+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "70c4343b-f39b-4fe7-9067-e5c6fa9894b6", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.08, "x": 0.42, "y": 0.38 }, "kind": "legend", "text": "10% ram, 55% dark, 35% fpu", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.254+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "36bc45bc-d973-4540-bd47-2347ed3123b2", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Core area allocation: 35%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-11478a313c32", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.23 }, "kind": "paragraph", "text": "However, GPUs were built for graphics workloads and evolved for high performance computing and AI workloads", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.254+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9d92e1a4-4b0b-498f-b17d-72ed01806be9", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.254+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6e8b02e2-016e-4ec7-b245-92adf9ad9c16", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 17, "type": "analysis", "title": "Data presented in two tables (Table II and Table III) comparing resource allocation for training vs inference.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 46, "componentCount": 5, "textChars": 508, "nDataPoints": 42, "notes": "Data presented in two tables (Table II and Table III) comparing resource allocation for training vs inference.", "elementsJson": [ "headline_text", "action_title", "paragraph", "comparison_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:54.726+00", "seconds": 3.205, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/17", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-17", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "At Facebook, for example, primary use case of GPUs is offline training rather than serving real-time data to users.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-4084b4a46361", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.94, "x": 0.03, "y": 0.28 }, "kind": "paragraph", "text": "While in most cases, training on GPUs tends to outperform training on CPUs, the abundance of readily-available CPU capacity in the datacenter makes it a useful and widely used platform. At Facebook, for example, primary use case of GPUs is offline training rather than serving real-time data to users.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.726+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "628f8e1c-8ef4-461d-a867-dc1f87ba4ed6", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.46, "x": 0.51, "y": 0.5 }, "kind": "table", "text": "However, online training is CPU-heavy", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.726+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cee96466-831d-435d-b49e-74fa82bb0813", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.46, "x": 0.03, "y": 0.5 }, "kind": "table", "text": "Offline training uses a mix of GPUs and CPUs", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.726+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "daf239ad-2615-4007-a394-fda9a935c61e", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.03, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.726+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c779518a-4ff9-4487-a11c-f96d026025bf", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "befa04c5-9ac5-4708-82b2-11a89f69b58f", "evidence": "The slide contains a table with data, which is a clear instantiation of the data-ink ratio tool.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:31.794421+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 17, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 18, "type": "analysis", "title": "The chart is a classic visualization of the end of Dennard scaling.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 43, "componentCount": 6, "textChars": 268, "nDataPoints": 100, "notes": "The chart is a classic visualization of the end of Dennard scaling.", "elementsJson": [ "scatter_plot", "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:55.113+00", "seconds": 3.242, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/18", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-18", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Processor clock frequencies are not getting faster and Moore’s Law can only take us so far", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-1337a38260eb", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.55, "x": 0.22, "y": 0.33 }, "kind": "chart", "text": "Clock frequency (MHz) vs Year", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Scatter plot", "slug": "scatter", "status": "active", "canonId": "019de52d-007d-72ec-8689-c79fb78e21a2", "version": 1, "description": null }, "framework": null }, "subkind": "scatter", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.113+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1d96c70e-a5ae-44c7-9833-13593899dfd1", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "clock frequency (MHz)", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-147585bb8c68", "parentComponentId": null }, { "bbox": { "h": 0.02, "w": 0.13, "x": 0.64, "y": 0.34 }, "kind": "source-note", "text": "Source: Stanford VLSI Group", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.113+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "07273489-841a-4ea9-85e8-07a2df38823a", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.75, "x": 0.02, "y": 0.23 }, "kind": "title", "text": "Processor clock frequencies are not getting faster and Moore's Law can only take us so far", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.113+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e1c1d25a-f676-44e2-ac6a-50c82c4af70c", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.113+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bc8f24bd-cecf-44ed-a011-673f69c1cc66", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "2ecef736-b4a3-4648-9206-97b137cd05a1", "evidence": "The slide contains a scatter chart, which is a clear instantiation of the chart selection guide tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:31.872341+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 18, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 19, "type": "analysis", "title": "The slide highlights the architectural shift towards memory-centric design (75% RAM) and demonstrates significant performance gains in inference speed.", "function": "present_solution", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "present_solution", "functionCanon": { "name": "Present solution", "slug": "present_solution", "status": "active", "canonId": "019de52d-15b8-71da-bd23-2b767850cbce", "version": 1, "description": null }, "density": "dense", "densityScore": 53, "componentCount": 7, "textChars": 385, "nDataPoints": 6, "notes": "The slide highlights the architectural shift towards memory-centric design (75% RAM) and demonstrates significant performance gains in inference speed.", "elementsJson": [ "headline_text", "action_title", "bar_chart_stacked", "bar_chart_horizontal", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:54.525+00", "seconds": 2.494, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/19", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-19", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "New architectures optimise for memory and compute to offer state-of-the-art performance running AI models", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-b1641e2258bb", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.55, "x": 0.42, "y": 0.35 }, "kind": "chart", "text": "DEEPBENCH LSTM RNN: INFERENCE", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.525+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cd855663-5eeb-4090-8422-f825fdc6306d", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.25, "x": 0.1, "y": 0.35 }, "kind": "chart", "text": "Core area and power distribution for IPU plan", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked bar chart", "slug": "bar-stacked", "status": "active", "canonId": "019de52c-feec-74bf-9028-15b74f84f0a7", "version": 1, "description": null }, "framework": null }, "subkind": "bar-stacked", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.525+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "217a5c45-2eca-40ae-9e7b-2390b4aa2df7", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Inferences per second: 242x", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-b7434111fe67", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.9, "x": 0.03, "y": 0.23 }, "kind": "paragraph", "text": "New architectures optimise for memory and compute to offer state-of-the-art performance running AI models", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.525+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9f98cf4b-ff05-4259-bba3-96ec8edcefae", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.4, "x": 0.5, "y": 0.85 }, "kind": "source-note", "text": "Graphcore C2 IPU accelerator PCIe card vs Nvidia P100-PCIE-12GB", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.525+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "12f6bd05-24fc-405a-b4f7-e8d34752ce32", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.03, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.525+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5fd518ad-3560-4a84-88de-2bc7e08848d2", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-689d46240d40", "evidence": "Title states '242x' performance from new architectures.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 19, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chartjunk Elimination", "slug": "chartjunk-elimination", "status": "active", "bestFor": null, "canonId": "019dd956-b49e-7453-ab01-b60f338eeda1", "version": 1, "bodyDocId": null, "description": "Remove decorative elements that don't convey data: 3D effects, shadows, unnecessary gridlines", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "6ed954bf-912b-4202-8510-c71359ca85e3", "evidence": "The slide contains a bar chart with clear and simple visuals, which is a clear instantiation of the chartjunk elimination tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:31.952625+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 19, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 20, "type": "analysis", "title": "The chart uses multipliers (81x, 4x, 3.3x, 6.1x) to emphasize the price disparity between standard CPUs and specialized AI hardware.", "function": "quantify_impact", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "quantify_impact", "functionCanon": { "name": "Quantify impact", "slug": "quantify_impact", "status": "active", "canonId": "019de52d-1630-720d-b57d-fa7f76bebd8d", "version": 1, "description": null }, "density": "balanced", "densityScore": 47, "componentCount": 6, "textChars": 422, "nDataPoints": 8, "notes": "The chart uses multipliers (81x, 4x, 3.3x, 6.1x) to emphasize the price disparity between standard CPUs and specialized AI hardware.", "elementsJson": [ "headline_text", "bar_chart_vertical", "footnote" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:54.666+00", "seconds": 2.545, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/20", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-20", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Offerings for AI command significantly higher prices", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-59a9fa130313", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.55, "x": 0.22, "y": 0.295 }, "kind": "chart", "text": "Exhibit 11: Offerings for AI command significantly higher prices. Google Compute Engine price/hour/single compute instance (i.e. per 1CPU, GPU, TPU, etc)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.666+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "03c92b92-d470-4c41-b86d-5f1165868fca", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Google Compute Engine price/hour: $6.50", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-5f8aadbd1c9c", "parentComponentId": null }, { "bbox": { "h": 0.02, "w": 0.3, "x": 0.22, "y": 0.95 }, "kind": "source-note", "text": "Source: Google, Goldman Sachs Global Investment Research.", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.666+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e8b1f61d-1c54-4eb8-845c-801af1724279", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.9, "x": 0.03, "y": 0.22 }, "kind": "title", "text": "But GPUs and novel silicon are costly to rent per hour, which means progress is limited by financial resources", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.666+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "98ea257b-5f9b-4e51-a453-aa037afb8410", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:54.666+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4530c13b-fcc9-4b82-b687-44e0e78adf84", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Loss Aversion", "slug": "loss-aversion", "status": "active", "bestFor": null, "canonId": "019dd956-9761-738f-8fd4-558a63ea744a", "version": 1, "bodyDocId": null, "description": "Losses are felt 2x more strongly than equivalent gains - frame stakes as potential loss", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "block", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-728aae2ca6b4", "evidence": "Frames cost as constraint that risks excluding players.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 55, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 20, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-6e9e06ffca00", "evidence": "Title 'GPUs and novel silicon are costly to rent... limited by financial resources'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 20, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "f01d68df-b1bf-4ecf-ac17-70e6a3b88f5e", "evidence": "The slide contains a bar chart with a clear narrative, which could be an instantiation of the data story arc tool.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:32.033938+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 20, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 21, "type": "analysis", "title": "The slide uses a line chart to show training convergence and a bar chart to show cost efficiency.", "function": "quantify_impact", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "quantify_impact", "functionCanon": { "name": "Quantify impact", "slug": "quantify_impact", "status": "active", "canonId": "019de52d-1630-720d-b57d-fa7f76bebd8d", "version": 1, "description": null }, "density": "balanced", "densityScore": 47, "componentCount": 6, "textChars": 418, "nDataPoints": 6, "notes": "The slide uses a line chart to show training convergence and a bar chart to show cost efficiency.", "elementsJson": [ "headline_text", "action_title", "line_chart", "bar_chart_grouped" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:55.215+00", "seconds": 2.911, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/21", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-21", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Whilst more costly per hour, new silicon (e.g. Google's TPUv2) allows for faster model training at lower final costs (vs. NVIDIA V100)", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-97ab733aef1e", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.4, "x": 0.55, "y": 0.35 }, "kind": "chart", "text": "Cloud cost to reach 75.7% top-1 accuracy", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Grouped bar chart", "slug": "bar-grouped", "status": "active", "canonId": "019de52c-ff3b-751e-a8ff-f21ff11097b1", "version": 1, "description": null }, "framework": null }, "subkind": "bar-grouped", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.215+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e4d77dd1-b7ac-4cd7-8b04-1af3161ef026", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.45, "x": 0.03, "y": 0.35 }, "kind": "chart", "text": "Top-1 accuracy learning rate of TPU vs GPU on ImageNet", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.215+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6885cb8f-6ffa-4114-a8c0-eeaa4bf5653e", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Cloud cost to reach 75.7% top-1 accuracy: $55", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-9baf97b37680", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.95, "x": 0.03, "y": 0.22 }, "kind": "title", "text": "Whilst more costly per hour, new silicon (e.g. Google's TPUv2) allows for faster model training at lower final costs (vs. NVIDIA V100)", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.215+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7abf4f02-5973-4e3f-bd02-b032354b7bbf", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.03, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.215+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ef690129-b64a-4d10-833e-0f1d0e82a20a", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Contrast Pairs", "slug": "contrast-pairs", "status": "active", "bestFor": null, "canonId": "019dd956-9b77-70d3-88b1-f34e26f60860", "version": 1, "bodyDocId": null, "description": "Creating tension through opposites: Before↔After, Problem↔Solution, Pain↔Relief", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "loop", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-787c2dff0c52", "evidence": "TPUv2 vs V100 explicit contrast pair.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 21, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-7742d096b03a", "evidence": "Title makes head-to-head TPUv2 vs V100 economic claim.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 21, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Waterfall chart", "slug": "waterfall-chart", "status": "active", "bestFor": null, "canonId": "e2799da3-725f-44a9-a093-570f1f388e64", "version": 1, "bodyDocId": null, "description": null, "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "4f1655bb-a7b7-4fd6-bc42-0795aac493b3", "evidence": "The slide contains a line chart and a bar chart, which could be an instantiation of the waterfall chart tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:32.113519+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 21, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 21, "from": 10, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4a0086ef4670", "evidence": "Twelve contiguous slides on GPU growth, Moore's Law, new architectures, hourly cost, ending in TPUv2 vs V100 cost case.", "position": 2, "objective": "Stack evidence that AI hardware is the binding constraint", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 22, "type": "analysis", "title": "The slide features a photograph of a TPUv3 pod.", "function": "establish_context", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "balanced", "densityScore": 32, "componentCount": 5, "textChars": 305, "nDataPoints": 2, "notes": "The slide features a photograph of a TPUv3 pod.", "elementsJson": [ "headline_text", "paragraph", "photo" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.069+00", "seconds": 4.729, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/22", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-22", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Each Cloud TPUv3 (4 chips) has 128GB of high-bandwidth memory 2x that of the Cloud TPUv2.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-27ef578ceed7", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.94, "x": 0.03, "y": 0.38 }, "kind": "image", "text": "Photograph of a TPUv3 pod", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.069+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "32bb9059-8d23-4cfe-b4d7-605d0ff217a7", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.6, "x": 0.03, "y": 0.22 }, "kind": "list", "text": "What's next for Google? The TPUv3 announced at Google I/O 2018\nEach Cloud TPUv3 (4 chips) has 128GB of high-bandwidth memory 2x that of the Cloud TPUv2.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.069+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0e91ded6-48cc-4386-a062-1760f5556b0b", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "High-bandwidth memory: 128GB", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-2bf8f509efbe", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.03, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.069+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "05016f29-3900-4ad7-a863-55164808857c", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "ee9d9b02-12c9-4528-91da-63b9f80cdb2d", "evidence": "The slide contains a list with bullet points, which is a clear instantiation of the list presentation tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:32.193304+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 22, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 23, "type": "evidence", "title": "Slide from State of AI 2018 report.", "function": "illustrate_case", "rawType": "client_example", "block": null, "metadata": { "slideType": "client_example", "slideTypeCanon": { "name": "Client example", "slug": "client_example", "status": "active", "canonId": "019de52d-0c87-77cb-9867-613ae22e95ac", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 73, "componentCount": 11, "textChars": 455, "nDataPoints": 5, "notes": "Slide from State of AI 2018 report.", "elementsJson": [ "headline_text", "paragraph", "infographic", "big_number" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:55.571+00", "seconds": 2.968, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/23", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-23", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "What's next for NVIDIA? The HGX-2, announced at NVIDIA GTC May 2018", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-3d4ee0605aef", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.3, "y": 0.35 }, "kind": "image", "text": "Diagram of HGX-2 hardware architecture", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Infographic", "slug": "infographic", "status": "active", "canonId": "019de52c-fe26-750c-9d7d-9f34e21324e8", "version": 1, "description": null }, "framework": null }, "subkind": "infographic", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ede7d0c6-6219-434b-b144-3f8cb5b66ca4", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.1, "x": 0.58, "y": 0.75 }, "kind": "metric", "text": "2 PFLOPS Total Compute", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0c15ed48-bb3b-4803-89c3-0eba9139cee8", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.15, "x": 0.55, "y": 0.35 }, "kind": "metric", "text": "16 NVIDIA Tesla V100 GPUs", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "14e56c42-9380-4060-ae3b-2fe5015ecdc1", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.1, "x": 0.45, "y": 0.75 }, "kind": "metric", "text": "0.5TB Aggregate High-Bandwidth GPU Memory", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8f80e9c2-d5bc-4daf-8118-22c7ad5f7d34", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.1, "x": 0.32, "y": 0.75 }, "kind": "metric", "text": "24X Higher GPU-to-GPU Bandwidth", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e6b16be6-eb75-4577-a53b-389b271c0dd3", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.15, "x": 0.55, "y": 0.48 }, "kind": "metric", "text": "12 NVIDIA NVSwitches", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "edb8cbc2-4780-4528-904d-7de92d399955", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Total Compute: 2 PFLOPS", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:08+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80c-410cd71af1fb", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.9, "x": 0.02, "y": 0.27 }, "kind": "paragraph", "text": "Multi-precision computing platform for scientific computing (high precision) and AI workloads (low precision).", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5fd4b096-9bef-4f52-a153-744fbe899494", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.6, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "What's next for NVIDIA? The HGX-2, announced at NVIDIA GTC May 2018", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a921b6b1-497b-4dc5-bc54-c7b5d0246593", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.571+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c918562d-bb29-4ccb-926b-d1a142b2d5de", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Slide visual craft", "slug": "visual-craft", "status": "active", "bestFor": null, "canonId": "019df269-2d56-7739-ab65-cc2e091a109b", "version": 1, "bodyDocId": "019df269-2d32-736a-ae59-0250c5540c9a", "description": "What separates a 5/5-visual-craft deck from a 3/5 one.", "familyLabel": "slide-recipe", "categoryName": "Slide", "categorySlug": "slide" }, "agent": "designer", "layer": "slide", "agents": [ "designer" ], "matchId": "b3092564-982b-4650-9c0b-ac6d154c5e76", "evidence": "The slide contains an infographic with clear and simple visuals, which is a clear instantiation of the visual craft tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:32.272862+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 23, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 24, "type": "analysis", "title": "The chart uses a combination of a bar chart for absolute revenue and a line chart for percentage share.", "function": "quantify_impact", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "quantify_impact", "functionCanon": { "name": "Quantify impact", "slug": "quantify_impact", "status": "active", "canonId": "019de52d-1630-720d-b57d-fa7f76bebd8d", "version": 1, "description": null }, "density": "balanced", "densityScore": 30, "componentCount": 5, "textChars": 239, "nDataPoints": 6, "notes": "The chart uses a combination of a bar chart for absolute revenue and a line chart for percentage share.", "elementsJson": [ "bar_chart_vertical", "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:58.075+00", "seconds": 5.427, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/24", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-24", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.08, "w": 0.95, "x": 0.03, "y": 0.22 }, "kind": "callout", "text": "NVIDIA's datacenter business breaks $2B run-rate, is growing >100% year on year and accounts for almost 20% of their group revenue", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.075+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0f4326f0-9f9c-46a8-af2a-490c04d46971", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.55, "x": 0.22, "y": 0.35 }, "kind": "chart", "text": "Datacenter revenue growth chart", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.075+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cf345011-99e0-4668-b8c0-6b822e28438c", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.15, "x": 0.62, "y": 0.35 }, "kind": "legend", "text": "Datacenter revenue; Datacenter as % of total", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.075+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "efdce0d5-5951-4a30-b471-d39509738eda", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Datacenter revenue: $2B", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:09+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-7c12bb2ba5b2", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.03, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.075+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "87ed1c4d-c466-4fda-8ba8-57f6e76eda9f", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-7c87341a72c5", "evidence": "Title 'NVIDIA's datacenter business breaks $2B run-rate, growing >100% YoY'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 24, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "799a0830-2a20-4e97-9300-63475d82b601", "evidence": "The slide contains a bar chart, which is a clear instantiation of the chart selection guide tool.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:32.352489+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 24, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 25, "type": "analysis", "title": "The chart uses a green line to represent stock price over time, with a specific annotation for the 'Deep learning works starts to work' inflection point.", "function": "quantify_impact", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "quantify_impact", "functionCanon": { "name": "Quantify impact", "slug": "quantify_impact", "status": "active", "canonId": "019de52d-1630-720d-b57d-fa7f76bebd8d", "version": 1, "description": null }, "density": "dense", "densityScore": 50, "componentCount": 7, "textChars": 291, "nDataPoints": 12, "notes": "The chart uses a green line to represent stock price over time, with a specific annotation for the 'Deep learning works starts to work' inflection point.", "elementsJson": [ "headline_text", "action_title", "line_chart", "kpi_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:55.643+00", "seconds": 2.989, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/25", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-25", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "NVIDIA's enterprise value has 10x in 3 years since the deep learning revolution ignited", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:11+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-5d73e453d23f", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.8, "x": 0.05, "y": 0.3 }, "kind": "chart", "text": "NVIDIA stock price 1999-2018", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.643+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "41a40137-268f-4952-adf3-1b09829203a2", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.25, "x": 0.35, "y": 0.48 }, "kind": "legend", "text": "Deep learning works starts to work", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.643+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "45cabaa9-46a3-4ab3-b413-e9ad0f1878a8", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Enterprise Value: 10x", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:11+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-63f182d7aafe", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.7, "x": 0.05, "y": 0.75 }, "kind": "table", "text": "Financial metrics table", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.643+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "82deac59-af70-4e3a-b291-7dbeba574ccd", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.65, "x": 0.05, "y": 0.22 }, "kind": "title", "text": "NVIDIA's enterprise value has 10x in 3 years since the deep learning revolution ignited", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.643+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "20fed51e-f4b7-4b2e-93b9-e3cc8c9a274b", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.643+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "454f961f-1962-4dd6-b5a0-bf62e4f416d1", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-833999375475", "evidence": "Title 'NVIDIA's enterprise value has 10x in 3 years'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 25, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 26, "type": "analysis", "title": "The slide uses a combination of bulleted strategic highlights and a dual-axis bar/line chart to show financial performance.", "function": "analyze_data", "rawType": "financial_analysis", "block": null, "metadata": { "slideType": "financial_analysis", "slideTypeCanon": { "name": "Financial analysis", "slug": "financial_analysis", "status": "active", "canonId": "019de52d-0b45-709c-bfcc-0cf83dd00f13", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 45, "componentCount": 6, "textChars": 365, "nDataPoints": 10, "notes": "The slide uses a combination of bulleted strategic highlights and a dual-axis bar/line chart to show financial performance.", "elementsJson": [ "headline_text", "bullet_list", "bar_chart_stacked", "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:56.528+00", "seconds": 3.658, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/26", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-26", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.04, "w": 0.65, "x": 0.03, "y": 0.22 }, "kind": "callout", "text": "Intel's datacenter group accounts for 30% of the company's group revenue", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.528+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7da5ff7d-df46-4a0c-bc08-3ddc083159c2", "parentComponentId": null }, { "bbox": null, "kind": "callout", "text": "The data center TAM is expected to be >$70 billion by 2022, of which we currently have less than a 40% market share.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:10+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-c5b52eb82e79", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.35, "x": 0.58, "y": 0.35 }, "kind": "chart", "text": "5-year trend of Revenue ($B) and Op Income ($B) from 2013 to 2017.", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": null, "framework": null }, "subkind": "bar-line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.528+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "42b8bfb6-c45f-4b31-82b9-349a243386a4", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.48, "x": 0.04, "y": 0.35 }, "kind": "list", "text": "Highlights and segment imperatives: growth, TAM, market share, and product launches.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.528+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "496f7b24-fd1a-4ebf-a22c-ffb37f1c765e", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Revenue: $19.1bn", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:10+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-c86bbaf79171", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.528+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7419beaa-6747-438e-9d05-101e24631b95", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-87d804342658", "evidence": "Title quantifies Intel datacenter 30% revenue share.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 26, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 27, "type": "analysis", "title": "The slide uses a simple table format to segment the AI hardware market landscape.", "function": "summarize", "rawType": "market_landscape", "block": null, "metadata": { "slideType": "market_landscape", "slideTypeCanon": { "name": "Market landscape", "slug": "market_landscape", "status": "active", "canonId": "019de52d-0b6d-75db-a946-9ac362e3afc2", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "sparse", "densityScore": 23, "componentCount": 4, "textChars": 347, "nDataPoints": 0, "notes": "The slide uses a simple table format to segment the AI hardware market landscape.", "elementsJson": [ "data_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:55.476+00", "seconds": 2.582, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/27", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-27", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Many companies are developing custom AI chips", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:10+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-fa2e941843e8", "parentComponentId": null }, { "bbox": { "h": 0.68, "w": 0.96, "x": 0.02, "y": 0.23 }, "kind": "table", "text": "IC Vendors: Intel, Qualcomm, Nvidia, Samsung, AMD, Xilinx, IBM, STMicroelectronics, NXP, MediaTek, HiSilicon, Rockchip; Tech Giants & HPC Vendors: Google, Amazon AWS, Microsoft, Apple, Aliyun, Alibaba Group, Tencent Cloud, Baidu, Baidu Cloud, HUA", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.476+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "09c03707-e8db-45a8-93fc-e1250c010418", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.35, "x": 0.18, "y": 0.15 }, "kind": "title", "text": "Many companies are developing custom AI chips", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.476+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c0065f5e-5747-4604-8fcb-2672477c2409", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.476+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ed4ff2a6-5de7-4086-9ede-34a287edc7c6", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-895f07012035", "evidence": "Title 'Many companies are developing custom AI chips'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 27, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Small Multiples", "slug": "small-multiples", "status": "active", "bestFor": null, "canonId": "019dd956-b32d-7586-b368-baa1dd388ed4", "version": 1, "bodyDocId": null, "description": "Repeating similar charts for comparison across categories or time", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8eca-8cd9df00b36a", "evidence": "Market landscape implies many parallel logos/cards.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 27, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 28, "type": "setup", "function": "compare_options", "rawType": "comparison_table", "block": null, "metadata": { "slideType": "comparison_table", "slideTypeCanon": { "name": "Comparison table", "slug": "comparison_table", "status": "active", "canonId": "019de52d-0a06-73f8-8813-fe022fab1de2", "version": 1, "description": null }, "function": "compare_options", "functionCanon": { "name": "Compare options", "slug": "compare_options", "status": "active", "canonId": "019de52d-167f-7038-aebc-70a9d2e0e33b", "version": 1, "description": null }, "density": "balanced", "densityScore": 34, "componentCount": 5, "textChars": 442, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text", "action_title", "comparison_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.244+00", "seconds": 4.343, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/28", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-28", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Large technology companies are hedging their hardware suppliers, but there are few options to choose from", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:10+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80d-e78ba9c685d1", "parentComponentId": null }, { "bbox": { "h": 0.02, "w": 0.3, "x": 0.23, "y": 0.94 }, "kind": "source-note", "text": "Source: Company data, Goldman Sachs Global Investment Research.", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.244+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dc4ca05c-7c61-4e4d-bd06-db7433cf5d30", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.54, "x": 0.23, "y": 0.305 }, "kind": "table", "text": "Exhibit 8: Cloud companies leverage multiple hardware architectures in their datacenters. Hardware used by various cloud companies based on public statements.", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.244+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7e51d9eb-70a3-480a-b5f7-3f87b1731234", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.9, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Large technology companies are hedging their hardware suppliers, but there are few options to choose from", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.244+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dbf89cd6-1737-428b-a011-c092c079c519", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.244+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "46db3da8-1715-4986-ae37-65776218420d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-909994feab22", "evidence": "Title 'Large technology companies are hedging their hardware suppliers'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 28, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 29, "type": "analysis", "title": "The slide uses a combination of qualitative news clippings and quantitative financial data to support the thesis.", "function": "quantify_impact", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "quantify_impact", "functionCanon": { "name": "Quantify impact", "slug": "quantify_impact", "status": "active", "canonId": "019de52d-1630-720d-b57d-fa7f76bebd8d", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 7, "textChars": 400, "nDataPoints": 7, "notes": "The slide uses a combination of qualitative news clippings and quantitative financial data to support the thesis.", "elementsJson": [ "headline_text", "bar_chart_vertical", "screenshot" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:58.876+00", "seconds": 5.952, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/29", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-29", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "We expect cloud companies to spend $76bn on capex in 2020", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:11+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-27964c6b372b", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.45, "x": 0.5, "y": 0.3 }, "kind": "chart", "text": "Exhibit 9: We expect cloud companies to spend $76bn on capex in 2020", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.876+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d54964fd-3960-43e5-a88f-064de9e591b4", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.02, "y": 0.3 }, "kind": "list", "text": "News headlines regarding Facebook, Intel, Microsoft, and Amazon developing AI chips.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.876+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4c703577-a93c-40a8-8fca-1538e5c2ccdf", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Cloud capex ($mn): $76bn", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:11+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-2b81b2307a22", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.5, "y": 0.9 }, "kind": "source-note", "text": "Source: Company data, Goldman Sachs Global Investment Research.", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.876+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e6895720-d877-4177-8b0c-64e691f0c307", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.8, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Cloud giants are creating dedicated AI hardware and significantly growing their capex budgets", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.876+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3833c468-e0e3-4fe6-9829-0f20bed33fd0", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "AI hardware", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.876+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9258b711-a145-448b-94f7-d4dc0da3ee86", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-97c9f2224b12", "evidence": "Title quantifies $76bn cloud capex by 2020.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 29, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "AIDA Model", "slug": "aida-model", "status": "active", "bestFor": null, "canonId": "019dd956-86fc-773c-813f-d416714b8413", "version": 1, "bodyDocId": null, "description": "Attention, Interest, Desire, Action - classic persuasion sequence", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "bb74dc5d-4d24-4b98-879f-b07178a9a0f9", "evidence": "The slide presents a clear and concise message, grabbing the audience's attention with a bold title and highlighting a key statistic, which is a common structure used in the AIDA model.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.473228+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 29, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 29, "from": 22, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-4c3e393d7324", "evidence": "TPUv3 specs, NVIDIA HGX-2, NVIDIA datacenter $2B, EV 10x, Intel DC, custom chip landscape, hyperscaler $76B capex.", "position": 3, "objective": "Demonstrate hardware vendors are scaling fast", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 30, "type": "setup", "title": "The slide illustrates how AI models interpret visual data by identifying objects as 'nouns'.", "function": "present_framework", "rawType": "diagnosis", "block": null, "metadata": { "slideType": "diagnosis", "slideTypeCanon": { "name": "Diagnosis", "slug": "diagnosis", "status": "active", "canonId": "019de52d-0cd7-7212-898f-0464041ff8d8", "version": 1, "description": null }, "function": "present_framework", "functionCanon": { "name": "Present framework", "slug": "present_framework", "status": "active", "canonId": "019de52d-1403-7667-9100-d2ac6709f750", "version": 1, "description": null }, "density": "balanced", "densityScore": 31, "componentCount": 5, "textChars": 381, "nDataPoints": 0, "notes": "The slide illustrates how AI models interpret visual data by identifying objects as 'nouns'.", "elementsJson": [ "screenshot", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:55.376+00", "seconds": 2.415, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/30", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-30", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "AI models associate pixels to objects (semantic segmentation) or identify what objects are shown (classification)", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:11+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-0c0150d1b51c", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.35, "x": 0.51, "y": 0.35 }, "kind": "image", "text": "Object classification example", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.376+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "277b53b2-0537-4788-8f3d-1988d7e11689", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.35, "x": 0.14, "y": 0.35 }, "kind": "image", "text": "Semantic image segmentation example", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.376+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3f6a1c97-ba73-4ab3-97b0-164210958488", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "AI models associate pixels to objects (semantic segmentation) or identify what objects are shown (classification)", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.376+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2a9eaa5e-0b12-46e2-a937-0ba1730c78b6", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Traditional computer vision describes visual scenes by learning to detect objects ('nouns')", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:55.376+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5862c772-7660-4d90-8fd0-f2bb592a3ee2", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-9aa29bd9b41c", "evidence": "Title 'Traditional computer vision describes scenes by detecting objects (nouns)'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 30, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-9e8bc4f23920", "evidence": "Nouns/verbs metaphor frames CV vs scene understanding.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 30, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 35, "from": 30, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-514142d7981d", "evidence": "Big picture (CV detects nouns) -> focus on verbs/common sense -> specific case (multi-viewpoint scenes).", "position": 4, "objective": "Zoom from object detection to full scene understanding", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 31, "type": "tension", "title": "The slide uses four examples of AI mislabeling images to demonstrate the lack of 'common sense world models' in AI.", "function": "frame_problem", "rawType": "problem_statement", "block": null, "metadata": { "slideType": "problem_statement", "slideTypeCanon": { "name": "Problem statement", "slug": "problem_statement", "status": "active", "canonId": "019de52d-0879-768d-bba8-8da8a5a121d4", "version": 1, "description": null }, "function": "frame_problem", "functionCanon": { "name": "Frame problem", "slug": "frame_problem", "status": "active", "canonId": "019de52d-13db-753e-ad70-e9b2d0dab398", "version": 1, "description": null }, "density": "overcrowded", "densityScore": 83, "componentCount": 11, "textChars": 1126, "nDataPoints": 0, "notes": "The slide uses four examples of AI mislabeling images to demonstrate the lack of 'common sense world models' in AI.", "elementsJson": [ "headline_text", "paragraph", "screenshot" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:56.653+00", "seconds": 3.386, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/31", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-31", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "This suggests that having a common sense world model of objects and people is required for an AI system to truly understand what's happening in a visual scene.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:13+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-7955bad9077f", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.2, "x": 0.51, "y": 0.5 }, "kind": "image", "text": "People with laptop labeled as woman holding teddy bear", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "35fe94cd-df8d-4623-9380-d5107103c843", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.2, "x": 0.28, "y": 0.5 }, "kind": "image", "text": "Ferret on remote labeled as cat on couch", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5d7306fb-c18c-41bb-9e7b-f2b517de4ef8", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.2, "x": 0.06, "y": 0.5 }, "kind": "image", "text": "Baby holding toothbrush labeled as baseball bat", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5d99add9-be2d-463e-9034-e282862b6722", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.2, "x": 0.74, "y": 0.5 }, "kind": "image", "text": "Empty road labeled as horse in middle of road", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9935d279-143e-43c1-b6d7-451eb8e1247a", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.2, "x": 0.06, "y": 0.86 }, "kind": "paragraph", "text": "a young boy is holding a baseball bat.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "23fdd42e-1e6f-4181-96c6-5ce596feb19d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.2, "x": 0.28, "y": 0.86 }, "kind": "paragraph", "text": "a cat is sitting on a couch with a remote control.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3517a98f-1a59-4288-88e0-ca73780ffc30", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.2, "x": 0.74, "y": 0.86 }, "kind": "paragraph", "text": "a horse is standing in the middle of a road.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bc6bbfdc-80ed-4dde-bae1-f1f12fa64630", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.2, "x": 0.51, "y": 0.86 }, "kind": "paragraph", "text": "a woman holding a teddy bear in front of a mirror.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bf80e007-a2df-4e48-a9be-3a492053969d", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "AI models make obvious mistakes when asked to describe a visual scene based on their understanding of objects. Image captioning helps expose the knowledge that computer vision systems learn by training on images labeled with the objects they contain. Such computer vision models make seemingly obvious mistakes when attempting to describe visual scenes. This suggests that having a common sense world model of objects and people is required for an AI system to truly understand what's happening in a visual scene.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dd1a7963-8cdb-473d-ab42-613010bb25ab", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "However, detecting objects in images is not enough to produce real scene understanding", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.653+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5f8a83d6-b540-44fd-8d3d-84c316439c2a", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-a05ddd60664e", "evidence": "Title 'detecting objects in images is not enough to produce real scene understanding'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 31, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "2228f7da-f69c-43ec-8ec7-801aa19269cb", "evidence": "The slide clearly articulates a problem, providing context and highlighting the limitations of current solutions, which is a key component of the problem statement canvas.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:38.568648+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 31, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 35, "from": 30, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-514142d7981d", "evidence": "Big picture (CV detects nouns) -> focus on verbs/common sense -> specific case (multi-viewpoint scenes).", "position": 4, "objective": "Zoom from object detection to full scene understanding", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 32, "type": "analysis", "title": "Slide from state.ai 2018 report.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 74, "componentCount": 12, "textChars": 492, "nDataPoints": 3, "notes": "Slide from state.ai 2018 report.", "elementsJson": [ "headline_text", "paragraph", "icon_grid", "screenshot" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.092+00", "seconds": 3.788, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/32", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-32", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "A promising approach to learning common sense uses deep learning and labeled videos of actions with objects", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:11+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-40c62c6c8c24", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.28, "x": 0.06, "y": 0.45 }, "kind": "image", "text": "Something something dataset", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2f207b39-f3ce-4b53-ba43-2515eddf2d93", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.28, "x": 0.66, "y": 0.45 }, "kind": "image", "text": "Moments in Time dataset", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "354bcde0-13d5-4b7a-ab6e-37ae1d9f90ac", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.28, "x": 0.36, "y": 0.45 }, "kind": "image", "text": "Kinetics dataset", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "763f9e96-69f8-41e2-80fb-f0772e729712", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.2, "x": 0.1, "y": 0.31 }, "kind": "image", "text": "twentybn", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3c2b1454-db26-4bac-b40c-0fe7b365a2ae", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.42, "y": 0.31 }, "kind": "image", "text": "DeepMind", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3d7c9c20-ae7c-4710-b02f-e4dc0608a546", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.25, "x": 0.69, "y": 0.31 }, "kind": "image", "text": "MIT-IBM Watson AI Lab", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e585dbfd-1de7-45ce-8f42-7eab212fead3", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.28, "x": 0.36, "y": 0.38 }, "kind": "metric", "text": "300k videos of 400 human actions", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "47a7346f-c32a-4880-bbde-abdf7a1ae176", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.28, "x": 0.66, "y": 0.38 }, "kind": "metric", "text": "1M+ videos of YouTube actions", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "70f88a9f-ff31-4113-b086-020d2ae77a02", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.28, "x": 0.06, "y": 0.38 }, "kind": "metric", "text": "1M+ videos of real world actions", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "714de004-7c04-4655-9db2-f5674da12064", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.9, "x": 0.04, "y": 0.23 }, "kind": "paragraph", "text": "A promising approach to learning common sense uses deep learning and labeled videos of actions with objects", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3141cb71-8cbc-4639-8d9f-8814eb12693b", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "True scene understanding requires understanding actions ('verbs') and common sense", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.092+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "638f92a1-bb7f-467e-92db-0bdee8eecc2b", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-a5bca6c0c634", "evidence": "Title 'True scene understanding requires understanding actions (verbs) and common sense'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 32, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-a9c29f584dc8", "evidence": "Verbs vs nouns metaphor extended.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 32, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Three-Act Structure", "slug": "three-act-structure", "status": "active", "bestFor": null, "canonId": "019dd956-8812-7796-9105-e7bc0e405681", "version": 1, "bodyDocId": null, "description": "Setup, Confrontation, Resolution - classic dramatic structure", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "441f482b-4600-4ddd-b37e-1e6ff2b3a324", "evidence": "The slide presents a clear progression of ideas, introducing a problem, describing a solution, and highlighting its potential impact, which follows a three-act structure.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:38.629451+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 32, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 35, "from": 30, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-514142d7981d", "evidence": "Big picture (CV detects nouns) -> focus on verbs/common sense -> specific case (multi-viewpoint scenes).", "position": 4, "objective": "Zoom from object detection to full scene understanding", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 33, "type": "setup", "title": "The diagram shows a cyclical process involving taxonomy creation, crowd-acting, model training, and customer testing, with feedback loops.", "function": "present_framework", "rawType": "implementation_plan", "block": null, "metadata": { "slideType": "implementation_plan", "slideTypeCanon": { "name": "Implementation plan", "slug": "implementation_plan", "status": "active", "canonId": "019de52d-0d75-700b-ae9f-f6069ba6e37a", "version": 1, "description": null }, "function": "present_framework", "functionCanon": { "name": "Present framework", "slug": "present_framework", "status": "active", "canonId": "019de52d-1403-7667-9100-d2ac6709f750", "version": 1, "description": null }, "density": "balanced", "densityScore": 34, "componentCount": 4, "textChars": 622, "nDataPoints": 0, "notes": "The diagram shows a cyclical process involving taxonomy creation, crowd-acting, model training, and customer testing, with feedback loops.", "elementsJson": [ "headline_text", "paragraph", "process_diagram" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:58.847+00", "seconds": 5.182, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/33", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-33", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "If a deep learning model can recognise and disambiguate nuanced actions from video, it should have internalised common sense about the world.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:13+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-92c82fbc4b0c", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.7, "x": 0.15, "y": 0.38 }, "kind": "diagram", "text": "Process flow showing 1. Taxonomy, 2. Crowd-acting, 3. Model training, 4. Customer testing with feedback loops.", "attrs": null, "canon": { "kind": { "name": "Diagram", "slug": "diagram", "status": "active", "canonId": "019de52c-f9f0-734a-81de-15382ff28c65", "version": 1, "description": "Schematic or relational diagram." }, "tool": null, "subkind": { "name": "Process flow", "slug": "process", "status": "active", "canonId": "019de52d-039c-718f-ad3a-efe9c4b3c838", "version": 1, "description": null }, "framework": null }, "subkind": "process", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.847+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "69c5626b-68a5-4a23-a0b9-3f00f53f70db", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.95, "x": 0.03, "y": 0.21 }, "kind": "paragraph", "text": "Enlist people to create videos that describe actions of interest, e.g. pretending to drop something off something. If a deep learning model can recognise and disambiguate nuanced actions from video, it should have internalised common sense about the world. This is also called “intuitive physics”.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.847+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6c0bba8e-b8b7-4d55-93e1-8efb128c0aa4", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.6, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "Building datasets for teaching machine learning models to understand video", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.847+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9e301047-3b23-4d91-a86a-20bac8ea8bfe", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-af9d57e86a31", "evidence": "Title describes dataset-building approach.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 33, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [ { "matchId": "52a32db1-36bf-415a-a9a6-8bebd8ab1b96", "evidence": "The diagram depicts a cyclical process with feedback loops between taxonomy, crowd-acting, model training, and testing.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:43:58.847+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 33, "frameworkName": "flywheel" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 35, "from": 30, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-514142d7981d", "evidence": "Big picture (CV detects nouns) -> focus on verbs/common sense -> specific case (multi-viewpoint scenes).", "position": 4, "objective": "Zoom from object detection to full scene understanding", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 34, "type": "evidence", "title": "The slide highlights the capability of TwentyBN-trained networks compared to standard image-trained networks.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "overcrowded", "densityScore": 79, "componentCount": 11, "textChars": 876, "nDataPoints": 0, "notes": "The slide highlights the capability of TwentyBN-trained networks compared to standard image-trained networks.", "elementsJson": [ "photo", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.309+00", "seconds": 8.414, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/34", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-34", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Examples of caption predictions generated by a deep learning model trained on crowd-acted data", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-4f00bd78081d", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.45, "x": 0.03, "y": 0.33 }, "kind": "image", "text": "Sequence of images showing coin stacking", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1a3b7dee-53d0-4b9f-99f1-4e433dff5ced", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.45, "x": 0.03, "y": 0.79 }, "kind": "image", "text": "Sequence of images showing cup removal", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6ad84b32-89f9-441e-92f0-8216f2018428", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.45, "x": 0.03, "y": 0.56 }, "kind": "image", "text": "Sequence of images showing flower pot interaction", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "952358bc-0000-4060-a5e8-ec433dd06efc", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.3, "x": 0.61, "y": 0.415 }, "kind": "image", "text": "Hand picking up remote control", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f27a7228-dd1e-4f39-b99d-5827e0d06376", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.8, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "Examples of caption predictions generated by a deep learning model trained on crowd-acted data", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "94596863-5259-4e0b-80f3-544ad744fec8", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.45, "x": 0.03, "y": 0.94 }, "kind": "paragraph", "text": "(e) Ground Truth: Removing cup, revealing little cup behind.\n(f) Model output: Removing mug, revealing cup behind.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "194c84de-0344-46fc-bb42-e5cf03b25522", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.45, "x": 0.03, "y": 0.71 }, "kind": "paragraph", "text": "(c) Ground Truth: Lifting up one end of flower pot, then letting it drop down.\n(d) Model output: Lifting up one end of bucket, then letting it drop down.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3108e9ae-cba6-4f84-94e5-4ed70de19cef", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.45, "x": 0.03, "y": 0.48 }, "kind": "paragraph", "text": "(a) Ground Truth: Stacking 4 coins.\n(b) Model output: Piling coins up.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a86be315-8613-47a8-8a0c-4f1d49d6e38f", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.4, "x": 0.54, "y": 0.85 }, "kind": "paragraph", "text": "Image-trained network: \"remote control\"\nTwentyBN-trained network: \"pretending to pick black remote control up\"", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d90e2a41-a0c9-443e-9374-403ced973a70", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Deep learning models can actually understand the verbs as well as the nouns in video", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.309+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0ec36a40-793f-4f07-a713-24789bb2bbc5", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-b0062aa4438d", "evidence": "Title 'Deep learning models can actually understand the verbs as well as the nouns'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 34, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 35, "from": 30, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-514142d7981d", "evidence": "Big picture (CV detects nouns) -> focus on verbs/common sense -> specific case (multi-viewpoint scenes).", "position": 4, "objective": "Zoom from object detection to full scene understanding", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 35, "type": "evidence", "title": "DeepMind research slide from 2018 State of AI report.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 41, "componentCount": 5, "textChars": 634, "nDataPoints": 0, "notes": "DeepMind research slide from 2018 State of AI report.", "elementsJson": [ "headline_text", "paragraph", "other", "other" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:56.707+00", "seconds": 2.778, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/35", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-35", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "If an ML system correctly predicts new viewpoints of the same scene, it has internalised knowledge of that scene", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-1ac51cfea18f", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.35, "x": 0.12, "y": 0.45 }, "kind": "image", "text": "Diagram A showing 3D scene with multiple camera viewpoints", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.707+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2093cc98-7037-42b6-9b72-7feaf62fc0aa", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.4, "x": 0.52, "y": 0.45 }, "kind": "image", "text": "Diagram B showing grid of observations vs predictions vs truth", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.707+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b1314b43-aabe-46fd-bcaf-da725c9dd1dd", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.96, "x": 0.02, "y": 0.18 }, "kind": "paragraph", "text": "If an ML system correctly predicts new viewpoints of the same scene, it has internalised knowledge of that scene. The “Generative Query Network” (GQN) can do this without human labels or domain knowledge, suggesting that it captures the identities, positions, colors, and counts of objects in the scenes it observes.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.707+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b27b8e2e-e340-4aee-b4d6-6fc93715fb22", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.96, "x": 0.02, "y": 0.11 }, "kind": "title", "text": "Machines can also understand visual scenes by learning to see from multiple viewpoints", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.707+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6c198848-cb4a-4b10-9db0-18a6f9b5f8f5", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-b6c66082ba7c", "evidence": "Title 'Machines can also understand visual scenes by learning from multiple viewpoints'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 35, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 35, "from": 30, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-514142d7981d", "evidence": "Big picture (CV detects nouns) -> focus on verbs/common sense -> specific case (multi-viewpoint scenes).", "position": 4, "objective": "Zoom from object detection to full scene understanding", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 36, "type": "appendix", "title": "The diagram shows the standard RL feedback loop: Agent receives observations and rewards, and performs actions.", "function": "present_framework", "rawType": "other", "block": null, "metadata": { "slideType": "other", "slideTypeCanon": { "name": "Other (fallback)", "slug": "other", "status": "active", "canonId": "019de52d-0af5-71ed-83de-d51ae245fc43", "version": 1, "description": null }, "function": "present_framework", "functionCanon": { "name": "Present framework", "slug": "present_framework", "status": "active", "canonId": "019de52d-1403-7667-9100-d2ac6709f750", "version": 1, "description": null }, "density": "balanced", "densityScore": 29, "componentCount": 5, "textChars": 343, "nDataPoints": 0, "notes": "The diagram shows the standard RL feedback loop: Agent receives observations and rewards, and performs actions.", "elementsJson": [ "process_diagram", "photo" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.489+00", "seconds": 3.196, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/36", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-36", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "A game is the world model used by a reinforcement learning (RL) system to learn behaviors by trial and error", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:18+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-a6b224712810", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.9, "x": 0.05, "y": 0.35 }, "kind": "diagram", "text": "Reinforcement Learning Loop", "attrs": null, "canon": { "kind": { "name": "Diagram", "slug": "diagram", "status": "active", "canonId": "019de52c-f9f0-734a-81de-15382ff28c65", "version": 1, "description": "Schematic or relational diagram." }, "tool": null, "subkind": { "name": "Process flow", "slug": "process", "status": "active", "canonId": "019de52d-039c-718f-ad3a-efe9c4b3c838", "version": 1, "description": null }, "framework": null }, "subkind": "process", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.489+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1706b22d-db21-4d80-84e0-157dbcc5dc5d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "A game is the world model used by a reinforcement learning (RL) system to learn behaviors by trial and error", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.489+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "af0a581e-be8a-465c-9568-904deefb4347", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.82, "y": 0.94 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.489+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b849abd6-8eca-4154-8378-a29912fc0c62", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.95, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "RL systems can learn goal-oriented behavior within simulated environments, i.e. games", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.489+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7bc81f0f-0681-4090-80eb-9de12fc34737", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [ { "matchId": "c14f3493-0f12-432c-9d39-da62e5118bb9", "evidence": "Diagram showing Agent-Environment interaction via Observations, Rewards, and Actions", "framework": null, "confidence": 1, "extraction": { "at": "2026-05-02 17:43:57.489+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 36, "frameworkName": "Reinforcement Learning Loop" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 39, "from": 36, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-55bec48734a5", "evidence": "AlphaZero, OpenAI Dota2, world-model RL — three parallel evidence pieces of RL progress.", "position": 5, "objective": "Show RL agents now beat human experts across domains", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 37, "type": "evidence", "title": "The slide uses two charts: a line chart showing training progress over 40 days and a bar chart comparing peak Elo ratings of various Go engines.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 64, "componentCount": 7, "textChars": 612, "nDataPoints": 15, "notes": "The slide uses two charts: a line chart showing training progress over 40 days and a bar chart comparing peak Elo ratings of various Go engines.", "elementsJson": [ "headline_text", "paragraph", "line_chart", "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:56.77+00", "seconds": 2.428, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/37", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-37", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "AlphaZero achieves superhuman performance after 40 days of training.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-11a3a3e4d116", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.4, "x": 0.55, "y": 0.45 }, "kind": "chart", "text": "Elo rating comparison of Go engines", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.77+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5e702e15-b78a-4620-b4b0-d4fd60519aa3", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.45, "x": 0.05, "y": 0.45 }, "kind": "chart", "text": "AlphaZero training progress over 40 days", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.77+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b85d5ba1-928a-4607-82ba-7bfaea110587", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Elo rating", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-15510838f3ca", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "AlphaZero is one neural network trained through self-play without human supervision or historical player data to predict moves and chances of winning from a particular board position", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.77+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0b45e94c-0436-4b02-af6e-9746218b0d52", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.33 }, "kind": "paragraph", "text": "Strikingly, the more elegant AlphaZero system surpasses all other versions of AlphaGo (which is based on two neural networks). AlphaZero achieves superhuman performance after 40 days of training.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.77+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9d062ee4-26f3-4530-8768-60655aee8653", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "AlphaZero showed that a deep RL system can learn from scratch to beat Go champions", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:56.77+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bedb2eca-ba8e-4d8f-ad5d-e69ae1689868", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-b80e848e0ccc", "evidence": "Title 'AlphaZero showed that a deep RL system can learn from scratch to beat Go champions'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 37, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Sinatra Test", "slug": "sinatra-test", "status": "active", "bestFor": null, "canonId": "019dd956-be4a-775f-8038-d4e7e58dfc5a", "version": 1, "bodyDocId": null, "description": "If you can make it there, you can make it anywhere - using flagship example", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-be170044e6f0", "evidence": "Flagship case (AlphaZero, 40 days) carries the RL claim.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 37, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 39, "from": 36, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-55bec48734a5", "evidence": "AlphaZero, OpenAI Dota2, world-model RL — three parallel evidence pieces of RL progress.", "position": 5, "objective": "Show RL agents now beat human experts across domains", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 38, "type": "evidence", "title": "The slide includes a screenshot of the game interface and a comparison table of training infrastructure and performance metrics.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 53, "componentCount": 5, "textChars": 837, "nDataPoints": 12, "notes": "The slide includes a screenshot of the game interface and a comparison table of training infrastructure and performance metrics.", "elementsJson": [ "headline_text", "paragraph", "screenshot", "comparison_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.572+00", "seconds": 3.053, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/38", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-38", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "That RL agents can collaborate in teams to beat teams of humans is notable given the space of possible actions agents can take and the large maps they can interact with.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-82a9120f0563", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.55, "x": 0.05, "y": 0.45 }, "kind": "image", "text": "Dota 2 gameplay screenshot with UI overlays", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Screenshot", "slug": "screenshot", "status": "active", "canonId": "019de52c-fd87-7509-8391-3d79fb339379", "version": 1, "description": null }, "framework": null }, "subkind": "screenshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.572+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9471560a-7cfa-4142-af82-58b62125c34b", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "OpenAI Five is a team of 5 agents that learn through RL-based self-play to optimize their gameplay policy. The agents each have their own neural networks trained through RL to yield long-term planning behavior in a gameplay environment that is partially-observable and high-dimensional. That RL agents can collaborate in teams to beat teams of humans is notable given the space of possible actions agents can take and the large maps they can interact with.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.572+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fdeca259-e2b6-45f7-b750-87209ab314dd", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.32, "x": 0.63, "y": 0.45 }, "kind": "table", "text": "Comparison table of OpenAI 1v1 Bot vs OpenAI Five infrastructure and training metrics", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.572+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a9278b55-815d-4484-b823-7dac3575dcb1", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.13 }, "kind": "title", "text": "OpenAI's multi-agent RL system learns to play complex real-time strategy game, Dota2", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.572+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "511b208a-29a4-48f8-9484-c2130ac619ea", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-c39aa2ae7b9a", "evidence": "Title 'OpenAI's multi-agent RL system learns to play Dota2'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 38, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 39, "from": 36, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-55bec48734a5", "evidence": "AlphaZero, OpenAI Dota2, world-model RL — three parallel evidence pieces of RL progress.", "position": 5, "objective": "Show RL agents now beat human experts across domains", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 39, "type": "evidence", "title": "The slide illustrates the 'World Models' architecture (Ha & Schmidhuber, 2018).", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 6, "textChars": 599, "nDataPoints": 7, "notes": "The slide illustrates the 'World Models' architecture (Ha & Schmidhuber, 2018).", "elementsJson": [ "headline_text", "paragraph", "process_diagram", "data_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.888+00", "seconds": 3.151, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/39", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-39", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Using this world model allows an AI agent to perform at its best", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-a9c959835ef4", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.05, "y": 0.45 }, "kind": "diagram", "text": "Schematic for building a world model", "attrs": null, "canon": { "kind": { "name": "Diagram", "slug": "diagram", "status": "active", "canonId": "019de52c-f9f0-734a-81de-15382ff28c65", "version": 1, "description": "Schematic or relational diagram." }, "tool": null, "subkind": { "name": "Process flow", "slug": "process", "status": "active", "canonId": "019de52d-039c-718f-ad3a-efe9c4b3c838", "version": 1, "description": null }, "framework": null }, "subkind": "process", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.888+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c554330d-a566-42cf-900e-daa0345187de", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Average Score over 100 Random Tracks: 906", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-ac584be7355f", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "Here, an RL agent learns optimal behaviors within a world model it imagined for itself. The agent observes the game environment, creates its own understanding of each frame (VAE), uses this understanding to predict the next frame (MDN-RNN) and then trains its behavior to optimize a goal (C) in the imagined environment.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.888+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a89b131e-76c4-412a-88db-c07e644083f1", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.5, "x": 0.45, "y": 0.45 }, "kind": "table", "text": "Using this world model allows an AI agent to perform at its best", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.888+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "99d2e0d5-e0d1-42ed-a127-3487a7d68318", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.85, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "RL agents can also build their own world models and be trained within them", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.888+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dde5fc03-5323-4e30-9e88-76c3d6c4f200", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-c41cb2294f10", "evidence": "Title 'RL agents can also build their own world models'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 39, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [ { "matchId": "76a3b04a-5ab0-4bee-a49a-8b29c22d5d22", "evidence": "Diagram shows VAE, MDN-RNN, and Controller (C) architecture", "framework": null, "confidence": 1, "extraction": { "at": "2026-05-02 17:43:57.888+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 39, "frameworkName": "World Models" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 39, "from": 36, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-55bec48734a5", "evidence": "AlphaZero, OpenAI Dota2, world-model RL — three parallel evidence pieces of RL progress.", "position": 5, "objective": "Show RL agents now beat human experts across domains", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 40, "type": "analysis", "title": "The chart uses a humorous, hand-drawn style to illustrate the sudden surge in research interest.", "function": "establish_context", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "balanced", "densityScore": 47, "componentCount": 6, "textChars": 412, "nDataPoints": 7, "notes": "The chart uses a humorous, hand-drawn style to illustrate the sudden surge in research interest.", "elementsJson": [ "headline_text", "bullet_list", "data_table", "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:00.357+00", "seconds": 5.502, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/40", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-40", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "After many years of scandals, the research community is finally working to stem bias in ML models", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-d1b9bcfde380", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.55, "y": 0.35 }, "kind": "chart", "text": "BRIEF HISTORY OF FAIRNESS IN ML", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.357+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a274dfe4-182b-4001-8b6f-20d478bc991a", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.45, "x": 0.05, "y": 0.35 }, "kind": "list", "text": "SOFTWARE SCANDALS: Prominent incidents that highlight the effect of algorithmic bias", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.357+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "64d00a78-ead0-4f0f-936a-574e6608a6b2", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Research papers on ML fairness", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-d7f58e9a6fe7", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.8, "x": 0.05, "y": 0.22 }, "kind": "title", "text": "After many years of scandals, the research community is finally working to stem bias in ML models", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.357+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "74fe4282-dbe7-4b46-8b0e-87020330a722", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Fairness in machine learning: How do we ensure our models are not biased?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.357+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7eb3ad16-e5fe-4232-b6ac-80a74b74c415", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-ca8bf72fb965", "evidence": "Title posed as question 'How do we ensure our models are not biased?'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 40, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 43, "from": 40, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5930175dc2d9", "evidence": "Big picture (fairness) -> specific examples (Turkish gender, racial) -> 5-type allocation bias table.", "position": 6, "objective": "Diagnose ML bias by isolating its sources", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 41, "type": "evidence", "title": "The slide uses screenshots of Google Translate to show how machine learning models can perpetuate societal stereotypes.", "function": "illustrate_case", "rawType": "client_example", "block": null, "metadata": { "slideType": "client_example", "slideTypeCanon": { "name": "Client example", "slug": "client_example", "status": "active", "canonId": "019de52d-0c87-77cb-9867-613ae22e95ac", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 26, "componentCount": 4, "textChars": 409, "nDataPoints": 0, "notes": "The slide uses screenshots of Google Translate to show how machine learning models can perpetuate societal stereotypes.", "elementsJson": [ "screenshot", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.018+00", "seconds": 2.133, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/41", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-41", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Turkish is a gender-neutral language, yet Google Translate swaps the gender of the pronouns when translating from English to Turkish and back to English", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-34f4dd384871", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.8, "x": 0.1, "y": 0.35 }, "kind": "image", "text": "Google Translate interface showing gender bias", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Screenshot", "slug": "screenshot", "status": "active", "canonId": "019de52c-fd87-7509-8391-3d79fb339379", "version": 1, "description": null }, "framework": null }, "subkind": "screenshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.018+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b984c764-97ec-4950-b786-07edb72d0311", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "Turkish is a gender-neutral language, yet Google Translate swaps the gender of the pronouns when translating from English to Turkish and back to English", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.018+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2d845c4d-8c2b-4934-a7a4-475ebd9c983f", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "An example of biased machine learning systems: Stereotyping", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.018+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "63a3427c-9ab7-4daa-ac85-a0c88daa915e", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-ccbd7bd7560a", "evidence": "Title 'An example of biased machine learning systems: Stereotyping'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 41, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Anchoring Bias", "slug": "anchoring-bias", "status": "active", "bestFor": null, "canonId": "019dd956-9254-73de-8f1b-c7250f8f4c9c", "version": 1, "bodyDocId": null, "description": "First piece of information becomes reference point for all subsequent judgments", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect", "Storyteller" ], "matchId": "d705c7de-1ddd-4aa4-8fa4-06c41bc36559", "evidence": "Turkish is a gender-neutral language, yet Google Translate swaps the gender of the pronouns when translating from English to Turkish and back to English", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.547572+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 41, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-d3fddd63b37b", "evidence": "Specific Turkish-Google Translate example.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 41, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Singularity Effect", "slug": "singularity-effect", "status": "active", "bestFor": null, "canonId": "019dd956-dcd3-7287-94d8-5a49496c2ddb", "version": 1, "bodyDocId": null, "description": "People care more about individuals than statistics - use specific examples", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-d68d2afdc37e", "evidence": "Single concrete bias example over abstract stats.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 41, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 43, "from": 40, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5930175dc2d9", "evidence": "Big picture (fairness) -> specific examples (Turkish gender, racial) -> 5-type allocation bias table.", "position": 6, "objective": "Diagnose ML bias by isolating its sources", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 42, "type": "evidence", "title": "The slide uses external media headlines and a personal anecdote to illustrate the problem of algorithmic bias.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 74, "componentCount": 8, "textChars": 1272, "nDataPoints": 0, "notes": "The slide uses external media headlines and a personal anecdote to illustrate the problem of algorithmic bias.", "elementsJson": [ "headline_text", "paragraph", "quote_block", "screenshot" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:01.422+00", "seconds": 6.482, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/42", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-42", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "When trained on datasets that do not appropriately reflect diversity of skin color, computer vision systems exhibit offensive racial bias", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-6bba1d4b02b9", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.45, "x": 0.48, "y": 0.68 }, "kind": "image", "text": "Google Photos Mistakenly Labels Black People 'Gorillas'", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Screenshot", "slug": "screenshot", "status": "active", "canonId": "019de52c-fd87-7509-8391-3d79fb339379", "version": 1, "description": null }, "framework": null }, "subkind": "screenshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.422+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0719dc81-5bab-4f68-b041-fd976838acde", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.45, "x": 0.02, "y": 0.68 }, "kind": "image", "text": "Facial Recognition Is Accurate, if You're a White Guy by Steve Lohr", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Screenshot", "slug": "screenshot", "status": "active", "canonId": "019de52c-fd87-7509-8391-3d79fb339379", "version": 1, "description": null }, "framework": null }, "subkind": "screenshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.422+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "12d6d07e-5749-4f4b-a7e1-152fa983796a", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.3, "x": 0.02, "y": 0.36 }, "kind": "image", "text": "When the Robot Doesn't See Dark Skin by Joy Buolamwini", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Screenshot", "slug": "screenshot", "status": "active", "canonId": "019de52c-fd87-7509-8391-3d79fb339379", "version": 1, "description": null }, "framework": null }, "subkind": "screenshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.422+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d7414afa-47c4-4b3b-8741-8429f9aea6a2", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "When trained on datasets that do not appropriately reflect diversity of skin color, computer vision systems exhibit offensive racial bias", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.422+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "53386d36-edf0-4c5f-b5c8-a8a51dbe0dfa", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.6, "x": 0.32, "y": 0.36 }, "kind": "quote", "text": "When I was a college student using A.I.-powered facial detection software for a coding project, the robot I programmed couldn't detect my dark-skinned face. I had to borrow my white roommate's face to finish the assignment. Later, working on another project as a graduate student at the M.I.T. Media Lab, I resorted to wearing a white mask to have my presence recognized.", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": null, "subkind": { "name": "Testimonial", "slug": "testimonial", "status": "active", "canonId": "019de52c-fc21-71b6-9ac3-8266a64945e8", "version": 1, "description": null }, "framework": null }, "subkind": "testimonial", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.422+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5f1b4913-380a-4916-bfb9-73735c6e97f8", "parentComponentId": null }, { "bbox": null, "kind": "quote", "text": "When I was a college student using A.I.-powered facial detection software for a coding project, the robot I programmed couldn't detect my dark-skinned face. I had to borrow my white roommate's face to finish the assignment. Later, working on another project as a graduate student at the M.I.T. Media Lab, I resorted to wearing a white mask to have my presence recognized. — Joy Buolamwini", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": { "name": "Authority citation", "slug": "authority-citation", "status": "active", "bestFor": null, "canonId": "019dd9e1-49ae-7557-89ad-3fca70469b45", "version": 1, "bodyDocId": null, "description": "Quoting an authority figure, study, or precedent to lend weight.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-6f76ebe68ea4", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Another example of biased machine learning systems: Racial bias", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.422+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0a612dc7-eeb6-4e2d-a8d5-eb33e9480d76", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-d87139e35c6e", "evidence": "Title 'Another example of biased machine learning systems: Racial bias'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 42, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Anchoring Bias", "slug": "anchoring-bias", "status": "active", "bestFor": null, "canonId": "019dd956-9254-73de-8f1b-c7250f8f4c9c", "version": 1, "bodyDocId": null, "description": "First piece of information becomes reference point for all subsequent judgments", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect", "Storyteller" ], "matchId": "a9e06996-6bef-4523-b486-ec40da386df7", "evidence": "When trained on datasets that do not appropriately reflect diversity of skin color, computer vision systems exhibit offensive racial bias", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.614502+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 42, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Emotional Appeal", "slug": "emotional-appeal", "status": "active", "bestFor": null, "canonId": "019dd956-bde9-7668-8be7-0097678fec59", "version": 1, "bodyDocId": null, "description": "Connecting to audience feelings rather than just logic", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-dc0843dbf1ed", "evidence": "Racial-bias example designed for moral charge.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 42, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 43, "from": 40, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5930175dc2d9", "evidence": "Big picture (fairness) -> specific examples (Turkish gender, racial) -> 5-type allocation bias table.", "position": 6, "objective": "Diagnose ML bias by isolating its sources", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 43, "type": "analysis", "title": "The slide uses a matrix to categorize different types of algorithmic bias observed in various tech products.", "function": "diagnose", "rawType": "diagnosis", "block": null, "metadata": { "slideType": "diagnosis", "slideTypeCanon": { "name": "Diagnosis", "slug": "diagnosis", "status": "active", "canonId": "019de52d-0cd7-7212-898f-0464041ff8d8", "version": 1, "description": null }, "function": "diagnose", "functionCanon": { "name": "Diagnose", "slug": "diagnose", "status": "active", "canonId": "019de52d-1657-7223-acc5-7437d878c850", "version": 1, "description": null }, "density": "balanced", "densityScore": 27, "componentCount": 4, "textChars": 432, "nDataPoints": 0, "notes": "The slide uses a matrix to categorize different types of algorithmic bias observed in various tech products.", "elementsJson": [ "headline_text", "paragraph", "data_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.577+00", "seconds": 4.599, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/43", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-43", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Bias typically stems from training data that fails to appropriately represent diversity or encodes biased labels", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-003291bf899c", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "Bias typically stems from training data that fails to appropriately represent diversity or encodes biased labels", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.577+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6607a723-7adf-4eb7-af9e-00ef9f695ef2", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.68, "x": 0.16, "y": 0.345 }, "kind": "table", "text": "Matrix of 7 AI bias examples mapped against 5 bias categories (denigration, stereotype, recognition, under-representation, ex-nomination).", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.577+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7bb36ba9-ec9c-4d9e-a01c-d7362ad5b8f4", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "ML models have 5 types of allocation bias that stem from training data", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.577+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "43f2d671-a1df-4179-8f4e-cf8841836b7d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "MECE Principle", "slug": "mece-principle", "status": "active", "bestFor": null, "canonId": "019dd956-7c60-71a8-8758-c79d914489c9", "version": 1, "bodyDocId": null, "description": "Mutually Exclusive, Collectively Exhaustive - ensuring complete, non-overlapping structure", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "block", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-e5bd648d5a3a", "evidence": "5-column matrix presents bias categories as MECE table.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "2x2 matrix", "slug": "matrix-2x2", "status": "active", "bestFor": null, "canonId": "019dd9e1-50ab-72d1-8afb-3ab0a38e32a8", "version": 1, "bodyDocId": null, "description": "BCG matrix, importance/urgency, any axes-and-quadrants frame.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "71c8db6e-2e62-4890-9fbf-8c23ab36054b", "evidence": "Matrix of 7 AI bias examples mapped against 5 bias categories", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:36.681189+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-e27c214e7f2d", "evidence": "Title 'ML models have 5 types of allocation bias that stem from training data'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [ { "matchId": "019dd95a-1ca5-70bb-bba0-69200c8b5702", "evidence": "Title declares 5-type taxonomy; columns: denigration/stereotype/recognition/under-representation/ex-nomination.", "framework": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:34.826634+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "frameworkName": "5 Types of Allocation Bias Matrix" }, { "matchId": "f5ba34a6-11e2-4604-a90c-a3c512b9069f", "evidence": "The slide uses a grid to categorize examples against defined bias types.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:43:59.577+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 43, "frameworkName": "matrix-nxn" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 43, "from": 40, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5930175dc2d9", "evidence": "Big picture (fairness) -> specific examples (Turkish gender, racial) -> 5-type allocation bias table.", "position": 6, "objective": "Diagnose ML bias by isolating its sources", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 44, "type": "evidence", "title": "The comic is a reference to the 'AI-Box Experiment' thought experiment.", "function": "illustrate_case", "rawType": "filler", "block": null, "metadata": { "slideType": "filler", "slideTypeCanon": { "name": "Filler", "slug": "filler", "status": "active", "canonId": "019de52d-0829-76f9-9abe-4c11ee7bf4ab", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "sparse", "densityScore": 20, "componentCount": 4, "textChars": 283, "nDataPoints": 0, "notes": "The comic is a reference to the 'AI-Box Experiment' thought experiment.", "elementsJson": [ "screenshot" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.015+00", "seconds": 1.945, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/44", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-44", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Many ML models, especially deep learning models, are often complex “black boxes”", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-ed6cd367d9ff", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.5, "x": 0.25, "y": 0.3 }, "kind": "image", "text": "Comic strip depicting the AI-Box experiment", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.015+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4f027071-7df7-41ad-ac55-1a5107de0253", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.22 }, "kind": "list", "text": "Many ML models, especially deep learning models, are often complex “black boxes”", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.015+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3c16d659-7e47-4b1c-ba86-29f5eac1a840", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Like all software, ML models need to be debugged, but understanding them is hard", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.015+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a473ad7f-6b11-4a10-acd1-5acb83a64f26", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Curiosity Gap", "slug": "curiosity-gap", "status": "active", "bestFor": null, "canonId": "019dd956-9703-74ef-8b85-96dfd8525025", "version": 1, "bodyDocId": null, "description": "People seek missing information - create open questions that demand resolution", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "block", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-eea570fe4a8e", "evidence": "Frames problem ('understanding them is hard') to set up later reveal.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 44, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-e9d49e4c82c6", "evidence": "Callout 'black boxes' metaphor for opaque models.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 44, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 50, "from": 44, "loop": { "name": "10_iceberg", "slug": "10-iceberg", "status": "active", "bestFor": "Consulting, complex problem solving, organizational change", "canonId": "019dd956-69f4-738c-852a-643798da0899", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause", "description": "Reveal that the visible problem is merely a symptom of a deeper root cause", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5ed4c9b352e3", "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 45, "type": "appendix", "title": "The slide uses a visual example of a dog playing a guitar to demonstrate how explainability techniques (like saliency maps or feature attribution) reveal what a", "function": "illustrate_case", "rawType": "other", "block": null, "metadata": { "slideType": "other", "slideTypeCanon": { "name": "Other (fallback)", "slug": "other", "status": "active", "canonId": "019de52d-0af5-71ed-83de-d51ae245fc43", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 51, "componentCount": 5, "textChars": 835, "nDataPoints": 3, "notes": "The slide uses a visual example of a dog playing a guitar to demonstrate how explainability techniques (like saliency maps or feature attribution) reveal what a model is 'looking at' to make a prediction.", "elementsJson": [ "headline_text", "paragraph", "photo", "footnote" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.715+00", "seconds": 2.502, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/45", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-45", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "In computer vision, a model can show us which pixels it used to infer a specific label (e.g. which pixels = “dog”) This way, we understand that the model has “learned” properly vs. predicted the right label for the wrong reason.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-4d30fe8a423e", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.8, "x": 0.1, "y": 0.35 }, "kind": "image", "text": "Four-part image showing original dog-guitar photo and three segmented explanations.", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.715+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f9ca01da-1d3d-4cf6-87e9-39d573df4591", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "In computer vision, a model can show us which pixels it used to infer a specific label (e.g. which pixels = “dog”) This way, we understand that the model has “learned” properly vs. predicted the right label for the wrong reason.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.715+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "03e99f06-36ef-440f-b5f0-67903ff84420", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.8, "x": 0.1, "y": 0.78 }, "kind": "source-note", "text": "Figure 4: Explaining an image classification prediction made by Google’s Inception neural network. The top 3 classes predicted are “Electric Guitar” (p = 0.32), “Acoustic guitar” (p = 0.24) and “Labrador” (p = 0.21)", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.715+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9cee3e5e-83cd-4b46-afff-ce1e0bc3a07d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Explainability helps validate that ML models perform well for the “right” reasons", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.715+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "70f4aa0b-1607-4f86-abed-82467b690068", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-f307bd88bbaf", "evidence": "Title 'Explainability helps validate that ML models perform well for the right reasons'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 45, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Visual Anchors", "slug": "visual-anchors", "status": "active", "bestFor": null, "canonId": "019dd956-c264-7357-a791-89fa5f02a400", "version": 1, "bodyDocId": null, "description": "Design elements that guide users' eyes through the content intentionally", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "a60d6033-b99e-40e4-a048-9da3377ae645", "evidence": "Figure 4: Explaining an image classification prediction made by Google’s Inception neural network", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:36.785603+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 45, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 50, "from": 44, "loop": { "name": "10_iceberg", "slug": "10-iceberg", "status": "active", "bestFor": "Consulting, complex problem solving, organizational change", "canonId": "019dd956-69f4-738c-852a-643798da0899", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause", "description": "Reveal that the visible problem is merely a symptom of a deeper root cause", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5ed4c9b352e3", "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 46, "type": "evidence", "title": "The slide showcases four examples of AI-generated explanations for image-based questions, using heatmaps to highlight visual evidence.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 36, "componentCount": 4, "textChars": 692, "nDataPoints": 0, "notes": "The slide showcases four examples of AI-generated explanations for image-based questions, using heatmaps to highlight visual evidence.", "elementsJson": [ "headline_text", "paragraph", "photo", "heatmap" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:58.263+00", "seconds": 2.786, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/46", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-46", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Joint textual rationale generation and attention visualization provides deeper insight into decisions", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-9d4275e1a8e7", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.75, "x": 0.13, "y": 0.45 }, "kind": "image", "text": "Four examples of PJ-X model outputs showing images with heatmaps and corresponding textual justifications.", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.263+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bd5a9dc2-d04e-4d61-924e-549157928122", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "Joint textual rationale generation and attention visualization provides deeper insight into decisions. For a given question and an image, the Pointing and Justification Explanation (PJ-X) model predicts the answer and multimodal explanations which both point to the visual evidence for a decision and provide textual justifications. Multimodal explanations results in better visual and textual explanations.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.263+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e3224ed8-e32f-4370-94c1-84ac5e7e634a", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Next step: Justifying decisions in plain language and pointing to the evidence", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:58.263+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d8f4d793-4568-489b-afa3-6de50c9dad75", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-f4bad5980e99", "evidence": "Title 'Next step: Justifying decisions in plain language'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 46, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 50, "from": 44, "loop": { "name": "10_iceberg", "slug": "10-iceberg", "status": "active", "bestFor": "Consulting, complex problem solving, organizational change", "canonId": "019dd956-69f4-738c-852a-643798da0899", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause", "description": "Reveal that the visible problem is merely a symptom of a deeper root cause", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5ed4c9b352e3", "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 47, "type": "analysis", "title": "The chart is a lollipop chart showing feature importance.", "function": "diagnose", "rawType": "diagnosis", "block": null, "metadata": { "slideType": "diagnosis", "slideTypeCanon": { "name": "Diagnosis", "slug": "diagnosis", "status": "active", "canonId": "019de52d-0cd7-7212-898f-0464041ff8d8", "version": 1, "description": null }, "function": "diagnose", "functionCanon": { "name": "Diagnose", "slug": "diagnose", "status": "active", "canonId": "019de52d-1657-7223-acc5-7437d878c850", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 6, "textChars": 561, "nDataPoints": 14, "notes": "The chart is a lollipop chart showing feature importance.", "elementsJson": [ "headline_text", "paragraph", "bar_chart_horizontal" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:03.222+00", "seconds": 7.701, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/47", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-47", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "The more a feature is important, the greater the model's prediction error as a result of the feature value change.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-32a93247f212", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.3, "y": 0.4 }, "kind": "chart", "text": "Lollipop chart showing feature importance for cervical cancer prediction", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Lollipop chart", "slug": "lollipop", "status": "active", "canonId": "019de52d-025d-72ca-b75a-a9cd5f6bedd0", "version": 1, "description": null }, "framework": null }, "subkind": "lollipop", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.222+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d3f8caef-6e01-496f-a8f1-abbe6f4d2ed7", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Feature Importance", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-34ae1f3b53c9", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "We can alter the value of a particular model feature to see how the overall model's prediction error changes. The more a feature is important, the greater the model's prediction error as a result of the feature value change.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.222+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4ee67ffc-4d7d-44f6-9e4c-4e0b524e5663", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Understanding feature importance gives us high level insight into a model's behavior", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.222+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c76f5f5f-deed-40ad-921d-2f08db1fba06", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.34, "x": 0.33, "y": 0.365 }, "kind": "title", "text": "Features important for predicting cervical cancer", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Subtitle", "slug": "subtitle", "status": "active", "canonId": "019de52c-fb09-739d-900c-815b0d9ac526", "version": 1, "description": null }, "framework": null }, "subkind": "subtitle", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.222+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a3a320e7-4255-4507-8c2a-c9e6061716cf", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Waterfall chart", "slug": "waterfall-chart", "status": "active", "bestFor": null, "canonId": "e2799da3-725f-44a9-a093-570f1f388e64", "version": 1, "bodyDocId": null, "description": null, "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "1403d9ce-3e85-4c5c-b98c-df7956e22af0", "evidence": "Lollipop chart showing feature importance for cervical cancer prediction", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:36.885106+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 47, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 50, "from": 44, "loop": { "name": "10_iceberg", "slug": "10-iceberg", "status": "active", "bestFor": "Consulting, complex problem solving, organizational change", "canonId": "019dd956-69f4-738c-852a-643798da0899", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause", "description": "Reveal that the visible problem is merely a symptom of a deeper root cause", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5ed4c9b352e3", "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 48, "type": "evidence", "title": "Shows two examples: a panda misclassified as a gibbon, and a macaw misclassified as a bookcase.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 39, "componentCount": 5, "textChars": 450, "nDataPoints": 4, "notes": "Shows two examples: a panda misclassified as a gibbon, and a macaw misclassified as a bookcase.", "elementsJson": [ "headline_text", "bullet_list", "paragraph", "infographic" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.905+00", "seconds": 2.309, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/48", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-48", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Adversarial examples cause computer vision models to make glaring mistakes!", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-f313384d1dab", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.5, "x": 0.25, "y": 0.35 }, "kind": "image", "text": "Visual representation of the adversarial attack process (Image + Noise = Misclassified Image)", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Infographic", "slug": "infographic", "status": "active", "canonId": "019de52c-fe26-750c-9d7d-9f34e21324e8", "version": 1, "description": null }, "framework": null }, "subkind": "infographic", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.905+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "260cf694-880d-47dc-bb56-cd581cb398ce", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "list", "text": "Adversarial examples cause computer vision models to make glaring mistakes! An unnoticeable universal noise filter applied to an image of a panda makes the model think it sees a gibbon.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.905+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4bab8570-899e-4cc6-ae3f-8c596ba81200", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "model confidence: 99.3%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-f5c4010c814a", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Imperceptible changes to data can alter a deep learning model's prediction", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.905+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d46dad8d-d771-4c45-8916-a94cd14f2194", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8eca-f84ec4d232d2", "evidence": "Title 'Imperceptible changes to data can alter a deep learning model's prediction' with 99.3% stat.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 48, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Unexpected Pattern", "slug": "unexpected-pattern", "status": "active", "bestFor": null, "canonId": "019dd956-bc0f-71af-8d32-3c1a2da38c40", "version": 1, "bodyDocId": null, "description": "Breaking a pattern to create surprise and attention", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8eca-fd335a7f063a", "evidence": "Adversarial example breaks expectation that small change = small effect.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 48, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 50, "from": 44, "loop": { "name": "10_iceberg", "slug": "10-iceberg", "status": "active", "bestFor": "Consulting, complex problem solving, organizational change", "canonId": "019dd956-69f4-738c-852a-643798da0899", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause", "description": "Reveal that the visible problem is merely a symptom of a deeper root cause", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5ed4c9b352e3", "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 49, "type": "evidence", "title": "The slide demonstrates adversarial machine learning, specifically targeted adversarial patches.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 31, "componentCount": 4, "textChars": 538, "nDataPoints": 0, "notes": "The slide demonstrates adversarial machine learning, specifically targeted adversarial patches.", "elementsJson": [ "headline_text", "paragraph", "photo", "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:57.915+00", "seconds": 2.136, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/49", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-49", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "The “toaster” patch maximally excites the computer vision model so that it always sees a toaster even when there is no “real” toaster in view.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:20+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-f1c39a050049", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.5, "x": 0.25, "y": 0.35 }, "kind": "image", "text": "Diagram showing a banana, a toaster sticker, and the resulting classification change.", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.915+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d74e4223-0afb-4571-8543-0642f968545d", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "A method for creating universal, robust, targeted adversarial image patches in the real world. The “toaster” patch maximally excites the computer vision model so that it always sees a toaster even when there is no “real” toaster in view.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.915+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7145496b-29ed-4538-9bd3-15513db11343", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Imperceptible changes to data can alter a deep learning model's prediction", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:57.915+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ee0350cc-1d30-4485-a01f-c2cf6d1101d1", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-01602bc9539b", "evidence": "Same title; toaster patch example.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 49, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-07a15920615d", "evidence": "Toaster patch is concrete, vivid example.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 49, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 50, "from": 44, "loop": { "name": "10_iceberg", "slug": "10-iceberg", "status": "active", "bestFor": "Consulting, complex problem solving, organizational change", "canonId": "019dd956-69f4-738c-852a-643798da0899", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause", "description": "Reveal that the visible problem is merely a symptom of a deeper root cause", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5ed4c9b352e3", "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 50, "type": "evidence", "title": "The slide uses a visual comparison to demonstrate a security vulnerability in AI.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 38, "componentCount": 5, "textChars": 545, "nDataPoints": 0, "notes": "The slide uses a visual comparison to demonstrate a security vulnerability in AI.", "elementsJson": [ "headline_text", "paragraph", "screenshot" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:01.923+00", "seconds": 5.997, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/50", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-50", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "A vision system that previously detected pedestrians at a zebra crossing is no longer able to “see” them.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-baa9f594ce0a", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.85, "x": 0.08, "y": 0.4 }, "kind": "image", "text": "Comparison of original image, network prediction, and adversarial target", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.923+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "aee3f9d4-0d47-4b9a-9f9c-ffc5436e193e", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "A vision system that previously detected pedestrians at a zebra crossing is no longer able to “see” them. This poses obvious security concerns when autonomous vehicles make it onto public roads.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.923+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a1da620d-bdcd-42b2-b6fc-aa85163bc39c", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.2, "y": 0.78 }, "kind": "source-note", "text": "Figure 2. Illustration of an adversary generating a dynamic target segmentation for hiding pedestrians.", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.923+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "90d840dc-5101-444d-9acb-58fd125153d8", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Adversarial attacks present serious safety challenges in the real world", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.923+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bc3d247d-5de4-4c71-a1b2-ea05daeaf75e", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-0ba566bd457e", "evidence": "Title 'Adversarial attacks present serious safety challenges in the real world'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 50, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Emotional Appeal", "slug": "emotional-appeal", "status": "active", "bestFor": null, "canonId": "019dd956-bde9-7668-8be7-0097678fec59", "version": 1, "bodyDocId": null, "description": "Connecting to audience feelings rather than just logic", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-0c27bc0c0126", "evidence": "Pedestrian-at-zebra-crossing made invisible — visceral.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 50, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 50, "from": 44, "loop": { "name": "10_iceberg", "slug": "10-iceberg", "status": "active", "bestFor": "Consulting, complex problem solving, organizational change", "canonId": "019dd956-69f4-738c-852a-643798da0899", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Symptom (Visible) -> The System (Hidden) -> The Root Cause", "description": "Reveal that the visible problem is merely a symptom of a deeper root cause", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-5ed4c9b352e3", "evidence": "Symptom (black box) -> hidden system (feature importance) -> root issues (adversarial attacks fool models in real world).", "position": 7, "objective": "Surface hidden risks beneath model accuracy", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 51, "type": "analysis", "title": "The slide uses a chronological sequence of technical diagrams to show the progression of AI model complexity.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 56, "componentCount": 7, "textChars": 587, "nDataPoints": 0, "notes": "The slide uses a chronological sequence of technical diagrams to show the progression of AI model complexity.", "elementsJson": [ "headline_text", "paragraph", "process_diagram", "screenshot" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.258+00", "seconds": 3.195, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/51", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-51", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": ">5 years of research into convolutional neural network architectures for computer vision applications Involving many researchers, institutions and proposed structural and computational innovations.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80f-66b36094bef2", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.2, "x": 0.75, "y": 0.35 }, "kind": "image", "text": "Inception v4 architecture diagram", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.258+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "34208358-1405-407f-ba0b-48c02903fb84", "parentComponentId": null }, { "bbox": { "h": 0.25, "w": 0.45, "x": 0.05, "y": 0.35 }, "kind": "image", "text": "AlexNet architecture diagram", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.258+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4675d7b4-3264-4806-91e2-520f703300f7", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.15, "x": 0.55, "y": 0.35 }, "kind": "image", "text": "ResNet architecture diagram", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.258+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c6d6273d-00f1-4272-ab98-0e1ecc6d012d", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.45, "x": 0.05, "y": 0.65 }, "kind": "image", "text": "VGGNet architecture diagram", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.258+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d1f6ea3f-cf15-474c-9209-e853505dc36f", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.85, "x": 0.05, "y": 0.22 }, "kind": "list", "text": ">5 years of research into convolutional neural network architectures for computer vision applications. Involving many researchers, institutions and proposed structural and computational innovations.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.258+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a159ae1e-cfd8-4e25-9d9d-e3f77f3b1514", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Improving deep learning model architecture requires iterative experimentation", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.258+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1e14afa8-8372-43a3-830b-d9bbddc5a1fb", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [ { "matchId": "08e6e216-27de-40a6-944e-f08ffa5929a6", "evidence": "Chronological presentation of model architectures from 2013 to 2016", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:43:59.258+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 51, "frameworkName": "timeline" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 55, "from": 51, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-619c71c15583", "evidence": "Architecture iteration -> AutoML automating engineers -> federated learning -> consensus mechanism.", "position": 8, "objective": "Catalogue frontier research trends (AutoML, federated)", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 52, "type": "analysis", "title": "The slide shows a bar chart of model accuracy over time (2013-2016) and a bubble chart showing accuracy vs. operations (complexity).", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 50, "componentCount": 6, "textChars": 431, "nDataPoints": 28, "notes": "The slide shows a bar chart of model accuracy over time (2013-2016) and a bubble chart showing accuracy vs. operations (complexity).", "elementsJson": [ "headline_text", "bar_chart_vertical", "bubble_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.403+00", "seconds": 2.776, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/52", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-52", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Leading to significant reductions in Top-1% accuracy on Large Scale Visual Recognition Challenge (ILSVRC)", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-d512218bc9ed", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.45, "x": 0.02, "y": 0.35 }, "kind": "chart", "text": "Bar chart showing Top-1 accuracy of various models from 2013 to 2016", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.403+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "32ff276c-fc0a-40f0-9437-953e274b3359", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.45, "x": 0.52, "y": 0.35 }, "kind": "chart", "text": "Bubble chart showing Top-1 accuracy vs Operations (G-Ops)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Bubble chart", "slug": "bubble", "status": "active", "canonId": "019de52d-00a4-756f-8978-8d5e48905dbc", "version": 1, "description": null }, "framework": null }, "subkind": "bubble", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.403+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "eb04d203-04cb-4f11-931d-66b78ad3a713", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Top-1 accuracy: 80%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-d94626735c67", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.8, "x": 0.02, "y": 0.21 }, "kind": "title", "text": "Leading to significant reductions in Top-1% accuracy on Large Scale Visual Recognition Challenge (ILSVRC)", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.403+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "add89f5d-b923-43cd-9d80-2af81dc8a70e", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Improving deep learning model architecture requires iterative experimentation", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.403+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "72d34a64-9819-470d-81c1-635c8cf6400f", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-132694e4f723", "evidence": "Title with '80%' stat on accuracy reductions.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 52, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chartjunk Elimination", "slug": "chartjunk-elimination", "status": "active", "bestFor": null, "canonId": "019dd956-b49e-7453-ab01-b60f338eeda1", "version": 1, "bodyDocId": null, "description": "Remove decorative elements that don't convey data: 3D effects, shadows, unnecessary gridlines", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "d54e1dc4-78c3-4172-af83-ed386f0a8b71", "evidence": "Bar chart showing Top-1 accuracy of various models from 2013 to 2016", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.083383+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 52, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 55, "from": 51, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-619c71c15583", "evidence": "Architecture iteration -> AutoML automating engineers -> federated learning -> consensus mechanism.", "position": 8, "objective": "Catalogue frontier research trends (AutoML, federated)", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 53, "type": "evidence", "title": "The slide shows two neural network cell architectures and a scatter plot comparing accuracy vs. computational cost.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 56, "componentCount": 6, "textChars": 623, "nDataPoints": 15, "notes": "The slide shows two neural network cell architectures and a scatter plot comparing accuracy vs. computational cost.", "elementsJson": [ "headline_text", "action_title", "paragraph", "process_diagram", "scatter_plot", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.332+00", "seconds": 2.458, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/53", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-53", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "AutoML traversed the architecture search space to find two new cell designs that could be integrated into a final model that outperformed all existing human-crafted models.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-c8813b797232", "parentComponentId": null }, { "bbox": { "h": 0.48, "w": 0.4, "x": 0.55, "y": 0.42 }, "kind": "chart", "text": "Accuracy vs. # Mult-Add operations", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Scatter plot", "slug": "scatter", "status": "active", "canonId": "019de52d-007d-72ec-8689-c79fb78e21a2", "version": 1, "description": null }, "framework": null }, "subkind": "scatter", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.332+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ac35b12d-d85d-4923-bd3a-1a5ee8fe20a6", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.45, "x": 0.05, "y": 0.45 }, "kind": "image", "text": "Normal and Reduction Cell diagrams", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Infographic", "slug": "infographic", "status": "active", "canonId": "019de52c-fe26-750c-9d7d-9f34e21324e8", "version": 1, "description": null }, "framework": null }, "subkind": "infographic", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.332+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2ab3e729-f3d3-41f4-9e5e-79c4a781943d", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "accuracy (precision @1)", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:15+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a80e-cd3c22173e6f", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "Google's AutoML automatically discovers the best model architecture for a computer vision task. AutoML traversed the architecture search space to find two new cell designs (Normal and Reduction, left figueres) that could be integrated into a final model (NASNet, right graph) that outperformed all existing human-crafted models.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.332+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fc803c48-e9f7-4ea0-9a97-d2c78e1edb89", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.35, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "AI to automate away AI engineers", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.332+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5e55633b-5922-46c2-9476-f77769d125ad", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-16354f7fb19c", "evidence": "Title 'AI to automate away AI engineers'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 53, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Unexpected Pattern", "slug": "unexpected-pattern", "status": "active", "bestFor": null, "canonId": "019dd956-bc0f-71af-8d32-3c1a2da38c40", "version": 1, "bodyDocId": null, "description": "Breaking a pattern to create surprise and attention", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-1bf7068e6cbc", "evidence": "Recursive twist — AI replacing AI engineers — surprises.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 53, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Waterfall chart", "slug": "waterfall-chart", "status": "active", "bestFor": null, "canonId": "e2799da3-725f-44a9-a093-570f1f388e64", "version": 1, "bodyDocId": null, "description": null, "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "38105218-80ce-4f5c-9713-9171e92b6162", "evidence": "chart/scatter: Accuracy vs. # Mult-Add operations", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:33.279949+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 53, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 55, "from": 51, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-619c71c15583", "evidence": "Architecture iteration -> AutoML automating engineers -> federated learning -> consensus mechanism.", "position": 8, "objective": "Catalogue frontier research trends (AutoML, federated)", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 54, "type": "evidence", "title": "The slide explains the technical mechanism of federated learning using OpenMined's framework.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 47, "componentCount": 5, "textChars": 903, "nDataPoints": 0, "notes": "The slide explains the technical mechanism of federated learning using OpenMined's framework.", "elementsJson": [ "paragraph", "process_diagram" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.513+00", "seconds": 2.632, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/54", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-54", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Large technology companies centralise immense amounts of user data. The community is now starting to push back by creating tools to decentralise data ownership.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-983fece6dfdc", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.5, "x": 0.25, "y": 0.5 }, "kind": "diagram", "text": "Open Mined Architecture", "attrs": null, "canon": { "kind": { "name": "Diagram", "slug": "diagram", "status": "active", "canonId": "019de52c-f9f0-734a-81de-15382ff28c65", "version": 1, "description": "Schematic or relational diagram." }, "tool": null, "subkind": { "name": "Process flow", "slug": "process", "status": "active", "canonId": "019de52d-039c-718f-ad3a-efe9c4b3c838", "version": 1, "description": null }, "framework": null }, "subkind": "process", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.513+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fbdbaf4a-cac1-40ed-b567-964026804249", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.02, "y": 0.02 }, "kind": "other", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Other", "slug": "other", "status": "active", "canonId": "019de52c-fa90-743c-8bf6-20d1fbe82b6d", "version": 1, "description": "Fallback when no canonical kind applies." }, "tool": null, "subkind": { "name": "Unclassified", "slug": "unclassified", "status": "active", "canonId": "019de52d-05ca-77db-81fb-db7d4234bbc0", "version": 1, "description": null }, "framework": null }, "subkind": "unclassified", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.513+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "19e3566b-ed61-4afa-b315-5026af3e731d", "parentComponentId": null }, { "bbox": { "h": 0.25, "w": 0.96, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "OpenMined: Train a model on lots of individual user devices such that their data never leaves their devices. Large technology companies centralise immense amounts of user data. The community is now starting to push back by creating tools to decentralise data ownership. In OpenMined, an AI model itself is encrypted by it’s owner such that the user cannot steal it. User data stays locally on a user’s device and is accessed to update the model’s parameters. These parameter changes from multiple users are aggregated back to the model owner for updating.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.513+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b69ac95b-9dfc-4fb3-8455-03031387c678", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Distributed “federated” learning to decentralise data acquisition and model training", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.513+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b4b75050-cad4-4768-b2af-d646e167cdc8", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-1c66fe5f8199", "evidence": "Title 'Distributed federated learning to decentralise data acquisition'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 54, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 55, "from": 51, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-619c71c15583", "evidence": "Architecture iteration -> AutoML automating engineers -> federated learning -> consensus mechanism.", "position": 8, "objective": "Catalogue frontier research trends (AutoML, federated)", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 55, "type": "analysis", "title": "The slide uses a process flow diagram to explain a technical concept.", "function": "present_framework", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "present_framework", "functionCanon": { "name": "Present framework", "slug": "present_framework", "status": "active", "canonId": "019de52d-1403-7667-9100-d2ac6709f750", "version": 1, "description": null }, "density": "balanced", "densityScore": 38, "componentCount": 6, "textChars": 349, "nDataPoints": 0, "notes": "The slide uses a process flow diagram to explain a technical concept.", "elementsJson": [ "process_diagram", "callout_box" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.907+00", "seconds": 2.958, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/55", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-55", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.1, "w": 0.3, "x": 0.35, "y": 0.75 }, "kind": "callout", "text": "Many users' updates are aggregated together", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.907+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1786b4a2-b0b7-482e-96fc-c553fbdfad87", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.25, "x": 0.05, "y": 0.4 }, "kind": "callout", "text": "Your keyboard model is personalised locally based on your usage.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.907+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b049040a-1577-4aed-979f-7980820074f0", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.25, "x": 0.7, "y": 0.4 }, "kind": "callout", "text": "Consensus change is agreed and shared to the core model.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.907+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f88c451c-2bb6-4db4-aca8-a898a91f1317", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.5, "x": 0.25, "y": 0.35 }, "kind": "diagram", "text": "Federated learning process flow", "attrs": null, "canon": { "kind": { "name": "Diagram", "slug": "diagram", "status": "active", "canonId": "019de52c-f9f0-734a-81de-15382ff28c65", "version": 1, "description": "Schematic or relational diagram." }, "tool": null, "subkind": { "name": "Process flow", "slug": "process", "status": "active", "canonId": "019de52d-039c-718f-ad3a-efe9c4b3c838", "version": 1, "description": null }, "framework": null }, "subkind": "process", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.907+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dea23cee-916a-419f-af48-b2a2e8707e0d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.22 }, "kind": "list", "text": "Google uses federated learning to train its mobile keyboard prediction models, Gboard", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.907+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f6f5db99-36a9-47f4-ba1f-13530c5602fd", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Federated learning to decentralise data acquisition and model training", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.907+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cbf086cb-1044-4f45-bc6e-57ffb3b8253f", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-22b21615e6d2", "evidence": "Title repeats federated framing; process diagram.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 55, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Process flow", "slug": "process-flow", "status": "active", "bestFor": null, "canonId": "019dd9e1-4dca-76bc-8135-401836a1c4c0", "version": 1, "bodyDocId": null, "description": "Workflow, steps, sequential diagram.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "4712ee03-225f-433d-8595-d348d006c361", "evidence": "diagram/process: Federated learning process flow", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:33.381408+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 55, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [ { "matchId": "68bb0cfe-265f-4656-b66e-a8a100f5dfa3", "evidence": "The slide depicts a step-by-step technical workflow.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:43:59.907+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 55, "frameworkName": "process" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 55, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 10, "name": "First Layer", "beatId": "019dd95a-0682-776c-8e34-be65c7627fe7", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "First Layer", "slug": "onion-first-layer", "status": "active", "canonId": "019dd9b8-0bd1-75b2-9773-18ea7796621c", "version": 1, "description": null }, "evidence": "Hardware, vision, RL, bias - technical research layer.", "position": 2, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 55, "from": 51, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-619c71c15583", "evidence": "Architecture iteration -> AutoML automating engineers -> federated learning -> consensus mechanism.", "position": 8, "objective": "Catalogue frontier research trends (AutoML, federated)", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 56, "type": "setup", "function": "transition", "rawType": "section_divider", "block": null, "metadata": { "slideType": "section_divider", "slideTypeCanon": { "name": "Section divider", "slug": "section_divider", "status": "active", "canonId": "019de52d-066a-7402-9554-6ae504999041", "version": 1, "description": null }, "function": "transition", "functionCanon": { "name": "Transition", "slug": "transition", "status": "active", "canonId": "019de52d-1365-7149-92aa-66df859e75fd", "version": 1, "description": null }, "density": "sparse", "densityScore": 14, "componentCount": 4, "textChars": 122, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.273+00", "seconds": 2.181, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/56", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-56", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.03, "w": 0.08, "x": 0.9, "y": 0.95 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.273+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3da8b580-7ff4-4cc6-9499-253a3e02292c", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.06, "x": 0.92, "y": 0.02 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.273+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "95b947e1-ddb3-463d-b219-cd53d964d592", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.96, "x": 0.02, "y": 0.02 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.273+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a82b7145-d0e4-4a4e-8de6-ea71f3ccf69c", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.24, "x": 0.38, "y": 0.45 }, "kind": "title", "text": "Section 2: Talent", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.273+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "114be9aa-3a90-4884-b1cc-a3d2798d2b20", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "f1dd0563-6012-4bb8-8362-c92362705192", "evidence": "type: section_divider", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:33.447831+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 56, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 57, "type": "resolution", "title": "The slide uses a choropleth map to represent talent density, supplemented by callouts for specific regional insights.", "function": "size_opportunity", "rawType": "geographic_map", "block": null, "metadata": { "slideType": "geographic_map", "slideTypeCanon": { "name": "Geographic map", "slug": "geographic_map", "status": "active", "canonId": "019de52d-09de-7153-8845-584c1a9d0c83", "version": 1, "description": null }, "function": "size_opportunity", "functionCanon": { "name": "Size opportunity", "slug": "size_opportunity", "status": "active", "canonId": "019de52d-147b-7568-8828-a74a27697618", "version": 1, "description": null }, "density": "dense", "densityScore": 62, "componentCount": 7, "textChars": 617, "nDataPoints": 6, "notes": "The slide uses a choropleth map to represent talent density, supplemented by callouts for specific regional insights.", "elementsJson": [ "headline_text", "action_title", "map", "callout_box", "other" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.706+00", "seconds": 5.597, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/57", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-57", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.3, "w": 0.25, "x": 0.7, "y": 0.4 }, "kind": "callout", "text": "Only 3k candidates in the world are available for work; Asian markets are catching up to the West; China has surpassed the USA in quantity of peer-reviewed publications per year", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.706+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9bf140ca-ebec-40f2-b6c0-4f2bd18f3271", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.25, "x": 0.05, "y": 0.4 }, "kind": "callout", "text": "20k Worldwide Researchers and Engineers (LinkedIn); 5k Worldwide High-Level Researchers (Conferences); USA serves as a hub for a largely global phenomenon, and has 10k open positions", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.706+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a73d52bb-23cc-454a-836f-2978c3d765ac", "parentComponentId": null }, { "bbox": null, "kind": "callout", "text": "USA serves as a hub for a largely global phenomenon, and has 10k open positions.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-3c7a5148464b", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.4, "x": 0.15, "y": 0.88 }, "kind": "chart", "text": "Total AI Talent legend", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.706+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6a355db5-231f-4f99-aaf4-c6769d7d23e2", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.6, "x": 0.2, "y": 0.25 }, "kind": "image", "text": "Global AI Talent Pool Heat Map", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Map", "slug": "map", "status": "active", "canonId": "019de52c-fe4d-77bd-b0a4-d72d58b9e5a4", "version": 1, "description": null }, "framework": null }, "subkind": "map", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.706+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bea2052e-5596-49e2-8d1a-7417634ec81e", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "PhD educated AI researchers and engineers: 22,000", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-41f41e431c9f", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Supply: Element AI estimates 22,000 PhD educated AI researchers and engineers", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.706+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b519ae89-4876-4e5f-94fb-c4d2175bc022", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-28dffa327633", "evidence": "First slide of talent-concentration pattern_hunter loop.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 57, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-26078fa4b136", "evidence": "Title 'Element AI estimates 22,000 PhD educated AI researchers'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 57, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 58, "type": "resolution", "title": "The slide uses a color-coded choropleth map to visualize talent density.", "function": "size_opportunity", "rawType": "geographic_map", "block": null, "metadata": { "slideType": "geographic_map", "slideTypeCanon": { "name": "Geographic map", "slug": "geographic_map", "status": "active", "canonId": "019de52d-09de-7153-8845-584c1a9d0c83", "version": 1, "description": null }, "function": "size_opportunity", "functionCanon": { "name": "Size opportunity", "slug": "size_opportunity", "status": "active", "canonId": "019de52d-147b-7568-8828-a74a27697618", "version": 1, "description": null }, "density": "balanced", "densityScore": 38, "componentCount": 6, "textChars": 267, "nDataPoints": 1, "notes": "The slide uses a color-coded choropleth map to visualize talent density.", "elementsJson": [ "map" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.589+00", "seconds": 2.448, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/58", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-58", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.6, "w": 0.7, "x": 0.15, "y": 0.25 }, "kind": "image", "text": "World map with color-coded regions representing researcher density", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Map", "slug": "map", "status": "active", "canonId": "019de52c-fe4d-77bd-b0a4-d72d58b9e5a4", "version": 1, "description": null }, "framework": null }, "subkind": "map", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1ee5e3eb-107c-4de5-85ef-0e3c9484b758", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "high-level researchers: 5,000", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-85c3f0a3365f", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.7, "x": 0.02, "y": 0.02 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "624d922c-fffc-4034-a8ba-0977c85b4f70", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.88, "y": 0.02 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b86749d7-ae78-4006-bbe6-95a3c7ed81a8", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.82, "y": 0.92 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e4f30781-00a9-41ea-bdf3-9a418e65e049", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Supply: Element AI estimates 5,000 high-level researchers worldwide", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "51c85894-0df3-4131-b417-97eb1980f71b", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-2dc3a646c539", "evidence": "Title '5,000 high-level researchers worldwide'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 58, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 59, "type": "evidence", "title": "The map highlights the US, Canada, UK, and China in red, with flow lines connecting the US to the other three countries.", "function": "illustrate_case", "rawType": "geographic_map", "block": null, "metadata": { "slideType": "geographic_map", "slideTypeCanon": { "name": "Geographic map", "slug": "geographic_map", "status": "active", "canonId": "019de52d-09de-7153-8845-584c1a9d0c83", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "sparse", "densityScore": 17, "componentCount": 4, "textChars": 204, "nDataPoints": 0, "notes": "The map highlights the US, Canada, UK, and China in red, with flow lines connecting the US to the other three countries.", "elementsJson": [ "map" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:01.668+00", "seconds": 4.447, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/59", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-59", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Largest inbound and outbound talent exchanges with the US", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-be4fb685ee55", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.6, "x": 0.2, "y": 0.25 }, "kind": "image", "text": "World map with flow lines", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Map", "slug": "map", "status": "active", "canonId": "019de52c-fe4d-77bd-b0a4-d72d58b9e5a4", "version": 1, "description": null }, "framework": null }, "subkind": "map", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.668+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "62b4e461-1e7a-43b7-9e22-c8244c13555a", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.28, "y": 0.94 }, "kind": "subtitle", "text": "Largest inbound and outbound talent exchanges with the US", "attrs": null, "canon": { "kind": { "name": "Subtitle", "slug": "subtitle", "status": "active", "canonId": "82d821ee-2793-42e7-8f16-c6131385a3ce", "version": 1, "description": "Header-zone supporting text near the title (= ex body.description with bbox)." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.668+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3b735e9c-5391-4eeb-9351-41090a60db03", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Talent concentration: America remains the hub for talent exchange", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.668+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "924a2ab5-7a8f-4dd2-84b5-ba2feb030ffa", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-3070433042ce", "evidence": "Title 'America remains the hub for talent exchange'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 59, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [ { "matchId": "8e39347e-fb1f-40b5-9d61-a67c12e37c04", "evidence": "Visual representation of geographic talent flows", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:44:01.668+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 59, "frameworkName": "map-spatial" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 60, "type": "setup", "title": "The chart uses house-shaped bars to represent headcount. Data points: Facebook (300), Tencent (400), Baidu (450), Google (1400), IBM (900), Microsoft (1000).", "function": "compare_peers", "rawType": "competitive_analysis", "block": null, "metadata": { "slideType": "competitive_analysis", "slideTypeCanon": { "name": "Competitive analysis", "slug": "competitive_analysis", "status": "active", "canonId": "019de52d-093e-776b-94c6-c1573f5f1a31", "version": 1, "description": null }, "function": "compare_peers", "functionCanon": { "name": "Compare peers", "slug": "compare_peers", "status": "active", "canonId": "019de52d-1453-72ba-a367-9ba3d34c1630", "version": 1, "description": null }, "density": "balanced", "densityScore": 40, "componentCount": 6, "textChars": 278, "nDataPoints": 6, "notes": "The chart uses house-shaped bars to represent headcount. Data points: Facebook (300), Tencent (400), Baidu (450), Google (1400), IBM (900), Microsoft (1000).", "elementsJson": [ "bar_chart_vertical", "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:03.587+00", "seconds": 6.328, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/60", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-60", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Google AI研发团队规模上千，全球居首", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-0ef466198921", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.6, "x": 0.2, "y": 0.3 }, "kind": "chart", "text": "Comparison of AI research team sizes by company", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.587+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0da1ace9-ec4c-4ef9-b101-312cd0e83fac", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "AI research headcount: 1400", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-110a05b2363b", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.8, "x": 0.02, "y": 0.02 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.587+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "83e2c545-7fff-4d9e-a19a-bc7d5bbc38fd", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.82, "y": 0.95 }, "kind": "source-note", "text": "state.ai 2018", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.587+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d41604a7-fc80-4cec-a8c1-94b98d75d7c8", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Talent concentration: Google is widely acknowledged as the leading employer of AI talent", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.587+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b4ccde6a-d5ad-4e04-914e-0c4d52d02c83", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-3521b6617347", "evidence": "Title 'Google is widely acknowledged as the leading employer of AI talent'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 60, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 61, "type": "analysis", "title": "The data is presented as a simple list of counts and institution names.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 29, "componentCount": 4, "textChars": 302, "nDataPoints": 15, "notes": "The data is presented as a simple list of counts and institution names.", "elementsJson": [ "data_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.492+00", "seconds": 2.217, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/61", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-61", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "metric", "text": "ICML paper mentions: 44", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-5e9e52ffe471", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.05, "y": 0.95 }, "kind": "source-note", "text": "Source: @karpathy", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.492+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5c55e2ff-b802-407b-a25d-2534e1561e78", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.66, "x": 0.17, "y": 0.24 }, "kind": "table", "text": "#mentions institution\n44 Google\n33 Microsoft\n32 CMU\n25 DeepMind\n23 MIT\n22 Berkeley\n22 Stanford\n16 Cambridge\n16 Princeton\n15 None\n14 Georgia Tech\n13 Oxford\n11 UT Austin\n10 Duke\n10 Facebook", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.492+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "91e87c97-d3d2-4700-a378-926bfeb4e17d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Talent concentration: 6.3% of 2017 ICML papers had a Google/DeepMind author", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.492+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "db587dd8-2e05-4b32-8aff-c12f6305474c", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-3843985ba57b", "evidence": "Title quantifies '6.3% of 2017 ICML papers had a Google/DeepMind author'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 61, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 62, "type": "analysis", "title": "The chart illustrates the increasing concentration of AI research talent at Google/DeepMind.", "function": "quantify_impact", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "quantify_impact", "functionCanon": { "name": "Quantify impact", "slug": "quantify_impact", "status": "active", "canonId": "019de52d-1630-720d-b57d-fa7f76bebd8d", "version": 1, "description": null }, "density": "balanced", "densityScore": 28, "componentCount": 5, "textChars": 215, "nDataPoints": 2, "notes": "The chart illustrates the increasing concentration of AI research talent at Google/DeepMind.", "elementsJson": [ "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:00.082+00", "seconds": 2.559, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/62", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-62", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.6, "w": 0.5, "x": 0.25, "y": 0.25 }, "kind": "chart", "text": "Percentage of ICML papers with a Google/DeepMind author", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.082+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5a3e5c2e-adc3-43af-8f86-b46c477acffc", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Percentage of ICML papers: 13%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:20+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-1d734039cc4e", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.85, "y": 0.95 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.082+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3e34e06d-5af6-4098-9d58-7b2946798d74", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.25, "x": 0.05, "y": 0.95 }, "kind": "source-note", "text": "Source: @karpathy, @dhruvguliani", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.082+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "04b3798b-c360-4f14-b206-c44691818f6b", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.12 }, "kind": "title", "text": "Talent concentration: Percentage of ICML papers with Google/DeepMind author doubles", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.082+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c5e735bb-f1fa-491a-949e-1b7c5d19f975", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-3ce963013302", "evidence": "Title 'Percentage of ICML papers with Google/DeepMind author doubles'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 62, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 63, "type": "analysis", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 38, "componentCount": 4, "textChars": 496, "nDataPoints": 20, "notes": null, "elementsJson": [ "data_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:03.056+00", "seconds": 5.4, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/63", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-63", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "metric", "text": "Number of NIPS papers: 60", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-58645fc88d32", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.2, "x": 0.05, "y": 0.95 }, "kind": "source-note", "text": "Source: @robbieallen", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.056+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "28cde9ef-c6d9-4352-b0b9-3b62016ced85", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.6, "x": 0.2, "y": 0.23 }, "kind": "table", "text": "1. google: 60 (8.8%)\n2. carnegie mellon university: 48 (7.1%)\n3. massachusetts institute of technology: 43 (6.3%)\n4. microsoft: 40 (5.9%)\n5. stanford university: 39 (5.7%)\n6. university of california, berkeley: 35 (5.2%)\n7. deepmind: 31 (4.6%)\n8. university of oxford: 22 (3.2%)\n9. university of illinois at urbana-champaign: 20 (2.9%)\n10. georgia institute of technology: 18 (2.7%)", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.056+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7d01d5da-e8f3-4f54-9cde-392d38fe8374", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Talent concentration: Google lead in contribution to 2017 NIPS papers", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.056+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cba8f7d8-ad92-4b5a-92b4-f93380577d5c", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-4211c677f508", "evidence": "Title 'Google lead in contribution to 2017 NIPS papers'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 63, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 64, "type": "analysis", "title": "The slide highlights the concentration of AI research talent in a few top-tier academic institutions and major tech corporations.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 32, "componentCount": 4, "textChars": 391, "nDataPoints": 11, "notes": "The slide highlights the concentration of AI research talent in a few top-tier academic institutions and major tech corporations.", "elementsJson": [ "numbered_list" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:43:59.954+00", "seconds": 2.141, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/64", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-64", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "metric", "text": "NIPS authorship count: 89", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:20+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-32c563139a94", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.05, "y": 0.95 }, "kind": "source-note", "text": "Source: @robbieallen", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.954+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4d9b3d45-d225-4125-ba5e-a871b36bf1ab", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.72, "x": 0.14, "y": 0.24 }, "kind": "table", "text": "1. carnegie mellon university: 89\n2. google: 78\n3. massachusetts institute of technology: 69\n4. deepmind: 68\n5. stanford university: 66\n6. university of california, berkeley: 60\n7. microsoft: 59\n8. eth zurich: 31\n9. university of oxford: 29\n10. duke university: 28\n11. princeton: 28", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.954+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5c7d0204-703b-48c6-a67e-afaf26be7519", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Talent Concentration: Google & DeepMind dominate NIPS authorship", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:43:59.954+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b70e0a9f-2026-4051-9cb9-46fa01af20e4", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Pareto Principle (80/20)", "slug": "pareto-principle-80-20", "status": "active", "bestFor": null, "canonId": "019dd956-8cc3-75ba-a35d-d6575b56c45d", "version": 1, "bodyDocId": null, "description": "80% of effects come from 20% of causes - focus on high-impact items", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "block", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-480ae1011fc7", "evidence": "Top-11 institutions ranked — concentration story.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 64, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-445ac238214f", "evidence": "Title 'Google & DeepMind dominate NIPS authorship'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 64, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 64, "from": 57, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6709b420daf9", "evidence": "Element AI counts (22k, 5k) -> talent flows -> Google headcount -> NIPS/ICML authorship dominance.", "position": 9, "objective": "Prove AI talent is scarce and concentrated at Google", "confidence": 82, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 65, "type": "analysis", "title": "The chart uses a 100% stacked bar format to compare salary growth trends by country.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 35, "componentCount": 4, "textChars": 398, "nDataPoints": 96, "notes": "The chart uses a 100% stacked bar format to compare salary growth trends by country.", "elementsJson": [ "headline_text", "bar_chart_100pct", "footnote" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:00.101+00", "seconds": 2.185, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/65", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-65", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.65, "w": 0.6, "x": 0.2, "y": 0.2 }, "kind": "chart", "text": "Salary / Compensation Changes in the Past 3 Years for Data Scientists and Machine Learning Engineers", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked bar chart (100%)", "slug": "bar-stacked-100pct", "status": "active", "canonId": "019de52c-ff14-71b8-8d88-860f8d7da84b", "version": 1, "description": null }, "framework": null }, "subkind": "bar-stacked-100pct", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.101+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8bf08c1b-5f38-4226-9dc5-aadd12808002", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Salary compensation change", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-2561e9fbc3a1", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.6, "x": 0.2, "y": 0.9 }, "kind": "source-note", "text": "N = 1549. All values are in US Dollars. Conversion rates from 2017 (when the data were collected) were used for conversion. Data are from The Kaggle 2017 The State of Data Science and Machine Learning study.", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.101+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d85a855c-2653-45c4-adef-78a9726a0dba", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.6, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Demand: Salaries for machine learning engineers continue to climb", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.101+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "faefe48b-d77c-405f-8ef8-221e565fdcbe", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "cccf6def-a3b7-45fb-9806-032281b8f953", "evidence": "chart/bar-stacked-100pct: Salary / Compensation Changes in the Past 3 Years for Data Scientists and Machine Learning Engineers", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:57:08.80941+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 65, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 67, "from": 65, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-689ec650b673", "evidence": "Salaries climb -> $300-500k base anecdotes -> $120M Google self-driving boss as tipping point.", "position": 10, "objective": "Quantify escalating talent compensation arms race", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 66, "type": "resolution", "title": "The slide uses external media citations to validate the trend of rising AI talent costs.", "function": "cite_precedent", "rawType": "key_takeaways", "block": null, "metadata": { "slideType": "key_takeaways", "slideTypeCanon": { "name": "Key takeaways", "slug": "key_takeaways", "status": "active", "canonId": "019de52d-06e7-7674-a891-9f7de3b11b6d", "version": 1, "description": null }, "function": "cite_precedent", "functionCanon": { "name": "Cite precedent", "slug": "cite_precedent", "status": "active", "canonId": "019de52d-142b-714b-86d4-58aad7b568ec", "version": 1, "description": null }, "density": "dense", "densityScore": 56, "componentCount": 6, "textChars": 674, "nDataPoints": 6, "notes": "The slide uses external media citations to validate the trend of rising AI talent costs.", "elementsJson": [ "quote_block", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:00.645+00", "seconds": 2.646, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/66", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-66", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Typical A.I. specialists, including both Ph.D.s fresh out of school and people with less education and just a few years of experience, can be paid from $300,000 to $500,000 a year or more in salary and company stock.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-88822d43866c", "parentComponentId": null }, { "bbox": { "h": 0.65, "w": 0.2, "x": 0.02, "y": 0.25 }, "kind": "image", "text": "Logos for The New York Times, Reuters, and Nature", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.645+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9e247ef6-462a-49f0-a516-ac6c45aacf2d", "parentComponentId": null }, { "bbox": { "h": 0.65, "w": 0.7, "x": 0.26, "y": 0.25 }, "kind": "list", "text": "Quotes from The New York Times, Reuters, and Nature regarding AI salary benchmarks and growth trends.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.645+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3152df45-0407-4f78-b7d7-22708ae50f69", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Annual salary: $1.9 million", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-90b4ba43721d", "parentComponentId": null }, { "bbox": null, "kind": "quote", "text": "Typical A.I. specialists, including both Ph.D.s fresh out of school and people with less education and just a few years of experience, can be paid from $300,000 to $500,000 a year or more in salary and company stock — The New York Times", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": { "name": "Authority citation", "slug": "authority-citation", "status": "active", "bestFor": null, "canonId": "019dd9e1-49ae-7557-89ad-3fca70469b45", "version": 1, "bodyDocId": null, "description": "Quoting an authority figure, study, or precedent to lend weight.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-8f1671b94fcb", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Demand: Anecdotally salaries continue to grow", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.645+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a1ce82f7-bc41-43d2-bc31-779398d2e03e", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-4f2e66e5c135", "evidence": "Title 'Anecdotally salaries continue to grow'; $300-500k quote.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 66, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "AIDA Model", "slug": "aida-model", "status": "active", "bestFor": null, "canonId": "019dd956-86fc-773c-813f-d416714b8413", "version": 1, "bodyDocId": null, "description": "Attention, Interest, Desire, Action - classic persuasion sequence", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "14050086-8286-46e2-bbd2-946ee577317e", "evidence": "quote: Typical A.I. specialists, including both Ph.D.s fresh out of school and people with less education and just a few years of experience, can be paid from $300,000 to $500,000 a year or more in salary and company sto", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:57:08.88717+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 66, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-50e8ece3290f", "evidence": "Specific salary band rather than abstract 'high pay'.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 66, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [ { "matchId": "9a0612bd-5f06-4635-ad0f-c3d034d27ee8", "evidence": "Uses reputable news sources to validate a market trend.", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:44:00.645+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 66, "frameworkName": "authority-citation" } ], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 67, "from": 65, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-689ec650b673", "evidence": "Salaries climb -> $300-500k base anecdotes -> $120M Google self-driving boss as tipping point.", "position": 10, "objective": "Quantify escalating talent compensation arms race", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 67, "type": "evidence", "title": "The slide uses screenshots of news articles to provide evidence of industry trends regarding talent compensation and litigation.", "function": "illustrate_case", "rawType": "client_example", "block": null, "metadata": { "slideType": "client_example", "slideTypeCanon": { "name": "Client example", "slug": "client_example", "status": "active", "canonId": "019de52d-0c87-77cb-9867-613ae22e95ac", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 7, "textChars": 426, "nDataPoints": 2, "notes": "The slide uses screenshots of news articles to provide evidence of industry trends regarding talent compensation and litigation.", "elementsJson": [ "headline_text", "screenshot" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:00.31+00", "seconds": 2.218, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/67", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-67", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Google paid its self-driving car boss $120 million — and then he left for Uber.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-6da9dbca89b8", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Compensation: $120 million", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-766c85c20706", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.96, "x": 0.02, "y": 0.02 }, "kind": "other", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Other", "slug": "other", "status": "active", "canonId": "019de52c-fa90-743c-8bf6-20d1fbe82b6d", "version": 1, "description": "Fallback when no canonical kind applies." }, "tool": null, "subkind": { "name": "Unclassified", "slug": "unclassified", "status": "active", "canonId": "019de52d-05ca-77db-81fb-db7d4234bbc0", "version": 1, "description": null }, "framework": null }, "subkind": "unclassified", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.31+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "eedad95b-bcce-4b7e-b7f6-520626802e53", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.5, "x": 0.23, "y": 0.25 }, "kind": "paragraph", "text": "Google paid its self-driving car boss $120 million — and then he left for Uber", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.31+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4b3c7493-f297-452e-885a-0122b210dc9e", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.5, "x": 0.23, "y": 0.7 }, "kind": "paragraph", "text": "UBER AND WAYMO ABRUPTLY SETTLE FOR $245 MILLION", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.31+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ca642491-e9af-4bdd-a4a8-40a1da7a6cd2", "parentComponentId": null }, { "bbox": null, "kind": "quote", "text": "UBER AND WAYMO ABRUPTLY SETTLE FOR $245 MILLION", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": { "name": "Authority citation", "slug": "authority-citation", "status": "active", "bestFor": null, "canonId": "019dd9e1-49ae-7557-89ad-3fca70469b45", "version": 1, "bodyDocId": null, "description": "Quoting an authority figure, study, or precedent to lend weight.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-728ccaf47dd0", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.75, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Demand: Compensation can be astronomical and relationships litigious", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.31+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "da48d531-543d-40d9-be9b-89e8b4fe906e", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-557714f6ece3", "evidence": "Title 'Compensation can be astronomical and relationships litigious'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 67, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Monroe's Motivated Sequence", "slug": "monroes-motivated-sequence", "status": "active", "bestFor": null, "canonId": "019dd956-8758-700f-83cf-d260a54ad24f", "version": 1, "bodyDocId": null, "description": "Five-step persuasion: Attention, Need, Satisfaction, Visualization, Action", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "f15a095d-a4e0-4e94-8b3a-87bb0872b445", "evidence": "paragraph/paragraph: Google paid its self-driving car boss $120 million — and then he left for Uber.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:57:08.965632+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 67, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Story Moments", "slug": "story-moments", "status": "active", "bestFor": null, "canonId": "019dd956-ae4e-768d-a18b-108e6cda511e", "version": 1, "bodyDocId": null, "description": "Design key emotional beats: The Shock, The Vision, The Proof, The Choice, The Call", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-5debd1d5d586", "evidence": "Single anecdote functions as 'the shock' moment.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 67, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Von Restorff Effect", "slug": "von-restorff-effect", "status": "active", "bestFor": null, "canonId": "019dd956-beab-73d8-9f62-30d0332e926e", "version": 1, "bodyDocId": null, "description": "Items that stand out are noticed and remembered - make key points visually distinct", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8ecb-5bcaaef47223", "evidence": "$120M figure stands out, designed to shock.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 67, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 67, "from": 65, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-689ec650b673", "evidence": "Salaries climb -> $300-500k base anecdotes -> $120M Google self-driving boss as tipping point.", "position": 10, "objective": "Quantify escalating talent compensation arms race", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 68, "type": "resolution", "title": "Includes logos for WiML and Black in AI.", "function": "summarize", "rawType": "key_takeaways", "block": null, "metadata": { "slideType": "key_takeaways", "slideTypeCanon": { "name": "Key takeaways", "slug": "key_takeaways", "status": "active", "canonId": "019de52d-06e7-7674-a891-9f7de3b11b6d", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "dense", "densityScore": 71, "componentCount": 7, "textChars": 1124, "nDataPoints": 4, "notes": "Includes logos for WiML and Black in AI.", "elementsJson": [ "headline_text", "bullet_list", "logo_grid", "footnote" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:01.2+00", "seconds": 3.036, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/68", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-68", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Diversity metrics for the industry are rarely publicised.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-ee786777abba", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.24, "x": 0.74, "y": 0.68 }, "kind": "image", "text": "WiML and Black in AI logos", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.2+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "60800791-659a-44e2-835d-f55a62ef0e95", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.95, "x": 0.02, "y": 0.27 }, "kind": "list", "text": "Key research labs are not yet making their workforce diversity statistics public.\nThere are limited diversity stats for major machine learning conferences publicly available.\nFor the largest machine learning conference by attendance, NIPS (Neural Information Processing Systems), there is data available on a single dimension of diversity (gender) for the past few years*.\nFor NIPS, the percentage of female attendees was 17% in 2017. This is lower than the technology industry more generally (for example, 31% of Google employees are women and 20% of people in a technical role at Google are women).\nThe percentage of women attending NIPS has risen slightly over the past few years from 13% in 2015 to 17% in 2017.\nThere are various initiatives aiming to increase diversity in machine learning:", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.2+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "504906c9-7160-4e2b-b7fc-9ba5925695d4", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "percentage of female attendees: 17%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-f12b24e98900", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.02, "y": 0.88 }, "kind": "source-note", "text": "*please let us know if you have similar statistics on other measures of diversity, such as race, that we can add to the report", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.2+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "39dc5519-f684-4280-a557-204101aebe96", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.48, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Diversity metrics for the industry are rarely publicised", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.2+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "736257b3-6afa-45ee-820f-5da0b31c4e4c", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.33, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Diversity in machine learning", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.2+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "749e5002-9ff0-47e6-abce-a6f517bb7cd3", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-605693573b66", "evidence": "Topic-with-stat title '17%' on diversity.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 68, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "2410d9be-5f15-4f41-89df-a525adb529c4", "evidence": "list/bullet: Key research labs are not yet making their workforce diversity statistics public.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:57:09.045763+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 68, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 69, "type": "analysis", "title": "The chart shows a slight upward trend in female attendance at the NIPS conference over three years.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "sparse", "densityScore": 21, "componentCount": 4, "textChars": 179, "nDataPoints": 3, "notes": "The chart shows a slight upward trend in female attendance at the NIPS conference over three years.", "elementsJson": [ "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:00.353+00", "seconds": 2.12, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/69", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-69", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.6, "w": 0.5, "x": 0.25, "y": 0.25 }, "kind": "chart", "text": "% of NIPS conference registrations from women", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.353+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6f150bfc-a57a-4bc9-9689-5e4f2daf92e6", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "percentage of women attending NIPS: 17%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:17+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a810-af7d9e44f802", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.82, "y": 0.92 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.353+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6d70afce-3174-42f6-9615-f505711c4ccf", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Diversity in machine learning: percentage of women attending the NIPS conference", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:00.353+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "84bb4119-7d56-4831-a7ba-076cb014dc9d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "b85f5e8d-bc34-45e2-aac6-684aad5d19af", "evidence": "chart/bar-vertical: % of NIPS conference registrations from women", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:57:09.121678+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 69, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 69, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 4, "name": "The Facts (What)", "beatId": "019dd95a-0682-776c-8e34-ad4df4fe3ce7", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Facts (What)", "slug": "triple-take-the-facts-what", "status": "active", "canonId": "019dd9b8-092b-754c-8eaa-ede3ffdcb060", "version": 1, "description": null }, "evidence": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data.", "position": 1, "isPrimaryArc": true, "parentBeatType": { "name": "Setup", "slug": "setup", "status": "active", "canonId": "019dd9b8-020e-7505-af9b-d9b9a4679bcc", "version": 1, "description": "Where the deck grounds the reader before doing anything else. Almost every arc opens here." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 70, "type": "setup", "function": "transition", "rawType": "section_divider", "block": null, "metadata": { "slideType": "section_divider", "slideTypeCanon": { "name": "Section divider", "slug": "section_divider", "status": "active", "canonId": "019de52d-066a-7402-9554-6ae504999041", "version": 1, "description": null }, "function": "transition", "functionCanon": { "name": "Transition", "slug": "transition", "status": "active", "canonId": "019de52d-1365-7149-92aa-66df859e75fd", "version": 1, "description": null }, "density": "sparse", "densityScore": 14, "componentCount": 4, "textChars": 124, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:01.073+00", "seconds": 2.738, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/70", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-70", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.03, "w": 0.96, "x": 0.02, "y": 0.03 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.073+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "14ef56bd-3724-447c-b709-17cdec6bfeda", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.07, "x": 0.91, "y": 0.03 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.073+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2d9298e3-aec2-484e-a8d5-8f5385425711", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.09, "x": 0.89, "y": 0.92 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.073+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ed5f791a-6039-407f-b97f-927817b445b6", "parentComponentId": null }, { "bbox": { "h": 0.045, "w": 0.268, "x": 0.366, "y": 0.455 }, "kind": "title", "text": "Section 3: Industry", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.073+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5fa39f93-4175-41b5-8aa6-8260439b870d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "da6b3cbc-9b5d-4596-b850-3b7f42e7f99f", "evidence": "paragraph/paragraph: Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:57:09.198897+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 70, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 71, "type": "analysis", "title": "The chart on the left shows patent counts by company (Microsoft, Google, Amazon, Facebook, Apple). The chart on the right shows R&D spend by company, categorize", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 60, "componentCount": 7, "textChars": 459, "nDataPoints": 45, "notes": "The chart on the left shows patent counts by company (Microsoft, Google, Amazon, Facebook, Apple). The chart on the right shows R&D spend by company, categorized by Tech vs Non-tech.", "elementsJson": [ "bar_chart_stacked", "bar_chart_horizontal" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:01.6+00", "seconds": 3.205, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/71", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-71", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "AI intellectual property is concentrated amongst few global players who also spend billions on R&D per year", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-275013be9e3a", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.55, "y": 0.35 }, "kind": "chart", "text": "Top US companies for R&D spending in FY2017", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.6+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "69ec04d0-47db-41ec-86c6-1f6bd23413f6", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.45, "x": 0.04, "y": 0.35 }, "kind": "chart", "text": "AI related patent application activity by filling date", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": null, "framework": null }, "subkind": "bar-stacked-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.6+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b0357353-079d-4194-adb4-10c5710512ae", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "R&D spending: $22.6B", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-2b1011ce83a2", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.9, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "AI intellectual property is concentrated amongst few global players who also spend billions on R&D per year", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.6+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0a7131b0-5391-47cd-9dae-c7441b34be3b", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.5, "x": 0.02, "y": 0.94 }, "kind": "source-note", "text": "*Google, Apple, Facebook, Amazon, Microsoft, Baidu, Alibaba, Tencent", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.6+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4a176ffa-7e0b-48bd-a1ec-f513f0fe5b22", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI takes the world stage: GAFAMBAT* are in the ring together", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.6+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c3e11811-0360-443f-ac32-b4621a98b2e9", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-67b724cb8b9d", "evidence": "Title 'AI takes the world stage: GAFAMBAT are in the ring together'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 71, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8ecb-6d6ab973373f", "evidence": "Stacked bar (patents) + horizontal bar (R&D spend).", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 71, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-6bfef7291ac7", "evidence": "Boxing 'in the ring' metaphor.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 71, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Visual Hierarchy", "slug": "visual-hierarchy", "status": "active", "bestFor": null, "canonId": "019dd956-b2ca-706b-937f-e52441086fd7", "version": 1, "bodyDocId": null, "description": "Order of perception through Size > Color > Position > Weight > Space", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8ecb-73b3aebb4967", "evidence": "Tech vs non-tech color-coded for emphasis.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 71, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 73, "from": 71, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6ee015b0dd0a", "evidence": "Patent filings concentrated -> R&D spend leadership -> cloud APIs -> TensorFlow framework war.", "position": 11, "objective": "Show GAFAMBAT capture AI infrastructure layer", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 72, "type": "analysis", "title": "The slide uses a comparative structure to show the parity in AI service offerings across the three major cloud providers.", "function": "present_solution", "rawType": "market_landscape", "block": null, "metadata": { "slideType": "market_landscape", "slideTypeCanon": { "name": "Market landscape", "slug": "market_landscape", "status": "active", "canonId": "019de52d-0b6d-75db-a946-9ac362e3afc2", "version": 1, "description": null }, "function": "present_solution", "functionCanon": { "name": "Present solution", "slug": "present_solution", "status": "active", "canonId": "019de52d-15b8-71da-bd23-2b767850cbce", "version": 1, "description": null }, "density": "dense", "densityScore": 55, "componentCount": 8, "textChars": 385, "nDataPoints": 0, "notes": "The slide uses a comparative structure to show the parity in AI service offerings across the three major cloud providers.", "elementsJson": [ "headline_text", "action_title", "icon_grid", "other" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:01.47+00", "seconds": 2.998, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/72", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-72", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Big cloud providers are building and exposing the building blocks of intelligence via API.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-d0aabffeacef", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.96, "x": 0.02, "y": 0.28 }, "kind": "image", "text": "Google ML services icons", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Icon grid", "slug": "icon-grid", "status": "active", "canonId": "019de52c-fdfe-749d-9990-1db465a3eed4", "version": 1, "description": null }, "framework": null }, "subkind": "icon-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.47+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8bf8456e-727d-4114-b3a5-fa37932a98d6", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.2, "x": 0.02, "y": 0.71 }, "kind": "list", "text": "And so is Microsoft...", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.47+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3fcb6831-3422-49b2-9b9e-a104a27b90ff", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.2, "x": 0.02, "y": 0.46 }, "kind": "list", "text": "Amazon is doing the same...", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.47+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6bada22e-9534-49cf-8b64-bfccbf127c53", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.5, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "Google is investing heavily to expose ML services through their cloud ecosystem", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.47+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b44d1b9f-5f7f-4c11-ac31-cd21e0f8e774", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.96, "x": 0.02, "y": 0.76 }, "kind": "paragraph", "text": "Microsoft service descriptions", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.47+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c506f31d-4ada-41ee-aeb3-d59cd9ba2cc1", "parentComponentId": null }, { "bbox": { "h": 0.18, "w": 0.96, "x": 0.02, "y": 0.51 }, "kind": "table", "text": "Amazon service offerings", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.47+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b2422f78-6f58-45bb-8cbe-3c221aa91811", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.96, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Big cloud providers are building and exposing the building blocks of intelligence via API", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:01.47+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f9974a12-bcc1-480d-9360-c0edf0dfe6ae", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-740ea173cacc", "evidence": "Title 'Big cloud providers... building blocks of intelligence via API'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 72, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Canvas framework", "slug": "canvas-framework", "status": "active", "bestFor": null, "canonId": "019dd9e1-5005-762c-82f8-265b08436ca1", "version": 1, "bodyDocId": null, "description": "Named canvas (business model canvas, value prop canvas, etc.).", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "4660af07-cfc2-4a30-848c-e634dc0b063a", "evidence": "list/bullet: Google is investing heavily to expose ML services through their cloud ecosystem", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:57:09.32286+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 72, "whyItWorks": null, "antipattern": null, "cardinality": "optional", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 73, "from": 71, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6ee015b0dd0a", "evidence": "Patent filings concentrated -> R&D spend leadership -> cloud APIs -> TensorFlow framework war.", "position": 11, "objective": "Show GAFAMBAT capture AI infrastructure layer", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 73, "type": "analysis", "title": "The slide uses two distinct metrics (GitHub stars and research citations) to validate the thesis of TensorFlow's market leadership.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 72, "componentCount": 7, "textChars": 992, "nDataPoints": 15, "notes": "The slide uses two distinct metrics (GitHub stars and research citations) to validate the thesis of TensorFlow's market leadership.", "elementsJson": [ "headline_text", "paragraph", "line_chart", "data_table", "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.348+00", "seconds": 3.225, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/73", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-73", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "This means Google acquires significant developer mindshare, creates an onramp onto their Cloud services, trains a generation of developers and researchers with their technology who contribute to improving it.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-ea6099cfd16a", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.42, "x": 0.55, "y": 0.45 }, "kind": "chart", "text": "Framework mentions in research publications", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.348+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1416c9b5-a7b5-4584-a5be-abd710efc95b", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.45, "x": 0.02, "y": 0.45 }, "kind": "chart", "text": "TensorFlow is extremely popular amongst developers", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.348+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "aa2ceb17-7ba7-4dc3-82c8-0d85ef819174", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "GitHub Star Count: 44508", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a811-ed70cccfe060", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.96, "x": 0.02, "y": 0.21 }, "kind": "paragraph", "text": "This means Google acquires significant developer mindshare, creates an onramp onto their Cloud services, trains a generation of developers and researchers with their technology who contribute to improving it. Their open source strategy also disarms potential competitors. However, practitioners feel intense uncertainty on how things will play out in the field. The wrong framework choice could have significant ramifications, not least refactoring costs.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.348+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d3c15176-9fe3-4a87-91ff-49e029e6e60a", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.15, "x": 0.35, "y": 0.52 }, "kind": "table", "text": "Framework GitHub Star Count: TensorFlow 44508, scikit-learn 16191, Caffe 15690, CNTK 9383, MXNet 7896, Torch 6285, Theano 5568", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.348+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "db95f7f0-b637-4a4d-aa9e-4799b6be6ac9", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.75, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Google’s TensorFlow is winning the ML framework war, but the grounds are shifting fast", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.348+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "970ef3a7-41cb-4b88-ba7c-9a9049c0bcbc", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-79bc24e8f5b2", "evidence": "Title 'Google's TensorFlow is winning the ML framework war'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 73, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Metaphor & Analogy", "slug": "metaphor-analogy", "status": "active", "bestFor": null, "canonId": "019dd956-adea-7729-bca1-70810d6238a9", "version": 1, "bodyDocId": null, "description": "Making abstract concepts concrete through familiar comparisons", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-7d9024dc625b", "evidence": "'Framework war' metaphor.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 73, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 73, "from": 71, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-6ee015b0dd0a", "evidence": "Patent filings concentrated -> R&D spend leadership -> cloud APIs -> TensorFlow framework war.", "position": 11, "objective": "Show GAFAMBAT capture AI infrastructure layer", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 74, "type": "tension", "title": "Includes a process diagram illustrating the drug development lifecycle from preclinical to clinical studies.", "function": "frame_problem", "rawType": "problem_statement", "block": null, "metadata": { "slideType": "problem_statement", "slideTypeCanon": { "name": "Problem statement", "slug": "problem_statement", "status": "active", "canonId": "019de52d-0879-768d-bba8-8da8a5a121d4", "version": 1, "description": null }, "function": "frame_problem", "functionCanon": { "name": "Frame problem", "slug": "frame_problem", "status": "active", "canonId": "019de52d-13db-753e-ad70-e9b2d0dab398", "version": 1, "description": null }, "density": "dense", "densityScore": 59, "componentCount": 9, "textChars": 263, "nDataPoints": 4, "notes": "Includes a process diagram illustrating the drug development lifecycle from preclinical to clinical studies.", "elementsJson": [ "headline_text", "action_title", "big_number", "process_diagram" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.11+00", "seconds": 2.869, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/74", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-74", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Why now? Today's drug development process is too slow and expensive", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-0c0135911f6f", "parentComponentId": null }, { "bbox": { "h": 0.48, "w": 0.54, "x": 0.23, "y": 0.45 }, "kind": "image", "text": "Drug development process flow diagram", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Infographic", "slug": "infographic", "status": "active", "canonId": "019de52c-fe26-750c-9d7d-9f34e21324e8", "version": 1, "description": null }, "framework": null }, "subkind": "infographic", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e6232fdf-178e-4299-bf35-5fb1799639ad", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.13, "x": 0.83, "y": 0.35 }, "kind": "metric", "text": "<10%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "62d38cae-b6fa-461f-9e2a-dd3960c5639e", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.19, "x": 0.33, "y": 0.35 }, "kind": "metric", "text": "10 times more!", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a2250d5f-a2c1-4ef5-af21-d07688e8b865", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.19, "x": 0.07, "y": 0.35 }, "kind": "metric", "text": "$2.6 billion per drug", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "de4b4fe2-0e30-4ed7-a459-af0ad70f95d0", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.19, "x": 0.6, "y": 0.35 }, "kind": "metric", "text": "10 years", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e38c9929-18eb-4979-936d-4c3a534b0c82", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "R&D cost: $2.6 billion", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-13968fb1c27a", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.63, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Why now? Today's drug development process is too slow and expensive", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3815af37-338a-4502-819e-47e85c2325b9", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.28, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Pharmaceutical industry", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7b543174-fdea-4dec-a5cc-47e3bdb68573", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Problem-Agitate-Solve (PAS)", "slug": "problem-agitate-solve-pas", "status": "active", "bestFor": null, "canonId": "019dd956-869f-7564-8d42-de54ed29ab90", "version": 1, "bodyDocId": null, "description": "Identify problem, agitate the pain, present the solution", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "block", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-8411d885a360", "evidence": "Why-now problem framing opens a P-A-S sequence for healthcare.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 74, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-804988813afe", "evidence": "Title 'Why now? Today's drug development process is too slow and expensive'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 74, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "72bf1512-1be1-46f8-a976-5a084d17ca6b", "evidence": "callout: Why now? Today's drug development process is too slow and expensive", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:57:09.440965+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 74, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [ { "matchId": "1f7b570e-7da8-4f46-8451-1a468634375f", "evidence": "The slide depicts the sequential stages of drug development from preclinical to clinical approval.", "framework": null, "confidence": 0.9, "extraction": { "at": "2026-05-02 17:44:02.11+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 74, "frameworkName": "process" } ], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 80, "from": 74, "loop": { "name": "14_day_in_life", "slug": "14-day-in-life", "status": "active", "bestFor": "Product demos, UX reviews, investor pitches focusing on pain points", "canonId": "019dd956-6b65-707f-9b14-89beaaab81fa", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The User's Struggle -> The Intervention (Your Solution) -> The Resolved State", "description": "Humanize abstract data by walking through a single user's struggle and resolution", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-735f28c87032", "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 75, "type": "analysis", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 51, "componentCount": 7, "textChars": 455, "nDataPoints": 0, "notes": null, "elementsJson": [ "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.023+00", "seconds": 2.512, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/75", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-75", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-6e065f4bd720", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.35, "x": 0.27, "y": 0.76 }, "kind": "image", "text": "Recursion, Qrativ, BenevolentAI", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.023+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7bbcbef8-0c6e-4d8b-ba25-1b7ee1763a67", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.65, "x": 0.27, "y": 0.45 }, "kind": "image", "text": "Atomwise, LabGenius, Insilico Medicine, Exscientia, GTN Ltd, Insitro", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.023+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8d962634-8ed2-417e-bc77-7092347fb62b", "parentComponentId": null }, { "bbox": { "h": 0.25, "w": 0.9, "x": 0.05, "y": 0.58 }, "kind": "list", "text": "Repurpose existing drugs: Discover how existing drugs can be repurposed for new conditions... Selected examples: [Logos]", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.023+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a7bcbb34-ccca-4c99-b224-d5bfd0b37b9a", "parentComponentId": null }, { "bbox": { "h": 0.25, "w": 0.9, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Develop new drugs: Teach a ML model to learn the rules of drug design... Selected examples: [Logos]", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.023+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c2814d27-d464-44c6-9b7b-b999298e3f2d", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.023+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "29c58153-fd83-427b-8e63-d8762ccb26ee", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.25, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Pharmaceutical industry", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.023+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "07b99c88-8ac4-4e70-9a4a-c1fa7dd12ab3", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "42dd50ba-b558-4481-8f96-53067bfd7a2e", "evidence": "list/bullet: Repurpose existing drugs: Discover how existing drugs can be repurposed for new conditions... Selected examples: [Logos]", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:57:09.52683+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 75, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 80, "from": 74, "loop": { "name": "14_day_in_life", "slug": "14-day-in-life", "status": "active", "bestFor": "Product demos, UX reviews, investor pitches focusing on pain points", "canonId": "019dd956-6b65-707f-9b14-89beaaab81fa", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The User's Struggle -> The Intervention (Your Solution) -> The Resolved State", "description": "Humanize abstract data by walking through a single user's struggle and resolution", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-735f28c87032", "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 76, "type": "analysis", "title": "The slide uses a line chart for longitudinal spending data and a stacked area chart for age-related health morbidity.", "function": "establish_context", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "balanced", "densityScore": 41, "componentCount": 6, "textChars": 259, "nDataPoints": 20, "notes": "The slide uses a line chart for longitudinal spending data and a stacked area chart for age-related health morbidity.", "elementsJson": [ "line_chart", "area_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:04.261+00", "seconds": 4.67, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/76", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-76", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Healthcare systems worldwide are costly and overburdened", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-4151113290e9", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.4, "x": 0.55, "y": 0.35 }, "kind": "chart", "text": "The older we get, the more health issues we have", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked area chart", "slug": "area-stacked", "status": "active", "canonId": "019de52c-ffb1-75fa-a614-76782f418cf1", "version": 1, "description": null }, "framework": null }, "subkind": "area-stacked", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.261+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cfe176b3-e8fb-47da-a0d0-aed5e98b115d", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.4, "x": 0.05, "y": 0.35 }, "kind": "chart", "text": "Healthcare spending as % of GDP growing since 80s", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.261+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "401f7bd5-b047-401f-9d85-3d96f60e5713", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Healthcare spending as % of GDP", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-4543c4e9b99b", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Why now? Healthcare systems worldwide are costly and overburdened", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.261+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0ced2809-fdfc-4378-baeb-7c8299a9a693", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.12, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Healthcare", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.261+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "44d3d4e1-067a-4757-9a92-d747c099f2d8", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-885ca802b5ea", "evidence": "Title 'Healthcare systems worldwide are costly and overburdened'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 76, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 80, "from": 74, "loop": { "name": "14_day_in_life", "slug": "14-day-in-life", "status": "active", "bestFor": "Product demos, UX reviews, investor pitches focusing on pain points", "canonId": "019dd956-6b65-707f-9b14-89beaaab81fa", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The User's Struggle -> The Intervention (Your Solution) -> The Resolved State", "description": "Humanize abstract data by walking through a single user's struggle and resolution", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-735f28c87032", "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 77, "type": "evidence", "title": "The slide uses a case study approach to frame the problem statement for AI application in healthcare.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 62, "componentCount": 9, "textChars": 341, "nDataPoints": 3, "notes": "The slide uses a case study approach to frame the problem statement for AI application in healthcare.", "elementsJson": [ "headline_text", "bullet_list", "big_number", "scatter_plot", "bar_chart_horizontal" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.383+00", "seconds": 2.703, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/77", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-77", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Early detection of leads to higher 5-year survival rates", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-889ac660cf24", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.5, "x": 0.45, "y": 0.55 }, "kind": "chart", "text": "Early detection of leads to higher 5-year survival rates", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Dot plot", "slug": "dot", "status": "active", "canonId": "019de52d-0235-738b-bd2b-78f8e24d7cb0", "version": 1, "description": null }, "framework": null }, "subkind": "dot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.383+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2b61ed6e-f935-46d8-819a-ecefb19962c1", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.35, "x": 0.05, "y": 0.55 }, "kind": "chart", "text": "Cancer treatment cost, median $/month", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Scatter plot", "slug": "scatter", "status": "active", "canonId": "019de52d-007d-72ec-8689-c79fb78e21a2", "version": 1, "description": null }, "framework": null }, "subkind": "scatter", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.383+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c629c7ec-4cee-4c33-a9d3-407bb14d1ee5", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.25, "x": 0.35, "y": 0.35 }, "kind": "metric", "text": "34,000 professionals", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.383+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "57c46e5e-6612-4944-b305-17febda7c2a3", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.25, "x": 0.05, "y": 0.35 }, "kind": "metric", "text": "Up to 30% of cases", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.383+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "632ae9f8-d60d-4f95-b538-ccca1e56e37c", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.3, "x": 0.65, "y": 0.35 }, "kind": "metric", "text": "30 million patients per year", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.383+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7e8162aa-d9cc-41fe-b50c-0852ee87c7d6", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "5-year survival rate: 30%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-8e34ae08c80b", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.75, "x": 0.04, "y": 0.22 }, "kind": "title", "text": "Breast cancer as a case study: Not enough doctors, diagnosing is hard and care is expensive", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.383+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "78aa7595-9358-4d9b-97ca-838e8fde95da", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.04, "y": 0.14 }, "kind": "title", "text": "Healthcare", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.383+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6476486b-4927-4a42-9592-a76072788065", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-8f11fc303e1d", "evidence": "Title 'Breast cancer as a case study: Not enough doctors, diagnosing is hard'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 77, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "AIDA Model", "slug": "aida-model", "status": "active", "bestFor": null, "canonId": "019dd956-86fc-773c-813f-d416714b8413", "version": 1, "bodyDocId": null, "description": "Attention, Interest, Desire, Action - classic persuasion sequence", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "7cc7e898-c2fc-4dd8-9c50-e419ecdd557e", "evidence": "The slide presents a problem (not enough doctors), agitation (diagnosing is hard and care is expensive), and a potential solution or call to action (early detection leads to higher 5-year survival rates).", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:34.922061+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 77, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Sinatra Test", "slug": "sinatra-test", "status": "active", "bestFor": null, "canonId": "019dd956-be4a-775f-8038-d4e7e58dfc5a", "version": 1, "bodyDocId": null, "description": "If you can make it there, you can make it anywhere - using flagship example", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-9225dd0e2e32", "evidence": "Breast cancer flagship case proves healthcare AI value.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 77, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 80, "from": 74, "loop": { "name": "14_day_in_life", "slug": "14-day-in-life", "status": "active", "bestFor": "Product demos, UX reviews, investor pitches focusing on pain points", "canonId": "019dd956-6b65-707f-9b14-89beaaab81fa", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The User's Struggle -> The Intervention (Your Solution) -> The Resolved State", "description": "Humanize abstract data by walking through a single user's struggle and resolution", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-735f28c87032", "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 78, "type": "analysis", "title": "The slide uses a list-based structure to categorize AI applications in healthcare, supported by company logos and a biological diagram.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 67, "componentCount": 8, "textChars": 742, "nDataPoints": 0, "notes": "The slide uses a list-based structure to categorize AI applications in healthcare, supported by company logos and a biological diagram.", "elementsJson": [ "headline_text", "bullet_list", "logo_grid", "infographic" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.956+00", "seconds": 3.261, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/78", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-78", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-6d6eb36b03ef", "parentComponentId": null }, { "bbox": { "h": 0.22, "w": 0.42, "x": 0.09, "y": 0.71 }, "kind": "image", "text": "Biological diagram of liquid biopsy process", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.956+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bd812d8f-f1c0-485b-ba83-52d82913fa0c", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.45, "x": 0.47, "y": 0.75 }, "kind": "image", "text": "Freenome, Grail, Guardant Health, Karius", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.956+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "16cf3092-0d50-4ab6-87f9-e401dc659d20", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.85, "x": 0.09, "y": 0.45 }, "kind": "image", "text": "Kheiron, Zebra, Arterys, Baylabs, Viz.ai, MaxQ, Cardiologs", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.956+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2e6e7830-8473-461b-8167-57282fa7b7a2", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.92, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Medical imaging: Train computer vision models on large numbers of labeled medical images (e.g. X-ray, ultrasound) with matched and clinically-validated patient diagnoses. Use this system to help doctors process more patient cases and make fewer diagnostic mistakes.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.956+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "19a8eda3-3f98-4fd7-ba82-2d6019af59d9", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.92, "x": 0.05, "y": 0.61 }, "kind": "list", "text": "Liquid biopsy: Isolate and analyse material such as cells or bacteria circulating in a patient's bloodstream. This approach allows for early non-invasive disease diagnosis as well as tracking response to therapy.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.956+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3f08ce44-3474-42bf-b195-e0ff76859bb2", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.956+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3837bddc-c097-4fa2-a2b4-d0091bad70a7", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Healthcare", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.956+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "40bd8c1d-0d8a-472a-8ec0-7c1c8db84fc5", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "2x2 matrix", "slug": "matrix-2x2", "status": "active", "bestFor": null, "canonId": "019dd9e1-50ab-72d1-8afb-3ab0a38e32a8", "version": 1, "bodyDocId": null, "description": "BCG matrix, importance/urgency, any axes-and-quadrants frame.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "5e6a4188-bfef-4a0d-ab6e-58966cf28462", "evidence": "Although not explicitly a 2x2 matrix, the logo grid could be argued to represent a simple matrix structure.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:35.043077+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 78, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "cf55e093-a729-4ff6-b7c1-cf7c0ab3f7dd", "evidence": "The slide presents information in a list format with bullet points, making it a list presentation.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.00183+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 78, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 80, "from": 74, "loop": { "name": "14_day_in_life", "slug": "14-day-in-life", "status": "active", "bestFor": "Product demos, UX reviews, investor pitches focusing on pain points", "canonId": "019dd956-6b65-707f-9b14-89beaaab81fa", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The User's Struggle -> The Intervention (Your Solution) -> The Resolved State", "description": "Humanize abstract data by walking through a single user's struggle and resolution", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-735f28c87032", "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 79, "type": "analysis", "title": "The slide shows a workflow for training and testing a deep learning model for tumor detection in pathology slides, alongside examples of medical imaging (OCT an", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 40, "componentCount": 7, "textChars": 233, "nDataPoints": 0, "notes": "The slide shows a workflow for training and testing a deep learning model for tumor detection in pathology slides, alongside examples of medical imaging (OCT and MRI).", "elementsJson": [ "process_diagram", "screenshot" ], "confidence": 0.9, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.768+00", "seconds": 6.025, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/79", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-79", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.65, "w": 0.38, "x": 0.11, "y": 0.21 }, "kind": "image", "text": "Workflow diagram showing training and testing of a deep learning model for tumor detection", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Infographic", "slug": "infographic", "status": "active", "canonId": "019de52c-fe26-750c-9d7d-9f34e21324e8", "version": 1, "description": null }, "framework": null }, "subkind": "infographic", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.768+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "143e3022-4b38-48de-8b90-b7a41a0dd5b6", "parentComponentId": null }, { "bbox": { "h": 0.37, "w": 0.41, "x": 0.52, "y": 0.21 }, "kind": "image", "text": "OCT scan image", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.768+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4a8d6f9e-ca12-4e7a-bd34-5ae544c8665d", "parentComponentId": null }, { "bbox": { "h": 0.31, "w": 0.35, "x": 0.55, "y": 0.6 }, "kind": "image", "text": "MRI scan image", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.768+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "69fa2d6c-8ed6-4331-9043-88f0320abc2b", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.88, "y": 0.95 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.768+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "12e786b0-c6f0-4e62-87b2-9cd488624c35", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.75, "x": 0.02, "y": 0.03 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.768+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e4cb5fc7-56b4-4e1e-8630-1f6ba1a6d0b3", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.88, "y": 0.03 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.768+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f5497c62-8371-4065-9a09-dc2d8fb09ed7", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Healthcare", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.768+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3a3a4cf4-6a9c-4869-b283-893ff5a2e316", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Narrative foundations (from the Storymakers methodology)", "slug": "narrative-foundations", "status": "active", "bestFor": null, "canonId": "019df269-258b-71bd-ad14-53598df04d2b", "version": 1, "bodyDocId": "019df269-2567-736e-8b9c-4ec77611334a", "description": "This skill focuses on the *adversarial* genre — activist campaigns, short", "familyLabel": "narrative-block", "categoryName": "Narrative & Transformation", "categorySlug": "narrative-transformation" }, "agent": "storyteller", "layer": "slide", "agents": [ "storyteller" ], "matchId": "28dc91c9-409d-44fb-b49c-39bb7e68d954", "evidence": "The slide has a structured approach with Introduction, Research, Talent, Industry, Politics, Predictions, and Conclusion sections.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:35.124053+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 79, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 80, "from": 74, "loop": { "name": "14_day_in_life", "slug": "14-day-in-life", "status": "active", "bestFor": "Product demos, UX reviews, investor pitches focusing on pain points", "canonId": "019dd956-6b65-707f-9b14-89beaaab81fa", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The User's Struggle -> The Intervention (Your Solution) -> The Resolved State", "description": "Humanize abstract data by walking through a single user's struggle and resolution", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-735f28c87032", "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 80, "type": "analysis", "title": "The chart uses a horizontal bar layout where each bar is composed of company logos.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 44, "componentCount": 6, "textChars": 350, "nDataPoints": 8, "notes": "The chart uses a horizontal bar layout where each bar is composed of company logos.", "elementsJson": [ "headline_text", "action_title", "bar_chart_horizontal", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.745+00", "seconds": 2.99, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/80", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-80", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Expect to see more activity as companies move their products through clinical trials and regulatory bodies", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:22+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-4a1395feb8d7", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.75, "x": 0.04, "y": 0.35 }, "kind": "chart", "text": "Horizontal bar chart showing company counts by year", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.745+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e55d66df-205b-40eb-9dfb-67dd95930afa", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Number of companies", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:22+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-4cd2d83c2da4", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.85, "x": 0.04, "y": 0.22 }, "kind": "title", "text": "Expect to see more activity as companies move their products through clinical trials and regulatory bodies", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.745+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "87e092db-b7cd-422b-93d6-cc87262909a4", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Healthcare", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.745+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7f013c0b-02b5-4b70-981e-c7664949b790", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.04, "y": 0.3 }, "kind": "title", "text": "Number of medical imaging AI companies founded per vintage", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Subtitle", "slug": "subtitle", "status": "active", "canonId": "019de52c-fb09-739d-900c-815b0d9ac526", "version": 1, "description": null }, "framework": null }, "subkind": "subtitle", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.745+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1945d1e1-d754-4c18-bdd8-474a80965960", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-97b35c34d4cb", "evidence": "Title 'Expect to see more activity as companies move products through trials'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 80, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "020daf1f-efda-4040-a9f8-e8bd4caf93ee", "evidence": "The slide uses a horizontal bar chart to show company counts by year, which is an appropriate chart type for categorical data.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.201868+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 80, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 80, "from": 74, "loop": { "name": "14_day_in_life", "slug": "14-day-in-life", "status": "active", "bestFor": "Product demos, UX reviews, investor pitches focusing on pain points", "canonId": "019dd956-6b65-707f-9b14-89beaaab81fa", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The User's Struggle -> The Intervention (Your Solution) -> The Resolved State", "description": "Humanize abstract data by walking through a single user's struggle and resolution", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-735f28c87032", "evidence": "Drug dev too slow ($2.6B) -> healthcare overburdened -> breast-cancer case study -> ML deployments and trials.", "position": 12, "objective": "Show pharma/healthcare struggle and AI intervention", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 81, "type": "analysis", "title": "Includes screenshots of SenseTime's surveillance software interface.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 70, "componentCount": 8, "textChars": 641, "nDataPoints": 4, "notes": "Includes screenshots of SenseTime's surveillance software interface.", "elementsJson": [ "headline_text", "action_title", "paragraph", "screenshot", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.687+00", "seconds": 5.827, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/81", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-81", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "The Chinese government continue to roll out CCTV surveillance software based on computer vision.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-aa1d0643b59e", "parentComponentId": null }, { "bbox": { "h": 0.42, "w": 0.38, "x": 0.53, "y": 0.49 }, "kind": "image", "text": "Long-range (>100m) Face Capturing / Recognition software interface", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Screenshot", "slug": "screenshot", "status": "active", "canonId": "019de52c-fd87-7509-8391-3d79fb339379", "version": 1, "description": null }, "framework": null }, "subkind": "screenshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.687+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "13ca7baf-40d3-4ffe-8a36-243561bd6232", "parentComponentId": null }, { "bbox": { "h": 0.42, "w": 0.38, "x": 0.08, "y": 0.49 }, "kind": "image", "text": "Short-range Face Capturing / Recognition software interface", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Screenshot", "slug": "screenshot", "status": "active", "canonId": "019de52c-fd87-7509-8391-3d79fb339379", "version": 1, "description": null }, "framework": null }, "subkind": "screenshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.687+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3035fc1f-26b3-403b-89c8-074b28d9cabe", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "CCTV camera count: 400 million", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:19+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-ae76e93a35e2", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.95, "x": 0.02, "y": 0.27 }, "kind": "paragraph", "text": "The Chinese government continue to roll out CCTV surveillance software based on computer vision. There are 170 million CCTV cameras as of late 2017. This network will grow to 400 million cameras in 3 years time.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.687+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "49bfd25d-edb4-40a6-8195-f09512b16de7", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.95, "x": 0.02, "y": 0.39 }, "kind": "paragraph", "text": "4 year-old SenseTime is leading the charge. It's valued >$4.5B since raising $620M Series C+ in May 2018.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.687+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "eab9cef2-aee1-4307-af1a-6e1c8f33d293", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Population-level surveillance is taking off in China", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.687+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9765382c-570b-499c-8691-dd5ca70c5036", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.28, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Government and defense", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.687+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f54f638a-3ddb-41ab-b3cc-1e129665c4a4", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-98a520264498", "evidence": "Title 'Population-level surveillance is taking off in China' with 400M figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 81, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Storytelling Effect", "slug": "storytelling-effect", "status": "active", "bestFor": null, "canonId": "019dd956-981b-74f9-9096-5c8df03dab5a", "version": 1, "bodyDocId": null, "description": "Stories are remembered up to 22x better than facts alone", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "78136482-56b8-45a2-a804-6bbfe2ec01ff", "evidence": "The slide tells a story about the growth of CCTV surveillance in China and its implications.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.282017+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 81, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 85, "from": 81, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7442641e422c", "evidence": "China surveillance -> Project Maven -> Cambridge Analytica -> privacy preservation tooling.", "position": 13, "objective": "Zoom into surveillance/privacy as the dark side of AI", "confidence": 68, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 82, "type": "analysis", "title": "The slide uses news snippets to illustrate the ethical tensions in AI development for defense.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "overcrowded", "densityScore": 83, "componentCount": 10, "textChars": 1004, "nDataPoints": 1, "notes": "The slide uses news snippets to illustrate the ethical tensions in AI development for defense.", "elementsJson": [ "headline_text", "bullet_list", "paragraph", "photo", "screenshot" ], "confidence": 0.9, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:03.514+00", "seconds": 3.464, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/82", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-82", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "In the US, companies including Google and Clarifai supplied AI technology to the Pentagon's Project Maven.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:20+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-cf150e13aed8", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.13, "x": 0.04, "y": 0.75 }, "kind": "image", "text": null, "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.514+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "173a9c21-0ada-4d8f-b6f2-ee406552949b", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.12, "x": 0.39, "y": 0.34 }, "kind": "image", "text": null, "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.514+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2a026268-08a5-4702-9496-717634fd680b", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.9, "x": 0.02, "y": 0.67 }, "kind": "list", "text": "In response, >4,500 Google employees signed a petition to quit if the company were to continue as a supplier", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.514+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "17c39684-29b1-411d-9cda-fb99303def89", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.9, "x": 0.02, "y": 0.23 }, "kind": "list", "text": "In the US, companies including Google and Clarifai supplied AI technology to the Pentagon's Project Maven", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.514+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "228c12e9-611a-4f8d-9367-40da503abc01", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.23, "x": 0.61, "y": 0.75 }, "kind": "paragraph", "text": "AI at Google: our principles - Google Blog. 7 Jun 2018 - We're announcing seven principles to guide our work in AI.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.514+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3a0cab19-5407-42dd-af79-8ec798deeee1", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.23, "x": 0.18, "y": 0.75 }, "kind": "paragraph", "text": "In wake of Project Maven backlash, Google unveils new AI policies. Today Google unveiled a new set of principles guiding its approach to artificial intelligence, including a pledge not to build AI weapons.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.514+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bad38566-8bf6-4176-ac0f-974fa9cab8da", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.38, "x": 0.04, "y": 0.34 }, "kind": "quote", "text": "Fired employee alleges in a lawsuit that Clarifai, an AI startup working on Project Maven, was hacked from Russia and did not promptly report it to the Pentagon", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": null, "subkind": { "name": "Pull quote", "slug": "pull-quote", "status": "active", "canonId": "019de52c-fbf9-76fe-97e1-a0b7540445c9", "version": 1, "description": null }, "framework": null }, "subkind": "pull-quote", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.514+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a9f9864b-970b-4ba7-8118-f0ba64ef8347", "parentComponentId": null }, { "bbox": null, "kind": "quote", "text": "Fired employee alleges in a lawsuit that Clarifai, an AI startup working on Project Maven, was hacked from Russia and did not promptly report it to the Pentagon — Tom Simonite / Wired", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": { "name": "Authority citation", "slug": "authority-citation", "status": "active", "bestFor": null, "canonId": "019dd9e1-49ae-7557-89ad-3fca70469b45", "version": 1, "bodyDocId": null, "description": "Quoting an authority figure, study, or precedent to lend weight.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:20+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a812-d24f935b7676", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.28, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Government and defense", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.514+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "32ce211b-48c7-4861-b3f4-ad369283707d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Problem-Agitate-Solve (PAS)", "slug": "problem-agitate-solve-pas", "status": "active", "bestFor": null, "canonId": "019dd956-869f-7564-8d42-de54ed29ab90", "version": 1, "bodyDocId": null, "description": "Identify problem, agitate the pain, present the solution", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "00b6b4ac-f5ed-4b1c-bf08-d5eb1407209f", "evidence": "The slide presents a problem (Project Maven backlash), agitates it (Google's involvement), and hints at a solution (Google's new AI principles).", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.362031+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 82, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 85, "from": 81, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7442641e422c", "evidence": "China surveillance -> Project Maven -> Cambridge Analytica -> privacy preservation tooling.", "position": 13, "objective": "Zoom into surveillance/privacy as the dark side of AI", "confidence": 68, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 83, "type": "analysis", "title": "The slide uses a visual collage of data-related icons and text labels to illustrate the breadth of personal data collection.", "function": "establish_context", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "balanced", "densityScore": 29, "componentCount": 5, "textChars": 327, "nDataPoints": 0, "notes": "The slide uses a visual collage of data-related icons and text labels to illustrate the breadth of personal data collection.", "elementsJson": [ "screenshot", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.612+00", "seconds": 2.447, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/83", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-83", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "In the wake of the Cambridge Analytica scandal, personal data privacy is now front and center", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-56e494373227", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.66, "x": 0.17, "y": 0.32 }, "kind": "image", "text": "Collage of data icons and labels with the text 'Palantir Knows Everything About You'", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Infographic", "slug": "infographic", "status": "active", "canonId": "019de52c-fe26-750c-9d7d-9f34e21324e8", "version": 1, "description": null }, "framework": null }, "subkind": "infographic", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.612+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d28107d9-e835-4412-87da-d8d781c90714", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.8, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "In the wake of the Cambridge Analytica scandal, personal data privacy is now front and center", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.612+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6ab44c54-7276-46a9-9b64-6ed9eee0f0ee", "parentComponentId": null }, { "bbox": null, "kind": "quote", "text": "Palantir Knows Everything About You", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": { "name": "Authority citation", "slug": "authority-citation", "status": "active", "bestFor": null, "canonId": "019dd9e1-49ae-7557-89ad-3fca70469b45", "version": 1, "bodyDocId": null, "description": "Quoting an authority figure, study, or precedent to lend weight.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-5adad925d408", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.28, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Government and defense", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.612+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c948c25a-176d-4ad5-a922-9ae1389df47d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Emotional Appeal", "slug": "emotional-appeal", "status": "active", "bestFor": null, "canonId": "019dd956-bde9-7668-8be7-0097678fec59", "version": 1, "bodyDocId": null, "description": "Connecting to audience feelings rather than just logic", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "2aba70c9-dcd8-4626-b304-81d1b51ce901", "evidence": "The slide uses a quote and an image to evoke an emotional response to the Cambridge Analytica scandal.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:35.44079+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 83, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 85, "from": 81, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7442641e422c", "evidence": "China surveillance -> Project Maven -> Cambridge Analytica -> privacy preservation tooling.", "position": 13, "objective": "Zoom into surveillance/privacy as the dark side of AI", "confidence": 68, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 84, "type": "analysis", "title": "Includes a counter-style infographic for 2017 data breaches and a stacked bar chart showing consumer confidence in institutions.", "function": "establish_context", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "dense", "densityScore": 55, "componentCount": 6, "textChars": 569, "nDataPoints": 30, "notes": "Includes a counter-style infographic for 2017 data breaches and a stacked bar chart showing consumer confidence in institutions.", "elementsJson": [ "headline_text", "paragraph", "big_number", "bar_chart_stacked" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:03.279+00", "seconds": 2.981, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/84", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-84", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Roughly half of Americans do not trust the federal government or social media sites to protect their data.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-92bd806907f0", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.41, "x": 0.57, "y": 0.49 }, "kind": "chart", "text": "Roughly half of Americans do not trust the federal government or social media sites to protect their data", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked bar chart (100%)", "slug": "bar-stacked-100pct", "status": "active", "canonId": "019de52c-ff14-71b8-8d88-860f8d7da84b", "version": 1, "description": null }, "framework": null }, "subkind": "bar-stacked-100pct", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.279+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1ef10d37-c056-42ab-abbe-1e5cb4e4306c", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.43, "x": 0.07, "y": 0.595 }, "kind": "chart", "text": "Data records compromised in 2017 infographic", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": null, "framework": null }, "subkind": "other", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.279+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4205b1e2-bc8f-4d78-8c74-baf3ce98ceb6", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Data records compromised: 2,600,968,280", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-9425510533e1", "parentComponentId": null }, { "bbox": { "h": 0.25, "w": 0.96, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "Why now? Massive data breaches such as Equifax’s heist of data about 146 million people has brought the privacy front of mind in industry. In Europe, the General Data Protection Regulation has come into effect since May 25th 2018...", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.279+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "64d8a811-f00a-4881-9fa3-9e813ae82382", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Privacy preservation and data anonymisation", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.279+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "607cb154-b2e9-4ea6-9f12-4590257550e1", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-9ef711004660", "evidence": "Concrete count 2,600,968,280 in title indicates breach scale.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 84, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chartjunk Elimination", "slug": "chartjunk-elimination", "status": "active", "bestFor": null, "canonId": "019dd956-b49e-7453-ab01-b60f338eeda1", "version": 1, "bodyDocId": null, "description": "Remove decorative elements that don't convey data: 3D effects, shadows, unnecessary gridlines", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "46daa846-80a7-47e4-9a5f-e15aea63b94d", "evidence": "The slide uses a simple and clear bar chart to present data on trust in government and social media sites.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.518968+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 84, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-a2c08559ad58", "evidence": "Exact integer '2,600,968,280' chosen over rounded version.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 84, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 85, "from": 81, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7442641e422c", "evidence": "China surveillance -> Project Maven -> Cambridge Analytica -> privacy preservation tooling.", "position": 13, "objective": "Zoom into surveillance/privacy as the dark side of AI", "confidence": 68, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 85, "type": "analysis", "title": "Part of the State of AI 2018 report.", "function": "summarize", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "dense", "densityScore": 56, "componentCount": 7, "textChars": 606, "nDataPoints": 0, "notes": "Part of the State of AI 2018 report.", "elementsJson": [ "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.755+00", "seconds": 2.405, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/85", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-85", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:20+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-4363f104410a", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.45, "x": 0.08, "y": 0.6 }, "kind": "image", "text": "PRIVITAR, DATAGUISE, BigID", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.755+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6e069844-7926-4edf-95c0-acc025b409a2", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.3, "x": 0.08, "y": 0.4 }, "kind": "image", "text": "Synthesized.io, statice", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.755+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ea5322a9-ae73-4e05-be43-2b41f12945ea", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Synthetic data generation: Training a machine learning model to learn the key statistical properties of a source dataset and using the model to generating synthetic data that preserves these properties.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.755+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5bf5d2a6-19c8-4ca1-adad-34e9a7f54c4a", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.48 }, "kind": "list", "text": "Obfuscating sensitive data: Detect sensitive data fields and anonymise them while preserving the important features of a dataset such that machine learning models can still learn useful information.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.755+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cdd10066-dc8a-4d05-ae97-e34dca634bb2", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.755+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "47605df0-25f0-45fd-9602-d53391fd53cf", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Privacy preservation and data anonymisation", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.755+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9df8235b-8933-4b35-a59d-8d6c44988735", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "2ffd5f9b-d92a-408d-b128-88705b793abb", "evidence": "The slide presents information in a list format with bullet points, making it a list presentation.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.596288+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 85, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 85, "from": 81, "loop": { "name": "06_zoom_in", "slug": "06-zoom-in", "status": "active", "bestFor": "Technical deep-dives, case studies, detailed analysis", "canonId": "019dd956-6846-74db-82be-dff5b058b072", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication", "description": "Start broad, then progressively focus on specific details that prove your point", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7442641e422c", "evidence": "China surveillance -> Project Maven -> Cambridge Analytica -> privacy preservation tooling.", "position": 13, "objective": "Zoom into surveillance/privacy as the dark side of AI", "confidence": 68, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 86, "type": "analysis", "title": "Part of the state.ai 2018 report.", "function": "establish_context", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "dense", "densityScore": 65, "componentCount": 8, "textChars": 532, "nDataPoints": 2, "notes": "Part of the state.ai 2018 report.", "elementsJson": [ "headline_text", "action_title", "paragraph", "bar_chart_stacked", "line_chart", "footnote", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.844+00", "seconds": 2.469, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/86", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-86", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Why now? Satellite data is decreasing in cost and increasing in resolution and frequency", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-c279776359c5", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.47, "x": 0.02, "y": 0.41 }, "kind": "chart", "text": "Worldwide commercial space launches by type", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked bar chart", "slug": "bar-stacked", "status": "active", "canonId": "019de52c-feec-74bf-9028-15b74f84f0a7", "version": 1, "description": null }, "framework": null }, "subkind": "bar-stacked", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.844+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ff4e344e-e92e-4714-b7a2-41ecae4a17ed", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.47, "x": 0.51, "y": 0.41 }, "kind": "chart", "text": "Weekly data collection by Planet", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.844+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1593aad6-e044-4089-a473-00af704431a8", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Weekly data collection", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-c4129f8c9b80", "parentComponentId": null }, { "bbox": { "h": 0.07, "w": 0.95, "x": 0.02, "y": 0.27 }, "kind": "paragraph", "text": "Driven by the rise of microsatellites, the decreasing costs of satellite components, the falling cost of launches and improvements in downlink infrastructure.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.844+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a4590595-5fc2-45a5-96f4-a05e5ddd0e44", "parentComponentId": null }, { "bbox": { "h": 0.02, "w": 0.4, "x": 0.02, "y": 0.88 }, "kind": "source-note", "text": "Data: FAA Office of Commercial Space Transportation; graphic by Bloomberg Businessweek.", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.844+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6c630d04-f15c-404e-bb86-a3c1acadd7cf", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.75, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Why now? Satellite data is decreasing in cost and increasing in resolution and frequency", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.844+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "09fc0ed5-2dd7-4a7c-88eb-d70b7aba905d", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.16, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Satellite data", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.844+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8f059537-1e0f-4452-9c51-11bdc5870ecf", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-a82f38546a5e", "evidence": "First slide of vertical-applications pattern_hunter loop.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 86, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-a64a500c77a4", "evidence": "Title 'Why now? Satellite data is decreasing in cost and increasing in resolution'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 86, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Before-After-Bridge", "slug": "before-after-bridge", "status": "active", "bestFor": null, "canonId": "019dd956-8604-7188-883f-42370288ad05", "version": 1, "bodyDocId": null, "description": "Narrative structure: Paint the before, show the after, explain the bridge", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "5ff146c7-6603-410a-bca8-85948b92aca3", "evidence": "The slide presents a before (high cost, low resolution) and after (decreasing cost, increasing resolution) scenario.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.673821+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 86, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 87, "type": "analysis", "title": "Part of the State of AI 2018 report.", "function": "summarize", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "dense", "densityScore": 72, "componentCount": 9, "textChars": 756, "nDataPoints": 0, "notes": "Part of the State of AI 2018 report.", "elementsJson": [ "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:03.298+00", "seconds": 2.725, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/87", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-87", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:23+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-6b9af8cc8ca4", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.6, "x": 0.08, "y": 0.78 }, "kind": "image", "text": "Selected examples: telluslabs, Descartes Labs, SATELLOGIC", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.298+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7c829446-41d4-4b53-9931-69b00d82936d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.4, "x": 0.08, "y": 0.58 }, "kind": "image", "text": "Selected examples: SPACEKNOW, GENSCAPE", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.298+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "94832af4-7679-4333-88ab-9af442d59aef", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.4, "x": 0.08, "y": 0.38 }, "kind": "image", "text": "Selected examples: CAI, VanderSat, Orbital Insight", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.298+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ac5adc95-1124-420a-8ef5-5b43eeb03226", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Insurance: Use real time imaging and historical data to automate claims, detect fraud and improve pricing models for property, catastrophe and crop insurance.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.298+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0ab2f937-42d6-416d-aede-652feedf28a2", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.48 }, "kind": "list", "text": "Finance: Automate assessment of ground truth data (traffic patterns, car counts in retail parking lots, drilling activity, construction activity etc) to find new sources of alpha in financial markets.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.298+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4d76bfa7-8e5f-4493-a5ec-9d3853848eac", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.68 }, "kind": "list", "text": "Agriculture: Use persistent daily imagery to monitor fields to understand changes in soil or crop health and forecast yields.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.298+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "923171b9-deb1-446a-b8a8-f3811a07aabc", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.05, "y": 0.22 }, "kind": "paragraph", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.298+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "14255718-486e-408d-9650-94ea7247fe6f", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.16, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Satellite data", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.298+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "96f2fae8-b2a2-4084-8aa6-faac40e93098", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "82ef5b33-954f-4892-9203-d1f9a092a312", "evidence": "The slide presents information in a list format with bullet points, making it a list presentation.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.753082+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 87, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 88, "type": "analysis", "title": "The slide showcases technical capabilities of satellite imagery providers (Planet, Orbital Insight, Descartes Labs) using RGB and NDVI analysis.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "sparse", "densityScore": 22, "componentCount": 5, "textChars": 186, "nDataPoints": 0, "notes": "The slide showcases technical capabilities of satellite imagery providers (Planet, Orbital Insight, Descartes Labs) using RGB and NDVI analysis.", "elementsJson": [ "photo", "infographic" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:02.914+00", "seconds": 2.327, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/88", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-88", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.68, "w": 0.92, "x": 0.05, "y": 0.23 }, "kind": "image", "text": "Satellite hardware and data visualization examples", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.914+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d702e972-1d99-4750-8237-84c61e1335ae", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.8, "x": 0.02, "y": 0.03 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.914+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1b269e15-6b74-4f87-9f1c-59a12162f14b", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.82, "y": 0.93 }, "kind": "source-note", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.914+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "40e1493a-5d7f-4193-80b2-66461c4da198", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.88, "y": 0.03 }, "kind": "source-note", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.914+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9f3b26c8-3215-4d04-83d3-cb7e3412a52a", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.35, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "Satellite data: Eyes in the sky", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:02.914+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d17c6294-5eb5-4125-ad92-f7cc553d9d05", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Narrative foundations (from the Storymakers methodology)", "slug": "narrative-foundations", "status": "active", "bestFor": null, "canonId": "019df269-258b-71bd-ad14-53598df04d2b", "version": 1, "bodyDocId": "019df269-2567-736e-8b9c-4ec77611334a", "description": "This skill focuses on the *adversarial* genre — activist campaigns, short", "familyLabel": "narrative-block", "categoryName": "Narrative & Transformation", "categorySlug": "narrative-transformation" }, "agent": "storyteller", "layer": "slide", "agents": [ "storyteller" ], "matchId": "cd5275ee-21bc-42c5-9967-c34f0fdc6500", "evidence": "The slide has a structured approach with Introduction, Research, Talent, Industry, Politics, Predictions, and Conclusion sections.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:35.834454+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 88, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 89, "type": "analysis", "title": "Part of the 'state.ai 2018' report series.", "function": "diagnose", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "diagnose", "functionCanon": { "name": "Diagnose", "slug": "diagnose", "status": "active", "canonId": "019de52d-1657-7223-acc5-7437d878c850", "version": 1, "description": null }, "density": "dense", "densityScore": 50, "componentCount": 6, "textChars": 482, "nDataPoints": 9, "notes": "Part of the 'state.ai 2018' report series.", "elementsJson": [ "line_chart", "area_chart", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:03.433+00", "seconds": 2.796, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/89", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-89", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Machine learning offers a flexible way to learn from past attacks and automate processes saving time for stretched security teams.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-162e62c3aeb3", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.43, "x": 0.52, "y": 0.45 }, "kind": "chart", "text": "% of organisations lacking cybersecurity skills", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Area chart", "slug": "area", "status": "active", "canonId": "019de52c-ff89-71e0-911e-5cab78b90f53", "version": 1, "description": null }, "framework": null }, "subkind": "area", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.433+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "251bd36c-1809-4f1f-842e-9c198fe72bbe", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.43, "x": 0.03, "y": 0.45 }, "kind": "chart", "text": "Global avg cost of cybercrime to organisations", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.433+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f80cf8ed-b70d-4e16-bb23-06069c929dd3", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Global avg cost of cybercrime: $11.7M", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-1bdddc09f909", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.94, "x": 0.03, "y": 0.22 }, "kind": "paragraph", "text": "Why now? Cloud computing, mobile devices, and more interconnected supply chains means the attack surface for cyber attacks is expanding. At the same time there is a growing shortage of cybersecurity personnel.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.433+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5e41f669-93f3-444b-b697-161e99a910cd", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.16, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "Cybersecurity", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.433+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4fc992bc-af70-498c-9365-c72b703d3fb3", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-ada4a571e7d8", "evidence": "Topic title 'Cybersecurity' with $11.7M cost stat.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 89, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "352e1766-28c9-45c4-87a8-7f47cd14488d", "evidence": "Cloud computing, mobile devices, and more interconnected supply chains means the attack surface for cyber attacks is expanding. At the same time there is a growing shortage of cybersecurity", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.254257+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 89, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Storytelling: audience calibration", "slug": "audience", "status": "active", "bestFor": null, "canonId": "019df269-23ea-723a-87c0-af1d78d702d2", "version": 1, "bodyDocId": "019df269-23c5-72ec-8b94-78fc1da7b656", "description": "The same thesis, written for four different audiences, produces four", "familyLabel": "narrative-block", "categoryName": "Narrative & Transformation", "categorySlug": "narrative-transformation" }, "agent": "storyteller", "layer": "slide", "agents": [ "storyteller" ], "matchId": "0faf014d-94fc-49f0-8af4-9d7195050ba3", "evidence": "Why now? Cloud computing, mobile devices, and more interconnected supply chains means the attack surface for cyber attacks is expanding.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.220822+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 89, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 90, "type": "analysis", "title": "Part of a larger report series; slide focuses on practical applications of ML in a specific industry vertical.", "function": "summarize", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "dense", "densityScore": 57, "componentCount": 7, "textChars": 642, "nDataPoints": 0, "notes": "Part of a larger report series; slide focuses on practical applications of ML in a specific industry vertical.", "elementsJson": [ "headline_text", "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:03.366+00", "seconds": 2.442, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/90", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-90", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-a8ac029521ad", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.7, "x": 0.08, "y": 0.45 }, "kind": "image", "text": "Darktrace, Jask, Deepinstinct, Cylance, Crowdstrike, Carbon Black", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.366+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d9f67e5f-c3db-4b00-a2e9-a835ff1ecf98", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.3, "x": 0.08, "y": 0.78 }, "kind": "image", "text": "StatusToday, Interset", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.366+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f4d962ba-e44d-4419-89f9-88d812cc5939", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.65 }, "kind": "list", "text": "Insider threat detection: Applying machine learning to large amounts of data on employee behaviour reduces the time to flag potential malicious intent.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.366+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1199247a-223c-4a16-91f6-acf4be62ba64", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Network and endpoint security: Supervised learning is used to detect malicious activity on an organisation’s network based on data from past attacks. Unsupervised learning is used to automatically learn what is normal and what is abnormal within a network on a an ongoing basis.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.366+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "11a0bad2-f0fd-4c14-8eba-7a177b94432b", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.03, "y": 0.22 }, "kind": "paragraph", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.366+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7fe10bee-cacd-4cfe-bbe6-41445c6538a1", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.03, "y": 0.15 }, "kind": "title", "text": "Cybersecurity", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:03.366+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "75b66691-465a-4259-93a8-38d0a5eed720", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "d8b08b24-dd68-48a2-82fe-3c9473939eac", "evidence": "Network and endpoint security: Supervised learning is used to detect malicious activity on an organisation’s network based on data from past attacks.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.323252+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 90, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 91, "type": "analysis", "title": "The slide uses a 'Why now?' framing to explain the rapid adoption of robotics in logistics.", "function": "diagnose", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "diagnose", "functionCanon": { "name": "Diagnose", "slug": "diagnose", "status": "active", "canonId": "019de52d-1657-7223-acc5-7437d878c850", "version": 1, "description": null }, "density": "dense", "densityScore": 64, "componentCount": 7, "textChars": 665, "nDataPoints": 7, "notes": "The slide uses a 'Why now?' framing to explain the rapid adoption of robotics in logistics.", "elementsJson": [ "bar_chart_vertical", "bar_chart_vertical", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:04.352+00", "seconds": 3.021, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/91", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-91", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "eCommerce growth decreases order size for item picking in warehouses and increases customer expectations around the speed of fulfilment.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-e5a8138e16ab", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.45, "x": 0.02, "y": 0.45 }, "kind": "chart", "text": "Number of robots working in Amazon fulfilment centres", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.352+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9a4b1605-5148-4029-8fde-51aadd696a2d", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.45, "x": 0.52, "y": 0.45 }, "kind": "chart", "text": "% of warehouse and logistics managers reporting inability to find hourly workers as a top concern", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.352+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a434879d-cf87-4eb7-9fb7-c102bb3ff3d6", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Number of robots: 120,000", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a813-e85cf189fbc4", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "Why now? eCommerce growth decreases order size for item picking in warehouses and increases customer expectations around the speed of fulfilment. Warehouse space and labour are both scarce driving more use of robotics. Retailers are also reacting to Amazon’s investment in this area following their acquisition of Kiva.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.352+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a288b159-5475-4dbd-93d7-b9bdca8cca11", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.88, "y": 0.95 }, "kind": "source-note", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.352+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9f1e057b-aaa4-4d46-957b-60468ce73ea4", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.25, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Warehouse automation", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.352+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8fdba1cb-c503-4e70-89ed-08ab03401808", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-b2a5dbb1945f", "evidence": "Topic 'Warehouse automation' with 120,000 figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 91, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Before-After-Bridge", "slug": "before-after-bridge", "status": "active", "bestFor": null, "canonId": "019dd956-8604-7188-883f-42370288ad05", "version": 1, "bodyDocId": null, "description": "Narrative structure: Paint the before, show the after, explain the bridge", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "45e90108-6056-4a85-886f-c284d027d9b1", "evidence": "Why now? eCommerce growth decreases order size for item picking in warehouses and increases customer expectations around the speed of fulfilment.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.391278+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 91, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [ { "matchId": "cb3bd79d-b718-4a68-8522-fa71e2c3132f", "evidence": "The slide uses a 'Why now?' structure to diagnose the drivers of a market trend.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:44:04.352+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 91, "frameworkName": "diagnostic-test" } ], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 92, "type": "analysis", "function": "present_solution", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "present_solution", "functionCanon": { "name": "Present solution", "slug": "present_solution", "status": "active", "canonId": "019de52d-15b8-71da-bd23-2b767850cbce", "version": 1, "description": null }, "density": "dense", "densityScore": 51, "componentCount": 7, "textChars": 453, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text", "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:04.161+00", "seconds": 2.611, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/92", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-92", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:21+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-0290c32125d6", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.35, "x": 0.08, "y": 0.6 }, "kind": "image", "text": "Ocado, Zalando", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.161+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "158504fa-8035-442e-b5a9-884da4dc6eb9", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.85, "x": 0.08, "y": 0.35 }, "kind": "image", "text": "Amazon Robotics, 6 River Systems, GreyOrange, RightHand Robotics, JD.com", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.161+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1ef8cad7-c317-4c8f-9ee8-c13330164e9a", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.65, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Robotics: Using robots and drones for picking, packing, inventory inspection.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.161+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "15945184-8494-44a2-94e3-74582bf8669c", "parentComponentId": null }, { "bbox": { "h": 0.07, "w": 0.85, "x": 0.05, "y": 0.48 }, "kind": "list", "text": "Warehouse management systems: Using machine learning within warehouse management software to optimise inventory, order picking and queues to minimise waste.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.161+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d164579c-167d-4c35-b3bc-b1f7bd8f26ad", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.03, "y": 0.22 }, "kind": "paragraph", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.161+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4c592b6d-e0c6-4149-aab9-5ae1837008e1", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.25, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "Warehouse automation", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.161+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3ad8086a-c829-484e-a6e1-8698031138c2", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "2b09dfec-b6c0-4a16-b5e9-ee68bf7a6997", "evidence": "Warehouse management systems: Using machine learning within warehouse management software to optimise inventory, order picking and queues to minimise waste.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.458999+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 92, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 93, "type": "analysis", "title": "The slide showcases GreyOrange, RightHand Robotics, and 6 River Systems.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 38, "componentCount": 7, "textChars": 171, "nDataPoints": 0, "notes": "The slide showcases GreyOrange, RightHand Robotics, and 6 River Systems.", "elementsJson": [ "photo", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:04.578+00", "seconds": 2.828, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/93", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-93", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.05, "w": 0.25, "x": 0.66, "y": 0.72 }, "kind": "image", "text": "6 River Systems logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.578+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "110d9f21-c2b0-44f8-bc3e-04d483943c5f", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.42, "y": 0.72 }, "kind": "image", "text": "RightHand Robotics logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.578+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "54609a17-b320-48c3-a855-e77df99392c4", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.18, "x": 0.09, "y": 0.72 }, "kind": "image", "text": "GreyOrange logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.578+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "66255c5e-752d-4cb8-acda-c92090352e7f", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.3, "x": 0.35, "y": 0.336 }, "kind": "image", "text": "RightHand Robotics arm", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.578+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "22fdc473-fbed-477b-b063-5bc556826817", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.3, "x": 0.03, "y": 0.336 }, "kind": "image", "text": "GreyOrange robot", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.578+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2a04656c-ed8d-4f3a-a787-a3d37ec8a235", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.3, "x": 0.67, "y": 0.336 }, "kind": "image", "text": "6 River Systems robot", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.578+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b2d96ef2-f505-4f24-8752-77cb7a13e047", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Warehouse automation: Products of all shapes and sizes", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.578+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b2dd8efe-f3dc-4831-8e1d-4d0cf6334157", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 94, "type": "analysis", "title": "The slide uses two charts to explain the economic rationale for increased automation in blue-collar sectors.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "overcrowded", "densityScore": 75, "componentCount": 8, "textChars": 722, "nDataPoints": 30, "notes": "The slide uses two charts to explain the economic rationale for increased automation in blue-collar sectors.", "elementsJson": [ "bar_chart_horizontal", "line_chart", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.506+00", "seconds": 3.663, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/94", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-94", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "A decrease in the cost of components (sensors, batteries) and improvements in computer vision mean that robots are increasingly cheaper than employing manual labour for various blue collar professions.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-2601cf54bbbc", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.4, "x": 0.05, "y": 0.42 }, "kind": "chart", "text": "Automation risk by job type (%)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.506+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "63a197a9-5e10-4886-a1cb-c58d4c4f7b74", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.4, "x": 0.52, "y": 0.42 }, "kind": "chart", "text": "Index of average robot prices and labour compensation in U.S. manufacturing", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.506+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8671610d-06c4-454c-a110-c9c60e8b136b", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Automation risk", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-2b6ba19f4f71", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.24 }, "kind": "paragraph", "text": "A decrease in the cost of components (sensors, batteries) and improvements in computer vision mean that robots are increasingly cheaper than employing manual labour for various blue collar professions.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.506+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b4aa262b-a353-429d-bb76-b656670b9464", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.4, "x": 0.52, "y": 0.9 }, "kind": "source-note", "text": "Source: Economist Intelligence Unit; IMB; Institut für Arbeitsmarkt- und Berufsforschung; International Robot Federation; US Social Security data; McKinsey analysis", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.506+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b81825e3-30da-4cfc-856f-e50595355889", "parentComponentId": null }, { "bbox": { "h": 0.02, "w": 0.1, "x": 0.05, "y": 0.88 }, "kind": "source-note", "text": "Source: OECD", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.506+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d1afa28a-0a69-42e2-a097-cd2f81645aed", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.28, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Blue collar manual work", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.506+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "69f56ec3-6822-4477-b6be-50c2db0fe424", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-b4fc28da99bf", "evidence": "Callout drives why-now: sensor cost down, vision better, robots cheaper than labour.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 94, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "98bacd4d-71eb-449c-9592-b77dd0bce44e", "evidence": "Index of average robot prices and labour compensation in U.S. manufacturing", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.52678+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 94, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 95, "type": "analysis", "title": "Part of a larger report series; uses a list-based structure to categorize industry applications.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 71, "componentCount": 9, "textChars": 704, "nDataPoints": 0, "notes": "Part of a larger report series; uses a list-based structure to categorize industry applications.", "elementsJson": [ "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:04.856+00", "seconds": 2.908, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/95", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-95", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:26+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-b2ad54370d7a", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.8, "x": 0.08, "y": 0.82 }, "kind": "image", "text": "Cobalt, Accelerated Dynamics, Aptonomy, Nightingale Security", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.856+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "29de0bab-e6a3-4e42-81f9-fe5c0f9c6b78", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.8, "x": 0.08, "y": 0.58 }, "kind": "image", "text": "Avidbots, Serbot, Xenex, Brain Corporation", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.856+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4ba15b32-7bfd-46cf-8557-356780fdb4fd", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.7, "x": 0.08, "y": 0.35 }, "kind": "image", "text": "Built Robotics, Construction Robotics, Clearpath Robotics", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.856+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d1a1059e-f44c-4ac5-90ca-e37b19f5feb8", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.48 }, "kind": "list", "text": "Cleaning: Self-driving cleaning robots for industrial spaces. This can include dangerous or hard to access spaces like windows, solar panels or infectious spaces.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.856+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6b7b0045-455b-4fa7-be9e-e9c547286133", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.72 }, "kind": "list", "text": "Security: Computer vision applied to security cameras combined with drones, robots and other sensors to replace aspects of a security guard's job.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.856+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bac1dd29-0640-4d33-93f8-516eabbe22e2", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Construction: Self-driving vehicles for digging and loading. Robots for bricklaying and other tasks.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.856+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c33e62f4-1d30-4eb0-988c-c71c873a9afb", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.856+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8691765e-81cd-4cd1-b1f9-5d07109b85cc", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.3, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Blue collar manual work", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.856+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "85f0d120-5946-421e-86b0-449ebe060c2a", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "9943f72f-f19e-4a15-8110-30eb7fbafc28", "evidence": "list/bullet: Cleaning: Self-driving cleaning robots for industrial spaces.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-17 03:20:35.043489+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 95, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 96, "type": "evidence", "function": "illustrate_case", "rawType": "client_example", "block": null, "metadata": { "slideType": "client_example", "slideTypeCanon": { "name": "Client example", "slug": "client_example", "status": "active", "canonId": "019de52d-0c87-77cb-9867-613ae22e95ac", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 39, "componentCount": 7, "textChars": 197, "nDataPoints": 0, "notes": null, "elementsJson": [ "photo", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:04.544+00", "seconds": 2.528, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/96", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-96", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.05, "w": 0.15, "x": 0.1, "y": 0.7 }, "kind": "image", "text": "Built Robotics logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.544+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "179b94a8-2fe1-4c99-a382-0b63b5271b3f", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.45, "y": 0.7 }, "kind": "image", "text": "Clearpath Robotics logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.544+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2c1bf340-526b-4ca2-95de-833e0427e392", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.75, "y": 0.7 }, "kind": "image", "text": "Avidbots logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.544+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f915eb71-1b74-4538-be37-68657e8d8399", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.3, "x": 0.02, "y": 0.295 }, "kind": "image", "text": "Built Robotics construction vehicle", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.544+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0734d9ae-59bd-4ac0-858b-9224b5ded3ad", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.3, "x": 0.35, "y": 0.295 }, "kind": "image", "text": "Clearpath Robotics vehicle", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.544+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0f53bb28-62a4-448b-8997-9a01a821db21", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.3, "x": 0.68, "y": 0.295 }, "kind": "image", "text": "Avidbots floor cleaner", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.544+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5ee831c4-6348-4337-b63d-6f94a1d3719d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Blue collar manual work: Examples of products in the market", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:04.544+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ce1d9564-73db-43d6-8217-5aa06f153ed4", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 97, "type": "analysis", "title": "The slide uses a combination of a line chart showing the food production gap and a pie chart showing investment distribution.", "function": "frame_problem", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "frame_problem", "functionCanon": { "name": "Frame problem", "slug": "frame_problem", "status": "active", "canonId": "019de52d-13db-753e-ad70-e9b2d0dab398", "version": 1, "description": null }, "density": "dense", "densityScore": 51, "componentCount": 6, "textChars": 531, "nDataPoints": 6, "notes": "The slide uses a combination of a line chart showing the food production gap and a pie chart showing investment distribution.", "elementsJson": [ "headline_text", "paragraph", "line_chart", "pie_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:07.801+00", "seconds": 5.62, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/97", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-97", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "We need to produce 70% more food calories to feed the world’s population by then.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-f8db5c4a8cbf", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.4, "x": 0.03, "y": 0.48 }, "kind": "chart", "text": "The need for boosting food production", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Area chart", "slug": "area", "status": "active", "canonId": "019de52c-ff89-71e0-911e-5cab78b90f53", "version": 1, "description": null }, "framework": null }, "subkind": "area", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.801+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6107e039-a018-4be6-9de8-025aa24cced5", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.4, "x": 0.55, "y": 0.48 }, "kind": "chart", "text": "Farms are investing in technology now", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Pie chart", "slug": "pie", "status": "active", "canonId": "019de52c-ffdb-71bb-a800-5c5ab9f1c373", "version": 1, "description": null }, "framework": null }, "subkind": "pie", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.801+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a1c0538b-cf7f-4036-a0fc-0e4f1651d085", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Food production gap: 70%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-fe209fa01a89", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.94, "x": 0.03, "y": 0.28 }, "kind": "paragraph", "text": "The world population is expected to grow from 7.6 billion to 9.6 billion by 2050. We need to produce 70% more food calories to feed the world's population by then. Robotics, control systems, connected devices in fields and greenhouses and new methods of farming must be developed to fill this food production gap.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.801+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "047fb7e4-16de-4e37-b23f-68759ece7a2f", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.03, "y": 0.15 }, "kind": "title", "text": "Agriculture: Indoor and outdoor farming", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.801+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dc6d6af4-3694-4067-91ad-ddb1a9789cfd", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-bb2599bcb4f3", "evidence": "Callout '70% more food calories needed by 2050'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 97, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data Story Arc", "slug": "data-story-arc", "status": "active", "bestFor": null, "canonId": "019dd956-9c30-728e-b974-d3fb123dcf5a", "version": 1, "bodyDocId": null, "description": "Context → Conflict → Insight → Implication structure for data narratives", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "9893e71e-aee2-4549-b04e-245cbbc5ab79", "evidence": "The world population is expected to grow from 7.6 billion to 9.6 billion by 2050. We need to produce 70% more food calories to feed the world's population by then.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:35.593063+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 97, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 98, "type": "analysis", "title": "Slide from state.ai 2018 report.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 64, "componentCount": 7, "textChars": 574, "nDataPoints": 60, "notes": "Slide from state.ai 2018 report.", "elementsJson": [ "bar_chart_stacked", "line_chart", "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.561+00", "seconds": 3.098, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/98", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-98", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Fewer farm workers on US farms, higher hourly wages per worker, but more automation leads to stable labor cost share of total gross farm revenue", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-801156e9c90c", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.32, "x": 0.02, "y": 0.38 }, "kind": "chart", "text": "Family and hired farmworkers on U.S. farms, 1950-2000", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked bar chart", "slug": "bar-stacked", "status": "active", "canonId": "019de52c-feec-74bf-9028-15b74f84f0a7", "version": 1, "description": null }, "framework": null }, "subkind": "bar-stacked", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.561+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a8d426c2-23c4-4338-a40e-e166ce5b6ebe", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.3, "x": 0.68, "y": 0.38 }, "kind": "chart", "text": "Labor costs as a share of total gross cash income for selected farm specializations, 2000-16", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.561+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3f48540f-0d8b-4268-8180-de0e61b3fefa", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.32, "x": 0.35, "y": 0.38 }, "kind": "chart", "text": "Real hourly wages for hired farmworkers, all agricultural workers, and AEWR, 1989-2017", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.561+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7a43050c-7b36-4ae7-8446-a36f71d05d0a", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Labor cost share", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-845e5bc9062d", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Fewer farm workers on US farms, higher hourly wages per worker, but more automation leads to stable labor cost share of total gross farm revenue", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.561+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "031e6b82-88fb-48cf-8dd2-d7c72f8b2694", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.4, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Agriculture: Indoor and outdoor farming", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.561+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "af8a26ce-9ff1-443e-b28b-de1d2d17642a", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-bf6cc421d2de", "evidence": "Title 'Fewer farm workers... higher wages... but more automation leads to stable labor cost share'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 98, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chartjunk Elimination", "slug": "chartjunk-elimination", "status": "active", "bestFor": null, "canonId": "019dd956-b49e-7453-ab01-b60f338eeda1", "version": 1, "bodyDocId": null, "description": "Remove decorative elements that don't convey data: 3D effects, shadows, unnecessary gridlines", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "450d2da4-c7aa-429c-ad7b-4821873c4138", "evidence": "Labor costs as a share of total gross cash income for selected farm specializations, 2000-16", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:35.660088+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 98, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 99, "type": "analysis", "title": "Part of the State of AI 2018 report.", "function": "summarize", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "dense", "densityScore": 73, "componentCount": 9, "textChars": 787, "nDataPoints": 0, "notes": "Part of the State of AI 2018 report.", "elementsJson": [ "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:06.218+00", "seconds": 3.61, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/99", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-99", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-b084fe528fa1", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.1, "x": 0.26, "y": 0.37 }, "kind": "image", "text": "OPTIMAL", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "81751f91-db2d-4283-86a7-2f6d555611d6", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.35, "x": 0.26, "y": 0.57 }, "kind": "image", "text": "infarm, Plenty, AGRICOOL", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ef0ed28b-dd41-41fd-9d86-854ec778bb24", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.45, "x": 0.26, "y": 0.76 }, "kind": "image", "text": "prospera, cainthus, Connecterra", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ef11c0d8-e338-4972-8e88-95e0f2b865bf", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.93, "x": 0.05, "y": 0.48 }, "kind": "list", "text": "Vertically-integrated farming: Compact, self-contained greenhouses for growing crops closer to the point of consumption. The farms have their climates that can be operated using similar ML-driven control systems.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2a98882c-1fba-474d-8b85-7a7d2cabd2d9", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.93, "x": 0.05, "y": 0.68 }, "kind": "list", "text": "Health inspection for crops and animals: Use computer vision and wearable sensors to learn models of plant and animal health and use them to detect anomalies.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4fc66b59-8295-4596-9415-01f6e9777477", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.93, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Greenhouse control systems: Use native sensors and actuators in greenhouses to collect data on growing conditions, learn a dynamic climate model and use it to optimise crop yield and energy consumption.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7ad22a9d-9428-48a6-9744-03832c174254", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.53, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2cde1bee-7777-4217-b53b-bb851387bc3e", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Agriculture: Indoor and outdoor farming", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "914c3bba-ae53-4f10-b4b5-2b52285479ed", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "4c6c8567-514d-4aee-9690-12dae8cc0d40", "evidence": "Vertically-integrated farming: Compact, self-contained greenhouses for growing crops closer to the point of consumption.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:35.726501+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 99, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 100, "type": "analysis", "title": "Includes company logos and photographic examples of agricultural robotics.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 61, "componentCount": 8, "textChars": 536, "nDataPoints": 0, "notes": "Includes company logos and photographic examples of agricultural robotics.", "elementsJson": [ "headline_text", "bullet_list", "logo_grid", "photo" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.129+00", "seconds": 2.462, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/100", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-100", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Crop picking robots: Build robots capable of mapping and navigating through crop fields while identifying and carefully picking ripe fruit automatically.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-85a14598a5d1", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.65, "x": 0.08, "y": 0.38 }, "kind": "image", "text": "Selected examples: Dogtooth, Abundant Robotics, FF Robotics, AGROBOT", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.129+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f66cc45d-ebc5-4c39-b448-0a493a132e15", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.26, "x": 0.35, "y": 0.48 }, "kind": "image", "text": "Abundant Robotics gripper", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.129+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2cd2f066-29ea-4b5d-bfca-c9a8635e64f0", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.26, "x": 0.07, "y": 0.48 }, "kind": "image", "text": "Dogtooth robot in field", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.129+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "488cb83c-d69d-49c5-aa73-904911326c95", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.26, "x": 0.65, "y": 0.48 }, "kind": "image", "text": "Infarm indoor unit", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.129+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e8cf00ca-1db2-4816-8b92-6e325d63f4b0", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Crop picking robots: Build robots capable of mapping and navigating through crop fields while identifying and carefully picking ripe fruit automatically.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.129+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "edaf6df5-55e6-4ba4-83f0-b1ddd8920934", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.5, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.129+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b7968be3-e883-4471-8f8d-e824687d8454", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Agriculture: Indoor and outdoor farming", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.129+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "88fa2671-a1fa-44e3-9224-f7fcac6fb035", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "8742df94-a17e-41ef-b0a1-bc52d2b4a7ae", "evidence": "components: list/bullet: Crop picking robots: Build robots capable of mapping and navigating through crop fields while identifying and carefully picking ripe fruit automatically.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-17 03:20:48.623151+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 100, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 101, "type": "tension", "title": "The slide uses a structured layout to present key statistics justifying the need for autonomous technology.", "function": "frame_problem", "rawType": "problem_statement", "block": null, "metadata": { "slideType": "problem_statement", "slideTypeCanon": { "name": "Problem statement", "slug": "problem_statement", "status": "active", "canonId": "019de52d-0879-768d-bba8-8da8a5a121d4", "version": 1, "description": null }, "function": "frame_problem", "functionCanon": { "name": "Frame problem", "slug": "frame_problem", "status": "active", "canonId": "019de52d-13db-753e-ad70-e9b2d0dab398", "version": 1, "description": null }, "density": "overcrowded", "densityScore": 81, "componentCount": 13, "textChars": 620, "nDataPoints": 9, "notes": "The slide uses a structured layout to present key statistics justifying the need for autonomous technology.", "elementsJson": [ "headline_text", "icon_grid", "other" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:06.534+00", "seconds": 3.755, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/101", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-101", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.15, "w": 0.15, "x": 0.3, "y": 0.55 }, "kind": "metric", "text": "$594 billion in harm from loss of life and injury each year", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "09bea732-33e5-40df-a9a3-698118ddc9c4", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.6, "y": 0.8 }, "kind": "metric", "text": "42 hours wasted in traffic each year per person", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1679947b-c433-44e3-a96d-4c6ebd36cffa", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.6, "y": 0.55 }, "kind": "metric", "text": "$160 billion in gas burned and time lost each year", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "17efacff-fe86-486b-9984-dd6ac7ea0947", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.75, "y": 0.3 }, "kind": "metric", "text": "2 out of 3 people will be involved in a drunk driving crash in their lifetime", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "18698c26-7ea5-43ea-b44d-7c9af472b25d", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.3, "y": 0.3 }, "kind": "metric", "text": "94 percent of U.S. crashes involve human error", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2d1857bc-0d8e-4966-84d1-3c66b8fbd2ed", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.45, "y": 0.8 }, "kind": "metric", "text": "79 percent of seniors age 65 and older living in car-dependent communities", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8095589b-29e3-401c-859e-13e0f4ac7411", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.45, "y": 0.55 }, "kind": "metric", "text": "$277 billion in annual economic costs", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a514b3a2-0191-4206-85ec-cc76f97a8588", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.45, "y": 0.3 }, "kind": "metric", "text": "1.2 million deaths worldwide due to vehicle crashes in 2013", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cc9ac992-a766-46d7-9d43-b57022f0dbcc", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.3, "y": 0.8 }, "kind": "metric", "text": "3 million Americans age 40 and older are blind or have low vision", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d8a9140c-0048-479d-812d-d077a1a602aa", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.15, "x": 0.6, "y": 0.3 }, "kind": "metric", "text": "37,461 road deaths in the U.S. in 2016 and 2.4 million injuries in 2015", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": null, "subkind": { "name": "Big number", "slug": "big-number", "status": "active", "canonId": "019de52c-fc70-70c8-b965-0584548a9204", "version": 1, "description": null }, "framework": null }, "subkind": "big-number", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ee0f1ed2-39ab-45aa-bab3-0c45a74ac71b", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Road deaths: 37,461", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-10d85e09a885", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.1, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Why now?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "09d532dc-724e-4df6-8bbf-9b4feb15cce3", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Autonomy", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.534+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "87892278-86c8-4ee4-be77-bf2b64a2ff15", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "7e2e35e9-af2e-4824-b1a9-113a8e0e3dc4", "evidence": "The slide presents a series of big numbers and metrics related to road safety, which is a typical structure for a problem statement.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.778221+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 101, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Storytelling Effect", "slug": "storytelling-effect", "status": "active", "bestFor": null, "canonId": "019dd956-981b-74f9-9096-5c8df03dab5a", "version": 1, "bodyDocId": null, "description": "Stories are remembered up to 22x better than facts alone", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "d3579080-887a-4fcb-a338-8770b3f346aa", "evidence": "The use of multiple big numbers and metrics creates a narrative around the problem of road safety.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.811698+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 101, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 102, "type": "analysis", "title": "Part of the 'state of AI 2018' report.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 72, "componentCount": 9, "textChars": 750, "nDataPoints": 0, "notes": "Part of the 'state of AI 2018' report.", "elementsJson": [ "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:06.407+00", "seconds": 3.571, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/102", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-102", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-563feedcb5aa", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.85, "x": 0.08, "y": 0.57 }, "kind": "image", "text": "Starship, Nuro, Amazon Prime Air, JD.com, Matternet", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.407+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "af1563ec-8e05-4252-84b5-aebbcce02549", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.85, "x": 0.08, "y": 0.37 }, "kind": "image", "text": "Waymo, Lyft, Cruise, Uber, Five AI, DiDi", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.407+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ee8a2466-0284-471f-8fcb-644b14324598", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.85, "x": 0.08, "y": 0.78 }, "kind": "image", "text": "Mapillary, Aurora, Unity, Nvidia, Oxbotica", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.407+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f4af1301-4d87-4cb1-9082-b6c3154d043c", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.92, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Autonomous vehicle ridesharing: Machine learning is often used across the entire stack from perception, localisation, mapping, planning, control, route optimisation and safety.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.407+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "17d59cb8-e83c-4333-b7da-fe4bbd6eeab8", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.92, "x": 0.05, "y": 0.49 }, "kind": "list", "text": "Autonomous last mile delivery: Same as above, except these vehicles are used to delivery goods locally by land or air.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.407+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "253c5a57-cd52-4bf7-9db6-410266fb231d", "parentComponentId": null }, { "bbox": { "h": 0.07, "w": 0.92, "x": 0.05, "y": 0.69 }, "kind": "list", "text": "Simulation environments, street level maps and software for autonomy: Using a mix of machine learning, computer vision, video game environments, photorealistic data generation and behavioral modelling.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.407+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "462ee95f-7923-4656-a21c-e276fb8e6441", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.03, "y": 0.22 }, "kind": "paragraph", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.407+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4537c70e-978a-4c04-a5e0-058be252b71f", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.12, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "Autonomy", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.407+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "aba8609e-746e-4f34-b169-cc5ce40377a3", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "1ba85ea4-1eca-4d61-af66-9e735195e3b9", "evidence": "The slide presents information in a list format with bullet points.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.876817+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 102, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 103, "type": "analysis", "title": "The slide uses a navigation bar at the top and a report identifier at the bottom right.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 37, "componentCount": 7, "textChars": 160, "nDataPoints": 0, "notes": "The slide uses a navigation bar at the top and a report identifier at the bottom right.", "elementsJson": [ "photo", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:10.481+00", "seconds": 7.638, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/103", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-103", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.08, "w": 0.15, "x": 0.75, "y": 0.7 }, "kind": "image", "text": "Baidu logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.481+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5d3aecab-c162-4990-90aa-d775d62f398d", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.15, "x": 0.43, "y": 0.7 }, "kind": "image", "text": "Waymo logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.481+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b01dfedb-a6ba-43c7-be37-028c1cc57af4", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.15, "x": 0.1, "y": 0.7 }, "kind": "image", "text": "Cruise logo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo", "slug": "logo", "status": "active", "canonId": "019de52c-fdaf-743d-94c4-d7c6a0e55302", "version": 1, "description": null }, "framework": null }, "subkind": "logo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.481+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ce14fcea-04c1-4d7c-80b6-9eccd83011ea", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.3, "x": 0.02, "y": 0.33 }, "kind": "image", "text": "Cruise autonomous vehicle", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.481+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "210f9639-ad8b-416e-b498-b463e35e52db", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.3, "x": 0.68, "y": 0.33 }, "kind": "image", "text": "Baidu autonomous vehicle", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.481+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2dc4e08a-2d12-4fd8-8d15-9df7210de0e3", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.3, "x": 0.35, "y": 0.33 }, "kind": "image", "text": "Waymo software visualization", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Screenshot", "slug": "screenshot", "status": "active", "canonId": "019de52c-fd87-7509-8391-3d79fb339379", "version": 1, "description": null }, "framework": null }, "subkind": "screenshot", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.481+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6a490390-beb6-412c-876b-9d7465afe693", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Autonomy: Vehicles and software products in the wild", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.481+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ac751393-cbe2-4843-9ac7-353d81db1088", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Slide visual craft", "slug": "visual-craft", "status": "active", "bestFor": null, "canonId": "019df269-2d56-7739-ab65-cc2e091a109b", "version": 1, "bodyDocId": "019df269-2d32-736a-ae59-0250c5540c9a", "description": "What separates a 5/5-visual-craft deck from a 3/5 one.", "familyLabel": "slide-recipe", "categoryName": "Slide", "categorySlug": "slide" }, "agent": "designer", "layer": "slide", "agents": [ "designer" ], "matchId": "b78c6fd2-6bcc-4dad-8e49-aaa3a6fce2e4", "evidence": "The slide uses images and logos to visually present information.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.945092+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 103, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 104, "type": "analysis", "title": "Part of the state.ai 2018 report.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 60, "componentCount": 7, "textChars": 482, "nDataPoints": 30, "notes": "Part of the state.ai 2018 report.", "elementsJson": [ "paragraph", "line_chart", "line_chart", "bar_chart_horizontal" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.207+00", "seconds": 2.228, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/104", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-104", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "There is an abundance of data in public markets, alternative sources and about users of financial products.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:30+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-44b7500e2898", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.25, "x": 0.72, "y": 0.45 }, "kind": "chart", "text": "Big data sources in use", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.207+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "236cf848-e178-496c-8732-cc4423ee05c7", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.3, "x": 0.03, "y": 0.45 }, "kind": "chart", "text": "Algorithmic trading as % of all trading", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.207+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "46da0d6f-af2f-4f8e-82ff-2f696ffb7767", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.3, "x": 0.38, "y": 0.45 }, "kind": "chart", "text": "Fraction of wealth lost to fees", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.207+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f07ad821-ad2b-47e7-80e5-f6e9e7b84867", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Fraction of wealth lost to fees: 7x", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:30+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-4b51733018e1", "parentComponentId": null }, { "bbox": { "h": 0.12, "w": 0.96, "x": 0.02, "y": 0.23 }, "kind": "paragraph", "text": "Why now? There is an abundance of data in public markets, alternative sources and about users of financial products. Moreover, consumers and investors are fatigued with overbearing fees to manage capital and provide products such as credit.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.207+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dc4b485a-8979-4b2c-88a3-359188bb385b", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.1, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Finance", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.207+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b69612f0-c5b5-441c-b394-dfd2fdd41c75", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-c2521190d621", "evidence": "Topic 'Finance' with 7x stat.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 104, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "3bdaa9cf-9c31-4d05-8371-d769bc943ff0", "evidence": "The slide includes various charts, such as line and bar charts.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.013838+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 104, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 105, "type": "analysis", "title": "Part of a larger report series (state.ai 2018).", "function": "present_solution", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "present_solution", "functionCanon": { "name": "Present solution", "slug": "present_solution", "status": "active", "canonId": "019de52d-15b8-71da-bd23-2b767850cbce", "version": 1, "description": null }, "density": "overcrowded", "densityScore": 75, "componentCount": 9, "textChars": 925, "nDataPoints": 0, "notes": "Part of a larger report series (state.ai 2018).", "elementsJson": [ "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:06.601+00", "seconds": 3.56, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/105", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-105", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-4208817ec939", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.08, "y": 0.4 }, "kind": "image", "text": "Wealthfront, Betterment, Nutmeg, Personal Capital", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.601+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2235abe3-aa12-4ffe-b08f-3b928514bf00", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.08, "y": 0.6 }, "kind": "image", "text": "Avant, Affirm, SoFi, Zopa, LendingClub", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.601+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a2e72ae1-c350-44d1-b961-e53e142cba5f", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.08, "y": 0.78 }, "kind": "image", "text": "Tractable, Ravelin, Signifyd, Shift Technology, Sift Science", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.601+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f82b6f1b-8614-40af-8f01-c529cfc0e46a", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.48 }, "kind": "list", "text": "Credit/Loans: The cost of calculating and underwriting risk is improved through automation and the discovery of novel features through machine learning that improve the overall efficiency of this process. Peer to peer lending has also benefited from these drivers.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.601+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "915d4d97-26e2-42f1-90cf-b70c1e78681d", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Wealth management: Software-driven automation of capital management, portfolio construction and tax optimisation. These services materially reduce the fees for consumers to invest their long-term savings.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.601+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9319bbb0-6563-4199-8877-6311cf2bdb52", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.68 }, "kind": "list", "text": "Fraud prevention: Using both supervised and unsupervised learning to detect known and novel fraudulent behaviors in electronic transactions, interpersonal communications, and claims images.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.601+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fe35ec81-e4e8-4a79-ae28-ed8f3d399728", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.02, "y": 0.23 }, "kind": "title", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.601+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f1d9e9da-0b8a-45af-afd4-3cc12d03601d", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.1, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Finance", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.601+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c6456627-f533-4b03-89b7-bddec424c240", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "332b462c-09d5-4236-9834-5d4d7f04a977", "evidence": "The slide presents information in a list format with bullet points.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.080904+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 105, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 106, "type": "analysis", "title": "The slide uses two charts to illustrate the inefficiency of current work patterns.", "function": "frame_problem", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "frame_problem", "functionCanon": { "name": "Frame problem", "slug": "frame_problem", "status": "active", "canonId": "019de52d-13db-753e-ad70-e9b2d0dab398", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 7, "textChars": 359, "nDataPoints": 18, "notes": "The slide uses two charts to illustrate the inefficiency of current work patterns.", "elementsJson": [ "paragraph", "bar_chart_horizontal", "bar_chart_grouped" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.921+00", "seconds": 2.827, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/106", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-106", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.1, "w": 0.15, "x": 0.8, "y": 0.75 }, "kind": "callout", "text": "1/3 day is structured, predictable, automated or automatable", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.921+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a302ff77-6bf1-4968-a967-f6185685ffb6", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.15, "x": 0.8, "y": 0.55 }, "kind": "callout", "text": "2/3 day is unstructured or unpredictable", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.921+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a777c19f-551f-495e-baf3-37cb4867f799", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.35, "x": 0.45, "y": 0.45 }, "kind": "chart", "text": "% of day spent in different modes of work", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Grouped bar chart", "slug": "bar-grouped", "status": "active", "canonId": "019de52c-ff3b-751e-a8ff-f21ff11097b1", "version": 1, "description": null }, "framework": null }, "subkind": "bar-grouped", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.921+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e298de9e-7304-45dc-9227-1d6e32e0a0c3", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.35, "x": 0.05, "y": 0.45 }, "kind": "chart", "text": "% of average week spent on tasks", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.921+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d991e3c1-ba1c-4697-a71e-cc6cf2e118e7", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "percentage of time: 2/3", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-cb21553af02d", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.25 }, "kind": "paragraph", "text": "Why now? Reducing operational process cost and complexity through software-defined automation is now a Board-level priority in the enterprise.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.921+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e46b39c9-e597-4975-904a-17c28a3d8544", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.25, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Enterprise automation", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.921+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "61b5812b-f918-45ef-bff4-2301bdfb12d3", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-c748ca9fd3b9", "evidence": "Callout '2/3 day is unstructured or unpredictable' frames automation gap.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 106, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "61cfaeb9-488a-4cff-89d2-816360663dc1", "evidence": "The slide includes various charts, such as bar and line charts.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.181991+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 106, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "7f9694e2-e18b-4cdf-a949-c72cf8a686c7", "evidence": "The slide presents a problem statement around enterprise automation.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.149179+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 106, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 107, "type": "analysis", "title": "The slide uses a list-based structure to categorize automation technologies.", "function": "present_framework", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "present_framework", "functionCanon": { "name": "Present framework", "slug": "present_framework", "status": "active", "canonId": "019de52d-1403-7667-9100-d2ac6709f750", "version": 1, "description": null }, "density": "overcrowded", "densityScore": 75, "componentCount": 9, "textChars": 934, "nDataPoints": 0, "notes": "The slide uses a list-based structure to categorize automation technologies.", "elementsJson": [ "bullet_list", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:06.532+00", "seconds": 3.31, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/107", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-107", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-9dad7e193ebf", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.8, "x": 0.08, "y": 0.44 }, "kind": "image", "text": "UiPath, BluePrism, Automation Anywhere, WorkFusion", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.532+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6bf43231-7333-483f-8300-863da5052a53", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.6, "x": 0.08, "y": 0.82 }, "kind": "image", "text": "Zapier, MuleSoft, IFTTT, tray.io", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.532+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8419f933-2308-4efd-9d91-3ced3bc7b038", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.6, "x": 0.08, "y": 0.63 }, "kind": "image", "text": "Hyperscience, omni:us, Ripcord", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.532+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c6b6285c-0c8c-454f-88c2-09cedcf142ae", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.72 }, "kind": "list", "text": "Software task automation: While not using machine learning per se, the ability to connect different API-driven software together to form workflows enables task automation in cloud-based enterprises.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.532+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "50928387-6d2d-43c5-b8ba-893a43b0a02c", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.9, "x": 0.05, "y": 0.53 }, "kind": "list", "text": "Document digitisation: Converting legacy paper documentation into digital records to simplify and automate office work. Based on (semi) supervised computer vision and optical character recognition.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.532+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7ca8a289-2e40-44f3-a7c0-34a081ef449d", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.28 }, "kind": "list", "text": "Robotic process automation: Creating automated “software robots” to replicate the repetitive desktop-based processes that human workers are otherwise doing. Computer vision and NLP can be used to understand what’s on the screen and flexible decision making will help solve more complex tasks.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.532+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e39f1202-e849-499b-b8b0-7564b8f0ac6f", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Where and how is machine learning being used effectively?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.532+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "045fa520-cd7c-4b59-8eba-86670279358a", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.25, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Enterprise automation", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.532+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bd2f3141-5a56-4204-b779-f68061eefad2", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "a24354e6-7cac-4718-b88d-52db0e133339", "evidence": "The slide presents information in a list format with bullet points.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.248672+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 107, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 108, "type": "analysis", "title": "The slide uses a 'Why now?' and 'Where and how is ML being used effectively?' structure to frame the industry trend.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 63, "componentCount": 6, "textChars": 1081, "nDataPoints": 4, "notes": "The slide uses a 'Why now?' and 'Where and how is ML being used effectively?' structure to frame the industry trend.", "elementsJson": [ "headline_text", "paragraph", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.339+00", "seconds": 5.102, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/108", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-108", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "That's research and development sped up by 167x!", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-697321a7e629", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.35, "x": 0.03, "y": 0.84 }, "kind": "image", "text": "Selected examples: Citrine Informatics, The Materials Project", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.339+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d6c401ab-a170-4727-a5ba-1945ba1a336c", "parentComponentId": null }, { "bbox": { "h": 0.32, "w": 0.94, "x": 0.03, "y": 0.48 }, "kind": "list", "text": "Where and how is ML being used effectively? Similar its application in drug discovery, machine learning can be used to learn the rules of material science discovery. For example, models can learn the structure of molecules and/or the stepwise process of efficiently testing these molecular properties. By using these techniques, researchers at Stanford Synchrotron Radiation Lightsource were able to create and screen 20,000 combinations of ingredients that form metallic glass in a single year. That's research and development sped up by 167x!", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.339+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "72666d2b-4b0f-4942-b50e-b00f62b6779a", "parentComponentId": null }, { "bbox": { "h": 0.22, "w": 0.94, "x": 0.03, "y": 0.21 }, "kind": "list", "text": "Why now? An enormous amount of experimental data has been generated on the properties of materials. Progress in materials science is a multiplier on broader engineering progress. But most materials are still found empirically, which limits the rate of progress. For example, scientists have manually investigated 6,000 combinations of ingredients that form metallic glass over the past 50 years.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.339+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b63d3240-6916-44d9-bdd3-bb67f78de654", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "R&D speedup: 167x", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-6d2a425a7f22", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.19, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "Material science", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.339+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5c3eb377-83c5-4b60-be6d-14fee626bcda", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-c86368581379", "evidence": "Title 'Material science' with callout '167x' R&D speedup.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 108, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Von Restorff Effect", "slug": "von-restorff-effect", "status": "active", "bestFor": null, "canonId": "019dd956-beab-73d8-9f62-30d0332e926e", "version": 1, "bodyDocId": null, "description": "Items that stand out are noticed and remembered - make key points visually distinct", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "019dd95a-0fd5-7148-8ecb-ce15efacf4fb", "evidence": "167x stat designed to stand out.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 108, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 108, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 56, "name": "Deeper Layer", "beatId": "019dd95a-0682-776c-8e34-c00b5f60e426", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Deeper Layer", "slug": "onion-deeper-layer", "status": "active", "canonId": "019dd9b8-0c69-7473-8222-610daa3e4df7", "version": 1, "description": null }, "evidence": "Talent and industry deployment across verticals.", "position": 3, "isPrimaryArc": false, "parentBeatType": { "name": "Development", "slug": "development", "status": "active", "canonId": "019dd9b8-0332-7008-b3aa-11b0840a57b1", "version": 1, "description": "The middle stretch in arcs that have one. Often where length problems creep in." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 108, "from": 86, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7a42bb8ead28", "evidence": "Parallel mini case studies: satellite, cybersec, warehouses, blue collar, agriculture, autonomy, finance, enterprise, materials.", "position": 14, "objective": "Catalogue vertical AI deployment evidence", "confidence": 72, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 109, "type": "setup", "function": "transition", "rawType": "section_divider", "block": null, "metadata": { "slideType": "section_divider", "slideTypeCanon": { "name": "Section divider", "slug": "section_divider", "status": "active", "canonId": "019de52d-066a-7402-9554-6ae504999041", "version": 1, "description": null }, "function": "transition", "functionCanon": { "name": "Transition", "slug": "transition", "status": "active", "canonId": "019de52d-1365-7149-92aa-66df859e75fd", "version": 1, "description": null }, "density": "sparse", "densityScore": 14, "componentCount": 4, "textChars": 124, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.558+00", "seconds": 2.312, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/109", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-109", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.03, "w": 0.96, "x": 0.02, "y": 0.05 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.558+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3db4e47e-2b3e-4358-a871-c5ecb3374d9f", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.06, "x": 0.92, "y": 0.05 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.558+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4cc5cb1d-e5a0-4550-8437-be8383cd1891", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.08, "x": 0.9, "y": 0.95 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.558+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5d107ffc-1c03-4bf9-8c21-87540c23789f", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.26, "x": 0.37, "y": 0.45 }, "kind": "title", "text": "Section 4: Politics", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.558+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5ef1f86b-cc64-4c30-b462-ad946d93f65a", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "106cb033-4f43-4930-885e-e40dee666f00", "evidence": "The slide clearly divides the presentation into sections.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:37.352664+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 109, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 110, "type": "setup", "title": "The slide provides context for subsequent data slides by defining the sample size and recruitment methods for Pew and Brookings surveys.", "function": "establish_context", "rawType": "appendix_methodology", "block": null, "metadata": { "slideType": "appendix_methodology", "slideTypeCanon": { "name": "Appendix methodology", "slug": "appendix_methodology", "status": "active", "canonId": "019de52d-07d9-726c-8599-167f2d2ffaa7", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 6, "textChars": 673, "nDataPoints": 2, "notes": "The slide provides context for subsequent data slides by defining the sample size and recruitment methods for Pew and Brookings surveys.", "elementsJson": [ "logo_grid", "bullet_list", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.917+00", "seconds": 2.57, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/110", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-110", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "We will review selected results from two major surveys of attitudes to AI and automation in the U.S.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-837a6e9f3d23", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.5, "x": 0.09, "y": 0.3 }, "kind": "image", "text": "Pew Research Center and The Brookings Institute logos", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Logo grid", "slug": "logo-grid", "status": "active", "canonId": "019de52c-fdd7-7409-93e5-81280a8ca215", "version": 1, "description": null }, "framework": null }, "subkind": "logo-grid", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.917+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c2b5a452-ebab-4acf-8716-c110e205b69c", "parentComponentId": null }, { "bbox": { "h": 0.25, "w": 0.9, "x": 0.02, "y": 0.65 }, "kind": "list", "text": "Brookings survey: Attitudes to AI. Conducted May 9-11 2018. Published May 2017. Survey of 1535 adult internet users in the U.S. Recruited through the Google Surveys platform.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.917+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3c038d83-ff14-44eb-829c-fdf2fb2de776", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.9, "x": 0.02, "y": 0.4 }, "kind": "list", "text": "Pew Research Center: Americans and Automation in Everyday Life. Conducted May 1-15 2017. Published October 2017. Survey of 4135 US adults. Recruited from landline and cellphone random-digit-dial surveys.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.917+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b304ee6c-72f1-47e0-8045-96745fe5fd06", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.8, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "We will review selected results from two major surveys of attitudes to AI and automation in the U.S.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.917+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fc8cbf01-0c5d-4248-bf92-fdd135378546", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Two Surveys", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.917+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a4eb2999-9381-4997-a493-46d605041312", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-d49c07a21d14", "evidence": "Opens public-attitudes pattern_hunter loop.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 110, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Concrete Language", "slug": "concrete-language", "status": "active", "bestFor": null, "canonId": "019dd956-bcd4-7398-9723-6aa72346170b", "version": 1, "bodyDocId": null, "description": "Using specific sensory details instead of abstract terms", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-d100588b1ba0", "evidence": "Sets up two specific surveys to be referenced.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 110, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "7ad4c0a9-1065-488b-b45a-ca39bb5bc4f0", "evidence": "The slide presents information in a list format with bullet points.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.420707+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 110, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 111, "type": "analysis", "title": "The chart is a diverging stacked bar chart showing likelihood of job replacement.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 6, "textChars": 527, "nDataPoints": 32, "notes": "The chart is a diverging stacked bar chart showing likelihood of job replacement.", "elementsJson": [ "headline_text", "action_title", "bar_chart_stacked", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.683+00", "seconds": 2.195, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/111", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-111", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "18% of Americans indicate that they personally know someone who has lost a job, or had their pay or hours reduced, as a result of workforce automation.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-5a6d4af72b4c", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.55, "x": 0.4, "y": 0.28 }, "kind": "chart", "text": "% of U.S. adults who think it is ___ likely that the following jobs will be replaced by robots or computers in their lifetimes", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked bar chart", "slug": "bar-stacked", "status": "active", "canonId": "019de52c-feec-74bf-9028-15b74f84f0a7", "version": 1, "description": null }, "framework": null }, "subkind": "bar-stacked", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.683+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e4fd7c81-2737-483d-8d43-3672f80a6197", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Likelihood of job replacement: 77%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-5eee87e703cf", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.35, "x": 0.03, "y": 0.33 }, "kind": "paragraph", "text": "18% of Americans indicate that they personally know someone who has lost a job, or had their pay or hours reduced, as a result of workforce automation", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.683+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6097a19a-6008-4c89-bb16-b02f37c2affb", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.82, "y": 0.94 }, "kind": "source-note", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.683+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "53a93b80-5192-4e00-8db4-96c5ea5cf9f7", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Pew Research Center", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.683+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3f6501fe-85e4-418b-b373-64f9f6f01b80", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-da378bc4ee1a", "evidence": "Title 'Growing awareness of automation impacting jobs' with 77%.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 111, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "f97b401b-1778-49e3-804f-42d10ed42981", "evidence": "The slide includes various charts, such as bar charts.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.486975+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 111, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 112, "type": "analysis", "title": "Data source: Pew Research Center. The chart compares two metrics: 'Lost a job' and 'Had pay or hours reduced' across 13 demographic categories.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 59, "componentCount": 7, "textChars": 459, "nDataPoints": 26, "notes": "Data source: Pew Research Center. The chart compares two metrics: 'Lost a job' and 'Had pay or hours reduced' across 13 demographic categories.", "elementsJson": [ "bar_chart_grouped", "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.961+00", "seconds": 2.46, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/112", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-112", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Young, part-time employed, hispanic and lower-income Americans report most impact", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-e760528496f9", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.55, "x": 0.4, "y": 0.35 }, "kind": "chart", "text": null, "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Grouped bar chart", "slug": "bar-grouped", "status": "active", "canonId": "019de52c-ff3b-751e-a8ff-f21ff11097b1", "version": 1, "description": null }, "framework": null }, "subkind": "bar-grouped", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.961+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "20d17687-2d9d-41ef-a571-9febc29c9f1d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.2, "x": 0.05, "y": 0.65 }, "kind": "legend", "text": "Lost a job; Had pay or hours reduced", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.961+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "56976db5-15bf-4e69-83b5-16c4aa96bb16", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Percentage of U.S. adults: 11%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a814-e9e6b6a2042b", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.35, "x": 0.05, "y": 0.48 }, "kind": "paragraph", "text": "% of U.S. adults in each group who say they have ever ____ because their employers replaced their positions (or some aspect of their jobs) with a machine, robot or computer program", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.961+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b9b9f569-4086-4018-b087-0afa4ae4178d", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.7, "x": 0.05, "y": 0.21 }, "kind": "title", "text": "Young, part-time employed, hispanic and lower-income Americans report most impact", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.961+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8d1797de-cc45-43da-b157-d56fd7b5cf2d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Pew Research Center", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.961+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "533faa0c-155a-401a-96d9-5e680ad17689", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-dc7a67b21a1c", "evidence": "Title 'Young, part-time, hispanic and lower-income Americans report most impact'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 112, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Storytelling Effect", "slug": "storytelling-effect", "status": "active", "bestFor": null, "canonId": "019dd956-981b-74f9-9096-5c8df03dab5a", "version": 1, "bodyDocId": null, "description": "Stories are remembered up to 22x better than facts alone", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "194d817b-1387-4f36-8492-e4a9b8d98e17", "evidence": "The slide tells a story around the data presented.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.554181+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 112, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 113, "type": "analysis", "title": "Data source: Pew Research Center. The chart uses a diverging bar format to show binary sentiment (likely vs not likely).", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 33, "componentCount": 5, "textChars": 281, "nDataPoints": 12, "notes": "Data source: Pew Research Center. The chart uses a diverging bar format to show binary sentiment (likely vs not likely).", "elementsJson": [ "bar_chart_stacked", "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:07.241+00", "seconds": 3.553, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/113", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-113", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Rising concerns about automation increasing inequality", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-9d26aa12a1db", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.45, "x": 0.05, "y": 0.3 }, "kind": "chart", "text": "Diverging bar chart showing survey results for negative and positive outcomes of automation.", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Stacked bar chart", "slug": "bar-stacked", "status": "active", "canonId": "019de52c-feec-74bf-9028-15b74f84f0a7", "version": 1, "description": null }, "framework": null }, "subkind": "bar-stacked", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.241+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e12f3a7a-6252-47a7-947e-e35cd1e90196", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Percentage of U.S. adults: 76%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:25+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-a265cec5b1b4", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.5, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Rising concerns about automation increasing inequality", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.241+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0b5041da-6fba-4dae-b2d2-02a481577a56", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Pew Research Center", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.241+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7aab366d-acb9-4b73-88d5-30bce88dc542", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-e25bfde09521", "evidence": "Title 'Rising concerns about automation increasing inequality'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 113, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "a1e36292-f66f-47fe-be42-ed05e4020c83", "evidence": "chart/bar-stacked: Diverging bar chart showing survey results for negative and positive outcomes of automation.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.413343+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 113, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 114, "type": "analysis", "title": "Data source: Pew Research Center. The chart shows a consistent trend where those impacted by automation hold stronger opinions or higher levels of concern/enthu", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 45, "componentCount": 6, "textChars": 375, "nDataPoints": 8, "notes": "Data source: Pew Research Center. The chart shows a consistent trend where those impacted by automation hold stronger opinions or higher levels of concern/enthusiasm regarding the concept.", "elementsJson": [ "bar_chart_grouped" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:06.595+00", "seconds": 2.855, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/114", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-114", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Those whose job has been impacted by automation favor more radical policies", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-e274675fa5e0", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.4, "x": 0.15, "y": 0.4 }, "kind": "chart", "text": "Grouped bar chart showing percentages for 'Impacted by automation' vs 'Not impacted' across four categories.", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Grouped bar chart", "slug": "bar-grouped", "status": "active", "canonId": "019de52c-ff3b-751e-a8ff-f21ff11097b1", "version": 1, "description": null }, "framework": null }, "subkind": "bar-grouped", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.595+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1e377dae-1f1b-4924-beb3-fe3ca3af92c3", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.3, "x": 0.18, "y": 0.42 }, "kind": "legend", "text": "Impacted by automation; Not impacted", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.595+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bbf9dca1-133b-496d-b9bd-afff72791567", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Percentage of U.S. adults: 45%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-e597a9f1b7ec", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.6, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Those whose job has been impacted by automation favor more radical policies", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.595+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "014897d9-241e-4056-bd2b-9095a320ddca", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Pew Research Center", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.595+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a409d447-7dd1-4c67-85d3-d294550ea93b", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "2x2 matrix", "slug": "matrix-2x2", "status": "active", "bestFor": null, "canonId": "019dd9e1-50ab-72d1-8afb-3ab0a38e32a8", "version": 1, "bodyDocId": null, "description": "BCG matrix, importance/urgency, any axes-and-quadrants frame.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "31c9de04-45b5-4554-be83-c10f955c40b6", "evidence": "chart/bar-grouped: Grouped bar chart showing percentages for 'Impacted by automation' vs 'Not impacted' across four categories.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.492121+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 114, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-e4275df09fbd", "evidence": "Title 'Those whose job has been impacted favor more radical policies'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 114, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 115, "type": "analysis", "title": "Source: Brookings Institute, State of AI 2018 report.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 39, "componentCount": 6, "textChars": 261, "nDataPoints": 4, "notes": "Source: Brookings Institute, State of AI 2018 report.", "elementsJson": [ "pie_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:06.122+00", "seconds": 2.379, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/115", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-115", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Overall optimism around AI...", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-d98405853e9b", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.5, "x": 0.22, "y": 0.03 }, "kind": "chart", "text": "How optimistic are you about artificial intelligence?", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Pie chart", "slug": "pie", "status": "active", "canonId": "019de52c-ffdb-71bb-a800-5c5ab9f1c373", "version": 1, "description": null }, "framework": null }, "subkind": "pie", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.122+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "54ead723-0825-4e29-aed3-a16ddb5b209a", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.2, "x": 0.6, "y": 0.35 }, "kind": "list", "text": "very optimistic\nsomewhat optimistic\nnot very optimistic\ndon't know or no answer", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.122+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "40be7666-0717-429c-a5c7-1b245ab38d4b", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Public optimism: 41%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-dd8fbd9653df", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.25, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Overall optimism around AI...", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.122+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a02af169-e5b4-479f-b771-c4f104752ab7", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Brookings Institute", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.122+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ce6e8ff2-472a-45f1-bef4-5baa20d0933d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-ea35fbb09931", "evidence": "Callout 'Overall optimism around AI' with 41% stat.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 115, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "6cbe9b34-07bb-4899-a3da-798722230479", "evidence": "chart/pie: How optimistic are you about artificial intelligence?", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.609979+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 115, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "f57f7134-e461-45d2-bf57-9a840002ae47", "evidence": "list/bullet: very optimistic, somewhat optimistic, not very optimistic, don't know or no answer", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:36.570194+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 115, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 116, "type": "analysis", "title": "Source: Brookings Institute, State of AI 2018 report.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 42, "componentCount": 6, "textChars": 328, "nDataPoints": 3, "notes": "Source: Brookings Institute, State of AI 2018 report.", "elementsJson": [ "pie_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.484+00", "seconds": 4.632, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/116", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-116", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "...and expectation that AI will \"make my life easier\"", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-1b3f6a9a2e88", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.6, "x": 0.2, "y": 0.035 }, "kind": "chart", "text": "Do you expect artificial intelligence to make your day-to-day life: easier (34%), harder (13%), don't know or no answer (53%)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Pie chart", "slug": "pie", "status": "active", "canonId": "019de52c-ffdb-71bb-a800-5c5ab9f1c373", "version": 1, "description": null }, "framework": null }, "subkind": "pie", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.484+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "53c84c7e-b558-42c1-ac6f-36e3d9f4e698", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Public sentiment toward AI: 34%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-1e029ce37616", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.15, "x": 0.82, "y": 0.93 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.484+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8844df4c-facd-49d9-9b94-1119e4af9b27", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.02, "y": 0.23 }, "kind": "title", "text": "...and expectation that AI will \"make my life easier\"", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.484+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0c0eb6f4-f997-4fb7-a1c6-7b2e414589f7", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Brookings Institute", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.484+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "16b50f52-d209-440f-8a27-869d96b0443b", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-ef108df14f19", "evidence": "Callout '...expectation that AI will make my life easier'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 116, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "1abf1dfb-56a6-487d-8e4f-8b537924a952", "evidence": "chart/pie: Do you expect artificial intelligence to make your day-to-day life: easier (34%), harder (13%), don't know or no answer (53%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.687958+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 116, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 117, "type": "analysis", "title": "Source: Brookings Institute, State of AI 2018 report.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 44, "componentCount": 6, "textChars": 381, "nDataPoints": 4, "notes": "Source: Brookings Institute, State of AI 2018 report.", "elementsJson": [ "pie_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:05.871+00", "seconds": 2.012, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/117", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-117", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "...but expectation that AI will reduce privacy", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-f37dac5c9bed", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.6, "x": 0.2, "y": 0.35 }, "kind": "chart", "text": "Do you expect artificial intelligence to: increase personal privacy (5%), have no effect on personal privacy (12%), reduce personal privacy (49%), don't know or no answer (34%)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Pie chart", "slug": "pie", "status": "active", "canonId": "019de52c-ffdb-71bb-a800-5c5ab9f1c373", "version": 1, "description": null }, "framework": null }, "subkind": "pie", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.871+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7a330f4f-0377-4236-9e5d-e903987af3fd", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Public expectation of AI impact on privacy: 49%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-f695a66ed153", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.82, "y": 0.92 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.871+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "157402ab-fcaa-4fd4-87dd-cda20f46ad29", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "...but expectation that AI will reduce privacy", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.871+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "28926fa2-6c58-4bfb-b462-8e4a891d3ef9", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Brookings Institute", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:05.871+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b6a14303-956b-4e7c-a4e8-8a519a3c3efc", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Contrast Pairs", "slug": "contrast-pairs", "status": "active", "bestFor": null, "canonId": "019dd956-9b77-70d3-88b1-f34e26f60860", "version": 1, "bodyDocId": null, "description": "Creating tension through opposites: Before↔After, Problem↔Solution, Pain↔Relief", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "loop", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecb-f59fbb53f70b", "evidence": "Optimism-then-fear tension across consecutive slides.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 117, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-f2991456b5d3", "evidence": "Title '...but expectation that AI will reduce privacy'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 117, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "a37aacfe-9ae2-460e-892c-41223ea498b8", "evidence": "chart/pie: Do you expect artificial intelligence to: increase personal privacy (5%), have no effect on personal privacy (12%), reduce personal privacy (49%), don't know or no answer (34%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.767055+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 117, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 118, "type": "analysis", "title": "Data source: Brookings Institute. Part of the State of AI 2018 report.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 31, "componentCount": 5, "textChars": 266, "nDataPoints": 4, "notes": "Data source: Brookings Institute. Part of the State of AI 2018 report.", "elementsJson": [ "pie_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:07.291+00", "seconds": 2.847, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/118", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-118", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "...and destroy jobs", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:26+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-c65e67c11da2", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.55, "x": 0.22, "y": 0.035 }, "kind": "chart", "text": "Do you expect artificial intelligence to: create jobs (12%), have no effect on jobs (13%), reduce jobs (38%), don't know or no answer (37%)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Pie chart", "slug": "pie", "status": "active", "canonId": "019de52c-ffdb-71bb-a800-5c5ab9f1c373", "version": 1, "description": null }, "framework": null }, "subkind": "pie", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.291+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1c6a69b5-0b6e-430e-8b9e-0fa650122149", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Public sentiment on AI job impact: 38%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:26+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a815-cb75e50e24a5", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.2, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "...and destroy jobs", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.291+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "623059cc-c670-45d3-a323-0d20f3af272e", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Brookings Institute", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.291+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e3a9f318-f4e2-4467-92ee-afa4b7cb0135", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "a265b959-90a7-4cf5-bf5d-67b50e72dcea", "evidence": "chart/pie: Do you expect artificial intelligence to: create jobs (12%), have no effect on jobs (13%), reduce jobs (38%), don't know or no answer (37%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.845885+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 118, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 119, "type": "analysis", "title": "Source: Brookings Institute, State of AI 2018 report.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 44, "componentCount": 6, "textChars": 373, "nDataPoints": 3, "notes": "Source: Brookings Institute, State of AI 2018 report.", "elementsJson": [ "headline_text", "pie_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:06.793+00", "seconds": 2.23, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/119", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-119", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "...and represents a threat to human beings", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-732ada8a7ddc", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.6, "x": 0.2, "y": 0.3 }, "kind": "chart", "text": "Do you think artificial intelligence represents a: threat to human beings (32%), no threat to human beings (24%), don't know or no answer (44%)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Pie chart", "slug": "pie", "status": "active", "canonId": "019de52c-ffdb-71bb-a800-5c5ab9f1c373", "version": 1, "description": null }, "framework": null }, "subkind": "pie", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.793+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bfbbb6d8-3e27-41ee-aaa7-6f2dce63e7da", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.2, "x": 0.6, "y": 0.35 }, "kind": "legend", "text": "threat to human beings, no threat to human beings, don't know or no answer", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.793+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c8c5d6d9-77e2-446c-9416-205abc9e718a", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Public sentiment: 32%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-759688989bcd", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "...and represents a threat to human beings", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.793+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c5d359a0-90e7-4888-a4f8-d5069e332b8b", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Brookings Institute", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:06.793+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5682fd7b-d7c8-4528-811b-30273fb0857a", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "836ae640-b274-46de-a7f8-8eb20c825400", "evidence": "chart/pie: Do you think artificial intelligence represents a: threat to human beings (32%), no threat to human beings (24%), don't know or no answer (44%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:36.923349+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 119, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 120, "type": "analysis", "title": "Data source: Brookings Institute. Slide from state.ai 2018 report.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 42, "componentCount": 6, "textChars": 337, "nDataPoints": 3, "notes": "Data source: Brookings Institute. Slide from state.ai 2018 report.", "elementsJson": [ "pie_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:07.012+00", "seconds": 2.254, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/120", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-120", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "...and should be regulated by government", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-2d7a382b67d3", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.35, "x": 0.22, "y": 0.33 }, "kind": "chart", "text": "Do you think government officials should: regulate artificial intelligence (42%), not regulate artificial intelligence (17%), don't know or no answer (41%)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Pie chart", "slug": "pie", "status": "active", "canonId": "019de52c-ffdb-71bb-a800-5c5ab9f1c373", "version": 1, "description": null }, "framework": null }, "subkind": "pie", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.012+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9e286eb2-ae72-42b7-b5fe-b19668c7f157", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Public sentiment on AI regulation: 42%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-3398e6e47798", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.82, "y": 0.92 }, "kind": "paragraph", "text": "state.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.012+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6cef4964-2325-4e84-829a-19f4dff551dd", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.35, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "...and should be regulated by government", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.012+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "08c59099-11d5-481d-a2b3-c590d404597c", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Brookings Institute", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.012+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1f3e939c-2893-4b05-be18-b5a78f8f8930", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "f0b9bd96-0271-48d7-bcfb-0c6b459544a7", "evidence": "chart/pie: Do you think government officials should: regulate artificial intelligence (42%), not regulate artificial intelligence (17%), don't know or no answer (41%)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.003641+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 120, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 121, "type": "analysis", "title": "Data source: Brookings Institute. The chart shows a significant portion (35%) of respondents do not know or provided no answer.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 34, "componentCount": 5, "textChars": 302, "nDataPoints": 7, "notes": "Data source: Brookings Institute. The chart shows a significant portion (35%) of respondents do not know or provided no answer.", "elementsJson": [ "bar_chart_horizontal", "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:09.18+00", "seconds": 4.317, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/121", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-121", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "While Americans believe the US is currently the world leader in AI...", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-435bf78f8a62", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.55, "x": 0.23, "y": 0.33 }, "kind": "chart", "text": "Which one is the leading country when it comes to artificial intelligence?", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.18+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9a6bc54b-e153-465a-a624-eb38d7b990ac", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Public perception of AI leadership: 21%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-4588265b139a", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.5, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "While Americans believe the US is currently the world leader in AI...", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.18+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8836a49e-cc8d-496f-8d5c-18f1bdf98cab", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Brookings Institute", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.18+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e85097f3-eaeb-41fd-adcd-52768eba1dfe", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-fab416adcdbe", "evidence": "Callout 'Americans believe US is currently the world leader'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 121, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "712f6f6c-9859-4b99-88db-8661a54a07c3", "evidence": "chart/bar-horizontal: Which one is the leading country when it comes to artificial intelligence?", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.085378+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 121, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 122, "type": "analysis", "title": "Data source: Brookings Institute. Part of the State of AI 2018 report.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 63, "componentCount": 8, "textChars": 430, "nDataPoints": 7, "notes": "Data source: Brookings Institute. Part of the State of AI 2018 report.", "elementsJson": [ "headline_text", "action_title", "bar_chart_horizontal" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.322+00", "seconds": 3.366, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/122", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-122", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "In 10 years, which one will be the leading country when it comes to artificial intelligence?", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:30+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-693f829aaeba", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.55, "x": 0.23, "y": 0.33 }, "kind": "chart", "text": "In 10 years, which one will be the leading country when it comes to artificial intelligence?", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.322+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cb5448fd-383d-49c5-99aa-5f3a56fb6451", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Public perception of AI leadership: 21%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:30+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-6cc6e478935a", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.8, "x": 0.02, "y": 0.02 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.322+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ddcc53ac-6d8c-4553-86bc-53b95b6b12dc", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.82, "y": 0.92 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.322+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "dfd12a95-f663-4814-a5a0-57c4cb2ae6a4", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.88, "y": 0.02 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.322+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f88115b3-d40d-45cd-b799-30ac87182dd6", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "...China will close the gap over the next ten years", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.322+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c4bda325-edba-4140-8fbd-85f2f06fc415", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Public Attitudes to Automation: Brookings Institute", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.322+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "657b165f-a1f1-4265-a03d-5268aec347f7", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecb-fdbf6e341b86", "evidence": "Title '...China will close the gap over the next ten years'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 122, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "97be04bd-57d0-4207-8077-2c348e642053", "evidence": "chart/bar-horizontal: In 10 years, which one will be the leading country when it comes to artificial intelligence?", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.163249+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 122, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 122, "from": 110, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-7f99ebc2936f", "evidence": "Multiple Pew/Brookings stats: 77% awareness, 76% inequality, 41% optimism, 32% threat, US/China leadership.", "position": 15, "objective": "Aggregate survey data on public attitudes to AI/automation", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 123, "type": "analysis", "title": "The slide uses a single data point (4.1%) to contrast with the narrative of automation-driven unemployment.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 38, "componentCount": 6, "textChars": 266, "nDataPoints": 1, "notes": "The slide uses a single data point (4.1%) to contrast with the narrative of automation-driven unemployment.", "elementsJson": [ "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:07.875+00", "seconds": 2.568, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/123", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-123", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Despite increased automation the US unemployment rate is at a 17 year low", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-14288a45c494", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.6, "x": 0.17, "y": 0.35 }, "kind": "chart", "text": "Unemployment Rate", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.875+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "81ec66b4-b7e3-42af-b4a7-576d5928582f", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Unemployment Rate: 4.1%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-1aa75731a30a", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.2, "x": 0.17, "y": 0.88 }, "kind": "source-note", "text": "Source: Bureau of Labor Statistics", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.875+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0729c5bf-c79a-4e80-96ec-8f9bf2ccc23c", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.65, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Despite increased automation the US unemployment rate is at a 17 year low", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.875+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4a99fcec-f90e-4e3d-89a7-199be114f277", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "How is the US labour market actually changing?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.875+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c1758670-2e1c-4f87-8c48-a8d088332746", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-01a189082ec7", "evidence": "Title 'Despite increased automation the US unemployment rate is at a 17 year low'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 123, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "ba0d0cdb-835e-4443-8135-58fea9ec00c3", "evidence": "chart/line: Unemployment Rate", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.24311+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 123, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Unexpected Pattern", "slug": "unexpected-pattern", "status": "active", "bestFor": null, "canonId": "019dd956-bc0f-71af-8d32-3c1a2da38c40", "version": 1, "bodyDocId": null, "description": "Breaking a pattern to create surprise and attention", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-06352be76c2f", "evidence": "Counter-intuitive pairing of automation rise + low unemployment.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 123, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 129, "from": 123, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-823fbc64276b", "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 124, "type": "analysis", "title": "The chart highlights the stagnation of routine cognitive and manual jobs compared to the growth of non-routine cognitive roles.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 38, "componentCount": 6, "textChars": 245, "nDataPoints": 4, "notes": "The chart highlights the stagnation of routine cognitive and manual jobs compared to the growth of non-routine cognitive roles.", "elementsJson": [ "headline_text", "action_title", "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:07.685+00", "seconds": 2.225, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/124", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-124", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Routine jobs have stagnated", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:28+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-b43551c49296", "parentComponentId": null }, { "bbox": { "h": 0.62, "w": 0.52, "x": 0.24, "y": 0.28 }, "kind": "chart", "text": "Jobs: Routine Vs. Nonroutine, Cognitive Vs. Manual", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.685+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "daed08ab-33e2-4e82-a51d-29fbfd8c46f0", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Employment (Millions)", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:28+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-b9bd261f4673", "parentComponentId": null }, { "bbox": { "h": 0.02, "w": 0.3, "x": 0.24, "y": 0.9 }, "kind": "source-note", "text": "SOURCE: Current Population Survey and author's calculations.", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.685+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0b43f36a-5e93-4f40-a60e-90795e5f99b6", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.25, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Routine jobs have stagnated", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.685+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "92d9cc95-be76-4fb6-94b9-d1617c61bedc", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.65, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "More broadly, how is the US labour market actually changing?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.685+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "475b80cd-870a-4776-9683-ce7ea44d9cd2", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-0afbe66e9cb6", "evidence": "Callout 'Routine jobs have stagnated'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 124, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "21ce9b37-ffc1-4f92-af93-0e2d7a7beb95", "evidence": "chart/line: Jobs: Routine Vs. Nonroutine, Cognitive Vs. Manual", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.5, "extraction": { "at": "2026-07-16 21:56:37.322896+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 124, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 129, "from": 123, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-823fbc64276b", "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 125, "type": "analysis", "title": "Source: The New York Times, stateof.ai 2018 report.", "function": "analyze_data", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 29, "componentCount": 5, "textChars": 251, "nDataPoints": 2, "notes": "Source: The New York Times, stateof.ai 2018 report.", "elementsJson": [ "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:07.897+00", "seconds": 2.39, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/125", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-125", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Wages have lagged the increase in jobs", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-9f336386291e", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.56, "x": 0.22, "y": 0.33 }, "kind": "chart", "text": "Line chart showing Unemployment rate (light blue) and Wage growth (black) from 2008 to 2018.", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.897+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "171abf9f-4970-4f86-9995-625dee979ad6", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Unemployment rate / Wage growth: 4.1%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:27+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-a0a1c593800f", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.35, "x": 0.05, "y": 0.22 }, "kind": "title", "text": "Wages have lagged the increase in jobs", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.897+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "07cbfa09-f8b5-4fe0-8459-7e306e25d897", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "How is the US labour market actually changing?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:07.897+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "72c6aaf9-30e7-4018-9465-461629f1eed8", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-0db031fedbd8", "evidence": "Callout 'Wages have lagged the increase in jobs'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 125, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "5539f9f1-80ad-480b-9281-476893842f35", "evidence": "chart/line: Line chart showing Unemployment rate (light blue) and Wage growth (black) from 2008 to 2018.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:42.917039+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 125, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 129, "from": 123, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-823fbc64276b", "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 126, "type": "analysis", "title": "The chart shows recession periods as grey vertical bars and the average weeks of unemployment as a red line.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 41, "componentCount": 6, "textChars": 302, "nDataPoints": 6, "notes": "The chart shows recession periods as grey vertical bars and the average weeks of unemployment as a red line.", "elementsJson": [ "headline_text", "line_chart", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.224+00", "seconds": 2.435, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/126", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-126", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Since 2010 there has been a marked change in how long unemployment lasts for", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:28+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-c86d21791a51", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.52, "x": 0.24, "y": 0.28 }, "kind": "chart", "text": "Average (Mean) Duration of Unemployment", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.224+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "c16872b8-8083-4645-8eea-fd7c6747694c", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Average weeks of unemployment: 21.3", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:28+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-ce79346f8695", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.6, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "Since 2010 there has been a marked change in how long unemployment lasts for", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.224+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3d6145d0-5911-478c-8e62-1507f3f139bb", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.05, "x": 0.72, "y": 0.27 }, "kind": "source-note", "text": "dshort.com June 2018 As of May", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.224+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f6d27491-d212-4a4c-bb3e-f33925b87893", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "How is the US labour market actually changing?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.224+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f3388fb8-a4b5-486a-abf9-ea66589a3fc7", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-13d91fab0a78", "evidence": "Callout 'marked change in how long unemployment lasts for'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 126, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 129, "from": 123, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-823fbc64276b", "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 127, "type": "analysis", "title": "Source: The Brookings Institute, State of AI 2018 report.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 37, "componentCount": 6, "textChars": 249, "nDataPoints": 2, "notes": "Source: The Brookings Institute, State of AI 2018 report.", "elementsJson": [ "headline_text", "action_title", "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.306+00", "seconds": 2.483, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/127", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-127", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Labour productivity and hourly compensation have diverged", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:28+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-e44305392425", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.7, "x": 0.15, "y": 0.315 }, "kind": "chart", "text": "Real Labor Productivity and Hourly Compensation, 1947–2017", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.306+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d157e1f0-fcf9-4c6d-92d1-be98a987e30f", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Index (1947 = 100)", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:28+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-e98d4b74d62b", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.82, "y": 0.92 }, "kind": "source-note", "text": "state.ai 2018", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.306+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8c7a61ab-d148-449e-b20f-5f5f45ab3d5f", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.45, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "Labour productivity and hourly compensation have diverged", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.306+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "68869484-1ebd-4dee-82ff-4a9f324870a5", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "How is the US labour market actually changing?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.306+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "025c9986-ed96-40c3-8575-0073f457cbb2", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-1790edc3ce78", "evidence": "Callout 'Labour productivity and hourly compensation have diverged'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 127, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "9c0a94d3-c480-4e2a-ad91-aa074b8c0692", "evidence": "chart/line: Real Labor Productivity and Hourly Compensation, 1947–2017.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.009831+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 127, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 129, "from": 123, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-823fbc64276b", "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 128, "type": "analysis", "title": "The chart highlights recessionary periods with vertical grey bars.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 28, "componentCount": 5, "textChars": 221, "nDataPoints": 1, "notes": "The chart highlights recessionary periods with vertical grey bars.", "elementsJson": [ "headline_text", "action_title", "line_chart", "logo_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:10.499+00", "seconds": 4.573, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/128", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-128", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Labour's share of income has been declining steadily", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:32+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-98cc1e4e8dbe", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.8, "x": 0.1, "y": 0.28 }, "kind": "chart", "text": "Line chart of US labor share of income (1973-2017)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.499+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b5232b5e-e4c5-4696-b01a-5bb0b49e82f9", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Labor share (percent)", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:32+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-9e73a2a11092", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.04, "y": 0.22 }, "kind": "title", "text": "Labour's share of income has been declining steadily", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.499+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7edfa8d6-edae-4c17-aeae-a3db7029deb4", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "How is the US labour market actually changing?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.499+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2aee583a-3046-4c19-90de-fc0787e51555", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-18294dab0b06", "evidence": "Callout 'Labour's share of income has been declining steadily'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 128, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "de6879c5-cd34-4285-9e84-700062ddb85e", "evidence": "chart/line: Line chart of US labor share of income (1973-2017).", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.069858+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 128, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 129, "from": 123, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-823fbc64276b", "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 129, "type": "analysis", "title": "The chart uses a U-shaped distribution of income volatility against annual income levels.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 51, "componentCount": 7, "textChars": 378, "nDataPoints": 1, "notes": "The chart uses a U-shaped distribution of income volatility against annual income levels.", "elementsJson": [ "headline_text", "action_title", "scatter_plot" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.515+00", "seconds": 2.544, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/129", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-129", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Workers are experiencing greater income volatility", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:28+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a816-fd19cafa478d", "parentComponentId": null }, { "bbox": { "h": 0.65, "w": 0.5, "x": 0.315, "y": 0.23 }, "kind": "chart", "text": "Average monthly change in income vs Annual Income", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Scatter plot", "slug": "scatter", "status": "active", "canonId": "019de52d-007d-72ec-8689-c79fb78e21a2", "version": 1, "description": null }, "framework": null }, "subkind": "scatter", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.515+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "34547c17-ce68-49eb-af91-ed04efa845c3", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.18, "x": 0.44, "y": 0.34 }, "kind": "legend", "text": "People with low incomes see huge swings in income on a monthly basis.", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.515+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "3c84b552-ef73-43bf-bcc0-6eb714937938", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.2, "x": 0.69, "y": 0.78 }, "kind": "legend", "text": "But even among the middle class, incomes can vary monthly by more than 30 percent.", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.515+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6616068f-7027-4e39-b824-d283c9468f12", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Average monthly change in income", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:28+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-014fa228475c", "parentComponentId": null }, { "bbox": { "h": 0.06, "w": 0.25, "x": 0.02, "y": 0.225 }, "kind": "title", "text": "Workers are experiencing greater income volatility", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Action title", "slug": "action-title", "status": "active", "canonId": "019de52c-fb31-734c-a0f5-a3935b73135d", "version": 1, "description": null }, "framework": null }, "subkind": "action-title", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.515+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8a1d683a-ff87-43f2-80a8-dc55fb6e23be", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "How is the US labour market actually changing?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.515+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "09572701-4985-4560-9dd9-9d4ea31f25fa", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-1ed224a59dbf", "evidence": "Callout 'Workers are experiencing greater income volatility'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 129, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "d5cc1dbd-4efb-42b7-bc14-60cfa0aae27f", "evidence": "chart/scatter: Average monthly change in income vs Annual Income", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.133113+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 129, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 129, "from": 123, "loop": { "name": "02_pattern_hunter", "slug": "02-pattern-hunter", "status": "active", "bestFor": "Time-pressed audiences, building consensus, when data is strong", "canonId": "019dd956-6695-771a-be2e-f1748ef041a4", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion", "description": "Group multiple pieces of evidence that together point to a pattern or conclusion", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-823fbc64276b", "evidence": "Five+ contiguous data slides: routine jobs flat, wages lag, unemployment duration, productivity divergence, income volatility.", "position": 16, "objective": "Stack indicators of structural shift in US labor market", "confidence": 78, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 130, "type": "tension", "title": "The slide uses a 'Don't worry' vs 'Worry' dichotomy to frame the debate on AI and labor.", "function": "frame_problem", "rawType": "problem_statement", "block": null, "metadata": { "slideType": "problem_statement", "slideTypeCanon": { "name": "Problem statement", "slug": "problem_statement", "status": "active", "canonId": "019de52d-0879-768d-bba8-8da8a5a121d4", "version": 1, "description": null }, "function": "frame_problem", "functionCanon": { "name": "Frame problem", "slug": "frame_problem", "status": "active", "canonId": "019de52d-13db-753e-ad70-e9b2d0dab398", "version": 1, "description": null }, "density": "dense", "densityScore": 58, "componentCount": 6, "textChars": 1135, "nDataPoints": 0, "notes": "The slide uses a 'Don't worry' vs 'Worry' dichotomy to frame the debate on AI and labor.", "elementsJson": [ "paragraph", "bullet_list" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.893+00", "seconds": 2.858, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/130", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-130", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "There are two poles of thought on how machine learning will affect the labour market.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-8c122ec68d8d", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.95, "x": 0.02, "y": 0.68 }, "kind": "list", "text": "\"Worry\" - This time it's different. In previous industrial revolutions we automated human muscular power and somewhat routine cognitive skills. With increasingly advanced machine learning we will replicate more and more of human intelligence, reducing the number of well paid jobs and adding fewer jobs than are destroyed.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "57bf23c4-9bd2-44eb-ba1e-aedbc8e6bcaf", "parentComponentId": null }, { "bbox": { "h": 0.2, "w": 0.95, "x": 0.02, "y": 0.45 }, "kind": "list", "text": "\"Don't worry\" - Historically technology has been a net job creator and it won't be different this time. Machine learning will create more jobs than it destroys and like previous industrial revolutions, most of those jobs will be new ones that we can't imagine today. Yes, we got Automated Teller Machines at banks, but we also got many new jobs that replaced the bank teller jobs that were lost.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fc6895fc-aa14-4caa-94da-263bdd2b44f1", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "It's hard to say for now. There are many confounding factors including globalisation/offshoring, reduced unionisation, increased financialisation of the economy, increased consolidation, and demographic shifts.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "20644f6f-1c24-4a18-b37e-3a5569f83c12", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.02, "y": 0.37 }, "kind": "paragraph", "text": "There are two poles of thought on how machine learning will affect the labour market:", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9638e9b2-8bf2-41f6-a278-7e0c6e880154", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.4, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "How much of this is due to automation?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6e895d32-940e-4f8d-b158-85b80b384661", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Curiosity Gap", "slug": "curiosity-gap", "status": "active", "bestFor": null, "canonId": "019dd956-9703-74ef-8b85-96dfd8525025", "version": 1, "bodyDocId": null, "description": "People seek missing information - create open questions that demand resolution", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Storyteller", "layer": "block", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-27de9843b78b", "evidence": "Open question opens analytical loop.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 60, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 130, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-20732ba33df9", "evidence": "Question title 'How much of this is due to automation?'", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 130, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Problem Statement Canvas", "slug": "problem-statement-canvas", "status": "active", "bestFor": null, "canonId": "019dd956-839d-744b-ad11-9bf3e3642a18", "version": 1, "bodyDocId": null, "description": "Structured definition of the problem: What? Who? Where? When? Scale?", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "ae209c37-bae3-4438-bfe5-fb5b7be0013f", "evidence": "There are two poles of thought on how machine learning will affect the labour market.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.6, "extraction": { "at": "2026-07-16 21:56:43.193929+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 130, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [ { "matchId": "c18cd0bb-0678-439b-8984-6d0e00470c11", "evidence": "The slide presents two opposing viewpoints ('Don't worry' vs 'Worry') to frame the debate.", "framework": null, "confidence": 0.8, "extraction": { "at": "2026-05-02 17:44:08.893+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 130, "frameworkName": "comparison_frame" } ], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 131, "type": "analysis", "title": "The chart uses a back-to-back bar configuration to compare job volume growth against relative wage levels.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 26, "componentCount": 4, "textChars": 244, "nDataPoints": 20, "notes": "The chart uses a back-to-back bar configuration to compare job volume growth against relative wage levels.", "elementsJson": [ "bar_chart_horizontal", "bar_chart_horizontal" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.942+00", "seconds": 2.827, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/131", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-131", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.6, "w": 0.22, "x": 0.385, "y": 0.21 }, "kind": "chart", "text": "Dual bar chart comparing job growth (left) and median wage (right) by occupation.", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.942+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "45a183db-bb11-41dc-baad-b54a826d89c2", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Median annual wage", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-5a1c40a1e47c", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.25, "x": 0.38, "y": 0.9 }, "kind": "source-note", "text": "Source: A Turner, 'Capitalism in an Age of Robots' (Institute for New Economic Thinking, 2018) © FT", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.942+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "cbda9020-a34b-4f24-9a75-4b3e38ce88a4", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.45, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "For now, many new jobs are relatively low paid", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.942+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "52ae1fa6-cbff-4fa9-b57f-06f273729bdd", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "d3cf1188-84ae-48ea-b614-900a0f345425", "evidence": "chart/bar-horizontal: Dual bar chart comparing job growth (left) and median wage (right) by occupation.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.254642+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 131, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 132, "type": "analysis", "title": "The chart distinguishes between historical data (solid blue) and forecast data (lighter blue/grey).", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 44, "componentCount": 6, "textChars": 330, "nDataPoints": 13, "notes": "The chart distinguishes between historical data (solid blue) and forecast data (lighter blue/grey).", "elementsJson": [ "headline_text", "bar_chart_vertical" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.798+00", "seconds": 2.626, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/132", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-132", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Install base growing 12% year-on-year", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-2b45685a5644", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.6, "x": 0.2, "y": 0.33 }, "kind": "chart", "text": "Estimated worldwide operational stock of industrial robots 2015-2016 and forecast for 2017*-2020*", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.798+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f4e30542-31c8-4fe8-8657-8073a6794617", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.3, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "Install base growing 12% year-on-year", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.798+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5fdcf593-8d5b-4a29-b287-7af8fa217dd9", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Operational stock of industrial robots: 2 million", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-2ee642eafbbc", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.65, "y": 0.91 }, "kind": "source-note", "text": "Source: IFR World Robotics 2017", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.798+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d78e7b1f-d692-4e44-9560-c86fa7972220", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "It is also still early, there are only 2 million industrial robots in the world", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.798+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "03e0f8ed-cf02-4458-9b93-33f0e284f55b", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-2a2855e78437", "evidence": "Title 'It is also still early, only 2 million industrial robots in the world'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 132, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "eb35aeac-5d14-4b0c-b983-bd1a0f3bd388", "evidence": "chart/bar-vertical: Estimated worldwide operational stock of industrial robots 2015-2016 and forecast for 2017*-2020*", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.315085+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 132, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 133, "type": "setup", "title": "Source: International Robotics Federation. Part of the State of AI 2018 report.", "function": "compare_peers", "rawType": "peer_benchmark", "block": null, "metadata": { "slideType": "peer_benchmark", "slideTypeCanon": { "name": "Peer benchmark", "slug": "peer_benchmark", "status": "active", "canonId": "019de52d-08f0-7158-a78d-d7b5331fefaf", "version": 1, "description": null }, "function": "compare_peers", "functionCanon": { "name": "Compare peers", "slug": "compare_peers", "status": "active", "canonId": "019de52d-1453-72ba-a367-9ba3d34c1630", "version": 1, "description": null }, "density": "balanced", "densityScore": 31, "componentCount": 5, "textChars": 260, "nDataPoints": 6, "notes": "Source: International Robotics Federation. Part of the State of AI 2018 report.", "elementsJson": [ "bar_chart_horizontal" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.56+00", "seconds": 2.388, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/133", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-133", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "US automation hasn't kept up with other advanced economies", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-403b5658314d", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.7, "x": 0.15, "y": 0.25 }, "kind": "chart", "text": "Robots per 10000 manufacturing employees", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.56+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0982faf4-d6f9-4c17-995e-847274787d8c", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Robots per 10,000 manufacturing employees: 631", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-476aa52244b3", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.2, "x": 0.25, "y": 0.85 }, "kind": "source-note", "text": "Data: International Robotics Federation", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.56+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2ea22fe0-b9a4-4326-9da0-bf83bb1c0f23", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.85, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "There are fewer robots in U.S. factories compared to other advanced economies", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.56+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "32ccbc43-e3fe-44ae-bee1-648adc14f82b", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-2eb49b036ec4", "evidence": "Title 'There are fewer robots in U.S. factories compared to other advanced economies'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 133, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Slide recipe: the peer-gap chart", "slug": "peer-gap-chart-recipe", "status": "active", "bestFor": null, "canonId": "019df269-2b04-75bc-ad85-e0179264bdd7", "version": 1, "bodyDocId": "019df269-2ae1-771d-9cd1-193a2e25f21c", "description": "The single most-copied slide in activism. This is the exact layout, used", "familyLabel": "slide-recipe", "categoryName": "Slide", "categorySlug": "slide" }, "agent": "designer", "layer": "slide", "agents": [ "designer" ], "matchId": "d481df83-bd91-4c66-944d-f51df0d7ab62", "evidence": "chart/bar-horizontal: Robots per 10000 manufacturing employees", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.8, "extraction": { "at": "2026-07-16 21:56:43.375047+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 133, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 134, "type": "analysis", "title": "The chart shows a massive divergence between the computer subsector and the rest of manufacturing.", "function": "diagnose", "rawType": "diagnosis", "block": null, "metadata": { "slideType": "diagnosis", "slideTypeCanon": { "name": "Diagnosis", "slug": "diagnosis", "status": "active", "canonId": "019de52d-0cd7-7212-898f-0464041ff8d8", "version": 1, "description": null }, "function": "diagnose", "functionCanon": { "name": "Diagnose", "slug": "diagnose", "status": "active", "canonId": "019de52d-1657-7223-acc5-7437d878c850", "version": 1, "description": null }, "density": "balanced", "densityScore": 35, "componentCount": 5, "textChars": 358, "nDataPoints": 3, "notes": "The chart shows a massive divergence between the computer subsector and the rest of manufacturing.", "elementsJson": [ "line_chart", "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.754+00", "seconds": 2.496, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/134", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-134", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.6, "w": 0.56, "x": 0.22, "y": 0.28 }, "kind": "chart", "text": "Line chart showing exponential growth in the computer subsector compared to flat growth in other manufacturing sectors.", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Line chart", "slug": "line", "status": "active", "canonId": "019de52c-ff62-778f-8f6c-a952ceae8677", "version": 1, "description": null }, "framework": null }, "subkind": "line", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.754+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2b7a7f49-e0e5-4f55-9d67-ada68ce51dcc", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Real output growth", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-bf031349cf47", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.7, "x": 0.15, "y": 0.22 }, "kind": "paragraph", "text": "Real output growth for manufacturing with and without the computers subsector", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.754+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1efe3f50-8a80-4fdb-b43a-4bbb4b90e443", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.35, "x": 0.22, "y": 0.94 }, "kind": "source-note", "text": "Data: Bureau of Economic Analysis, Houseman calculations", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.754+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "80db5df2-e8ec-4811-b67e-127b0ad54054", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Do huge productivity gains in the computer sector mask stagnation in U.S. manufacturing?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.754+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9dc0e440-1877-4798-8611-fff7f10c1af1", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "87576d9a-eec1-4094-8387-8a753dedd938", "evidence": "chart/line: Line chart showing exponential growth in the computer subsector compared to flat growth in other manufacturing sectors.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.434951+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 134, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 135, "type": "analysis", "title": "The chart shows a significant decline in retail employees in 2016 compared to Amazon's growth.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 29, "componentCount": 4, "textChars": 308, "nDataPoints": 15, "notes": "The chart shows a significant decline in retail employees in 2016 compared to Amazon's growth.", "elementsJson": [ "bar_chart_grouped", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.948+00", "seconds": 2.626, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/135", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-135", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.5, "w": 0.7, "x": 0.15, "y": 0.3 }, "kind": "chart", "text": "Change in employees and robots at AMZN and AMZN-related retail", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Grouped bar chart", "slug": "bar-grouped", "status": "active", "canonId": "019de52c-ff3b-751e-a8ff-f21ff11097b1", "version": 1, "description": null }, "framework": null }, "subkind": "bar-grouped", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.948+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e24e3ed7-96c6-4ab9-ac9d-b54ca9ecc92a", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.18, "y": 0.32 }, "kind": "legend", "text": "Retail employees, AMZN employees, Total employees, AMZN robots, Total employees + robots", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.948+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fb6a9e65-63fe-4172-8b0e-75673b2230ac", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Change in employees and robots", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-a4f80721ee1c", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "One recent piece of analysis found that while Amazon is rapidly hiring people and robots, taken as a whole retail is losing jobs", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.948+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d22e0685-0ea3-4b93-9de9-707878d657fc", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Data-Ink Ratio", "slug": "data-ink-ratio", "status": "active", "bestFor": null, "canonId": "019dd956-afc8-70da-b378-92834d36fde6", "version": 1, "bodyDocId": null, "description": "Maximize data, minimize non-data ink - remove chartjunk and decoration", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "26147f2d-bba1-41d9-9817-ba71c0e25157", "evidence": "chart/bar-grouped: Change in employees and robots at AMZN and AMZN-related retail", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:43.494859+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 135, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 136, "type": "setup", "title": "Part of the 'State of AI 2018' report, specifically the 'Politics' section.", "function": "present_solution", "rawType": "strategic_options", "block": null, "metadata": { "slideType": "strategic_options", "slideTypeCanon": { "name": "Strategic options", "slug": "strategic_options", "status": "active", "canonId": "019de52d-0c0d-745a-8f0a-50898a91cc12", "version": 1, "description": null }, "function": "present_solution", "functionCanon": { "name": "Present solution", "slug": "present_solution", "status": "active", "canonId": "019de52d-15b8-71da-bd23-2b767850cbce", "version": 1, "description": null }, "density": "sparse", "densityScore": 24, "componentCount": 3, "textChars": 458, "nDataPoints": 0, "notes": "Part of the 'State of AI 2018' report, specifically the 'Politics' section.", "elementsJson": [ "bullet_list" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:08.998+00", "seconds": 2.597, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/136", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-136", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.23 }, "kind": "list", "text": "Universal Basic Income (UBI) or Basic Income: Has received substantial media coverage over the past years. We review various trials that are now being rolled out", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.998+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4cabd726-28b9-458c-a9c0-a4280dc2301e", "parentComponentId": null }, { "bbox": { "h": 0.15, "w": 0.9, "x": 0.05, "y": 0.43 }, "kind": "list", "text": "Universal Basic Services (UBS): A less mainstream idea that was recently fleshed out by the Institute for Global Prosperity at UCL. We highlight the proposal as an interesting new alternative or complement to UBI.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.998+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "92f56291-263a-4d5d-94f4-92d537918968", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "If automation does reduce net employment and/or wages what new policies will emerge?", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:08.998+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fd14c457-44f6-40e3-8843-9ef7db36ee08", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 137, "type": "analysis", "function": "summarize", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "balanced", "densityScore": 48, "componentCount": 4, "textChars": 1034, "nDataPoints": 6, "notes": null, "elementsJson": [ "headline_text", "paragraph", "bullet_list" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:11.897+00", "seconds": 5.299, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/137", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-137", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "'Basic Income' aims to mitigate technological unemployment with guaranteed payments to cover basic needs", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-6e7aeecd746c", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.9, "x": 0.03, "y": 0.28 }, "kind": "list", "text": "Finland's basic income trial is running with 2,000 randomly selected participants receiving €560 per month. Will conclude in December 2018. Analysis of the effects will take place in 2019.\nOntario basic income pilot began enrolling participants in April 2018. Will be restricted to 4,000 lower income participants.\nFive municipal experiments in the Netherlands with basic income commenced in late 2017.\nBarcelona launched B-MINCOME experiment in October 2017 with 2000 low income households.\nUS Charity GiveDirectly launched trial in Kenya in November 2017. More than 21,000 people will eventually receive some type of cash transfer, with more than 5,000 receiving a long-term basic income.\nY Combinator research published proposal for randomised control trial with 3000 adults in the United States", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.897+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d9ec5e2f-5527-465d-a165-0b9a0eb2fb58", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.9, "x": 0.03, "y": 0.22 }, "kind": "paragraph", "text": "'Basic Income' aims to mitigate technological unemployment with guaranteed payments to cover basic needs", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.897+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e3cce2b3-f3cb-4d2e-846f-6861f383bb1f", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.3, "x": 0.03, "y": 0.14 }, "kind": "title", "text": "Basic Income trials roll out", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.897+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f8c69b07-75f1-4491-9065-5824f0adb3a4", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-313effc0b6a5", "evidence": "Title 'Basic Income trials roll out'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 137, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "b2a38234-4191-4e53-a415-27fee0ba17ee", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:37.983879+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 137, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 138, "type": "analysis", "title": "The table compares the cost and value of providing specific services (UBS) versus a UBI equivalent.", "function": "quantify_impact", "rawType": "data_table", "block": null, "metadata": { "slideType": "data_table", "slideTypeCanon": { "name": "Data table", "slug": "data_table", "status": "active", "canonId": "019de52d-0bbd-71b9-9918-417f904abbf8", "version": 1, "description": null }, "function": "quantify_impact", "functionCanon": { "name": "Quantify impact", "slug": "quantify_impact", "status": "active", "canonId": "019de52d-1630-720d-b57d-fa7f76bebd8d", "version": 1, "description": null }, "density": "balanced", "densityScore": 47, "componentCount": 5, "textChars": 581, "nDataPoints": 20, "notes": "The table compares the cost and value of providing specific services (UBS) versus a UBI equivalent.", "elementsJson": [ "data_table", "icon_grid" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:09.9+00", "seconds": 3.12, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/138", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-138", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "‘Universal Basic Services’ (UBS) would build on existing state provision of services", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-ccbcd0937970", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Total Cost: £42.16 Bn", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-d326f1bd4e92", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.2 }, "kind": "paragraph", "text": "'Universal Basic Services' (UBS) would build on existing state provision of services. Seeks to expand public services (e.g. a National Health Service) to other major categories of consumer spend (transport, food, shelter). Interesting model for countries with a meaningful welfare state.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.9+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "155cbc18-2f4f-48a6-b396-da4a674530ec", "parentComponentId": null }, { "bbox": { "h": 0.48, "w": 0.58, "x": 0.21, "y": 0.43 }, "kind": "table", "text": "Table showing UBS categories (Shelter, Food, Transport, Information) against metrics (Definition, Cost, D1 effect, User value, UBI equiv)", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.9+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ee64521f-bc4d-4691-bd5a-891b81f3963f", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.5, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Universal Basic Services proposed by UCL researchers", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.9+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "29c806dc-6bb9-4adb-a86e-19cc724e6176", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-34b742690b57", "evidence": "Title 'Universal Basic Services proposed by UCL researchers' with £42.16Bn.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 138, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 138, "from": 130, "loop": { "name": "27_cost_of_inaction", "slug": "27-cost-of-inaction", "status": "active", "bestFor": "Urgent budget requests, compliance, risk mitigation", "canonId": "019dd956-70d1-7395-a15b-857ba858b394", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point", "description": "Quantify what happens if the audience does nothing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-858c9fef18af", "evidence": "Two poles of thought -> early signals (low-paid jobs, robot count) -> Basic Income/UBS as proposed policies.", "position": 17, "objective": "Frame inaction risk and policy responses to automation", "confidence": 65, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 139, "type": "analysis", "title": "The slide highlights the competitive nature of national AI strategies, specifically focusing on China's 2017 announcement.", "function": "establish_context", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "dense", "densityScore": 52, "componentCount": 5, "textChars": 887, "nDataPoints": 3, "notes": "The slide highlights the competitive nature of national AI strategies, specifically focusing on China's 2017 announcement.", "elementsJson": [ "headline_text", "bullet_list", "screenshot" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:09.845+00", "seconds": 2.957, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/139", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-139", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "By 2030, China will “become the world’s premier artificial intelligence innovation center and foster a new national leadership and establish the key fundamentals for an economic great power.”", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:32+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-7f15bc6918bb", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.28, "x": 0.68, "y": 0.25 }, "kind": "image", "text": "Infographic detailing China's 2030 AI development plan milestones for 2020, 2025, and 2030.", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Infographic", "slug": "infographic", "status": "active", "canonId": "019de52c-fe26-750c-9d7d-9f34e21324e8", "version": 1, "description": null }, "framework": null }, "subkind": "infographic", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.845+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e87dec52-5e14-48fd-a311-12f0307e6d38", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.65, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "China 2030 (announced July 2017)\nPartly a reaction to Obama White House report on AI (in 2016)\nNew state funded $2.1 billion AI park in Beijing\nCall for researchers to be making major breakthroughs by 2025\nBy 2030, China will “become the world’s premier artificial intelligence innovation center and foster a new national leadership and establish the key fundamentals for an economic great power.”\nBaidu announces new lab in collaboration with Chinese government\nGoal: to build a $150 billion AI industry by 2030", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.845+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4bfa3546-f83c-4fa3-a29b-dd2f35a3acfa", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Industry value target: $150 billion", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:32+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-8185c69a6515", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI Nationalism: flurry of National AI strategies announced", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.845+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e6a52e73-f008-48d1-8bd2-18f45532e2e5", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Inductive Reasoning", "slug": "inductive-reasoning", "status": "active", "bestFor": null, "canonId": "019dd956-99c6-705e-94cc-76a504c3da37", "version": 1, "bodyDocId": null, "description": "Grouping similar evidence to form a conclusion (Evidence → Pattern → Conclusion)", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-3e0e39c4a499", "evidence": "First slide of why-now AI nationalism evidence stack.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 139, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-38a83ee7ef14", "evidence": "Title 'AI Nationalism: flurry of National AI strategies announced' with $150B China.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 139, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "1b3854e3-fccf-4bd8-96a0-a8c39c260340", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.083649+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 139, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 143, "from": 139, "loop": { "name": "15_why_now", "slug": "15-why-now", "status": "active", "bestFor": "Sales pitches, fundraising, requesting immediate budget approval", "canonId": "019dd956-6c00-70cd-bd42-6bacc4af726f", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Context (Trends) -> The Trigger Event -> The Window of Opportunity", "description": "Create temporal urgency by proving that the window of opportunity is opening or closing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8b5e319977cd", "evidence": "Five contiguous slides with $150B China, €1.5B France, $2B Korea, €20B EU, £603M UK strategies.", "position": 18, "objective": "Argue national AI strategies create urgent geopolitical window", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 140, "type": "evidence", "title": "The slide uses a photo of Emmanuel Macron to anchor the case study on French AI policy.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 53, "componentCount": 6, "textChars": 650, "nDataPoints": 1, "notes": "The slide uses a photo of Emmanuel Macron to anchor the case study on French AI policy.", "elementsJson": [ "headline_text", "bullet_list", "photo" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:09.711+00", "seconds": 2.736, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/140", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-140", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.7, "w": 0.48, "x": 0.49, "y": 0.22 }, "kind": "image", "text": "Emmanuel Macron speaking at the AI for Humanity event", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.711+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "63cbd813-fd93-4ff6-b72d-349e68e8ddb2", "parentComponentId": null }, { "bbox": { "h": 0.55, "w": 0.45, "x": 0.02, "y": 0.28 }, "kind": "list", "text": "“My goal is to recreate a European sovereignty in AI” - Macron\n€1.5 billion committed over 5 years\nNew AI research centres in Paris opened by Facebook, Google, Samsung, DeepMind, Fujitsu\nPlan to open up of data collected by state-owned organizations such as France’s centralized healthcare system\nSeparately, France announces that foreign takeovers of AI companies will be subject to government approval", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.711+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "177bb7fb-a992-41b5-a4ac-ad61a5f086cf", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Investment commitment: €1.5 billion", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-fdda2e2e9d9d", "parentComponentId": null }, { "bbox": null, "kind": "quote", "text": "My goal is to recreate a European sovereignty in AI - Macron", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": { "name": "Authority citation", "slug": "authority-citation", "status": "active", "bestFor": null, "canonId": "019dd9e1-49ae-7557-89ad-3fca70469b45", "version": 1, "bodyDocId": null, "description": "Quoting an authority figure, study, or precedent to lend weight.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:29+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a817-fae36e96e29f", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI Nationalism: flurry of National AI strategies announced", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.711+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e5a4413f-f1f6-4fcc-b979-d20b81092a13", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.35, "x": 0.02, "y": 0.22 }, "kind": "title", "text": "French AI Strategy (announced March 2018)", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Subtitle", "slug": "subtitle", "status": "active", "canonId": "019de52c-fb09-739d-900c-815b0d9ac526", "version": 1, "description": null }, "framework": null }, "subkind": "subtitle", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.711+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "55651fd3-0564-4092-82eb-38277381b144", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-40cadf752666", "evidence": "Title repeats AI Nationalism with €1.5B France figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 140, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 143, "from": 139, "loop": { "name": "15_why_now", "slug": "15-why-now", "status": "active", "bestFor": "Sales pitches, fundraising, requesting immediate budget approval", "canonId": "019dd956-6c00-70cd-bd42-6bacc4af726f", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Context (Trends) -> The Trigger Event -> The Window of Opportunity", "description": "Create temporal urgency by proving that the window of opportunity is opening or closing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8b5e319977cd", "evidence": "Five contiguous slides with $150B China, €1.5B France, $2B Korea, €20B EU, £603M UK strategies.", "position": 18, "objective": "Argue national AI strategies create urgent geopolitical window", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 141, "type": "setup", "title": "The slide uses a photo of Lee Sedol playing AlphaGo to contextualize the South Korean AI focus.", "function": "establish_context", "rawType": "country_insights", "block": null, "metadata": { "slideType": "country_insights", "slideTypeCanon": { "name": "Country insights", "slug": "country_insights", "status": "active", "canonId": "019de52d-0e8d-7722-b4f8-bcbe32cb77b7", "version": 1, "description": null }, "function": "establish_context", "functionCanon": { "name": "Establish context", "slug": "establish_context", "status": "active", "canonId": "019de52d-1314-74e6-9074-d213dc0bdcdf", "version": 1, "description": null }, "density": "balanced", "densityScore": 47, "componentCount": 6, "textChars": 446, "nDataPoints": 4, "notes": "The slide uses a photo of Lee Sedol playing AlphaGo to contextualize the South Korean AI focus.", "elementsJson": [ "headline_text", "bullet_list", "photo" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:09.301+00", "seconds": 2.278, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/141", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-141", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Overall goal to reach “the global top 4 by 2022”.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:30+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-2e7f8886309b", "parentComponentId": null }, { "bbox": { "h": 0.45, "w": 0.4, "x": 0.56, "y": 0.318 }, "kind": "image", "text": "Lee Sedol playing AlphaGo", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.301+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "242741e4-f97e-4846-a955-83265c54449a", "parentComponentId": null }, { "bbox": { "h": 0.35, "w": 0.45, "x": 0.04, "y": 0.28 }, "kind": "list", "text": "Expands the existing 2016 AI plan to $2 billion through 2022\nAnnounces 6 new AI institutes\nPlan to award 4,500 domestic AI scholarships by 2022\n$1 billion fund for semiconductors through 2029\nOverall goal to reach “the global top 4 by 2022”", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.301+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "540be204-cb31-497b-8d3e-79d3faae651e", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Investment amount: $2 billion", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:30+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-31f2edeb56f8", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.5, "x": 0.04, "y": 0.14 }, "kind": "title", "text": "AI Nationalism: flurry of National AI strategies announced", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.301+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "5e829feb-3747-4366-9be5-0e03d11f7a05", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.35, "x": 0.04, "y": 0.22 }, "kind": "title", "text": "South Korea (announced March 2016 & May 2018)", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Subtitle", "slug": "subtitle", "status": "active", "canonId": "019de52c-fb09-739d-900c-815b0d9ac526", "version": 1, "description": null }, "framework": null }, "subkind": "subtitle", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.301+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "83a1bc58-1ca9-4c3d-98f1-771bcd121ba7", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-45a83a93428b", "evidence": "$2B Korea, callout 'global top 4 by 2022'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 141, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "2eab976a-16b0-499f-9380-2152fb7b8013", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.184084+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 141, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 143, "from": 139, "loop": { "name": "15_why_now", "slug": "15-why-now", "status": "active", "bestFor": "Sales pitches, fundraising, requesting immediate budget approval", "canonId": "019dd956-6c00-70cd-bd42-6bacc4af726f", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Context (Trends) -> The Trigger Event -> The Window of Opportunity", "description": "Create temporal urgency by proving that the window of opportunity is opening or closing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8b5e319977cd", "evidence": "Five contiguous slides with $150B China, €1.5B France, $2B Korea, €20B EU, £603M UK strategies.", "position": 18, "objective": "Argue national AI strategies create urgent geopolitical window", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 142, "type": "analysis", "title": "Part of a larger report series; focuses on EU policy.", "function": "summarize", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "balanced", "densityScore": 40, "componentCount": 5, "textChars": 471, "nDataPoints": 2, "notes": "Part of a larger report series; focuses on EU policy.", "elementsJson": [ "headline_text", "bullet_list", "infographic" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:09.236+00", "seconds": 2.169, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/142", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-142", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.25, "w": 0.4, "x": 0.56, "y": 0.42 }, "kind": "image", "text": "European Commission Digital Single Market Artificial Intelligence for Europe graphic", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Infographic", "slug": "infographic", "status": "active", "canonId": "019de52c-fe26-750c-9d7d-9f34e21324e8", "version": 1, "description": null }, "framework": null }, "subkind": "infographic", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.236+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9c4c9781-74f1-456b-8805-e59677f56b28", "parentComponentId": null }, { "bbox": { "h": 0.3, "w": 0.5, "x": 0.02, "y": 0.28 }, "kind": "list", "text": "Called for €20 billion investment\nPledged to increase spending to €1.5 billion through 2020 via EU research programme Horizon 2020\nCommits to presenting ethical guidelines on AI by end 2018\nPlan to update rules on use of public sector data to train ML systems", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.236+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "548abc84-1df3-487e-90a9-29539acada21", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Investment: €20 billion", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-9708f7bdc13f", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.4, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "European Commission plan (announced April 2018)", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.236+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d06cdcf0-d7bd-40b7-a8d9-f074d003b751", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.6, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI Nationalism: flurry of National AI strategies announced", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.236+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "050cf3ac-666a-485f-a9af-91d932efc835", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-4a6a305b0f82", "evidence": "€20B EU AI investment figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 142, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "b994cc10-2e9e-461e-adff-8e23af92bdb3", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.251146+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 142, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 143, "from": 139, "loop": { "name": "15_why_now", "slug": "15-why-now", "status": "active", "bestFor": "Sales pitches, fundraising, requesting immediate budget approval", "canonId": "019dd956-6c00-70cd-bd42-6bacc4af726f", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Context (Trends) -> The Trigger Event -> The Window of Opportunity", "description": "Create temporal urgency by proving that the window of opportunity is opening or closing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8b5e319977cd", "evidence": "Five contiguous slides with $150B China, €1.5B France, $2B Korea, €20B EU, £603M UK strategies.", "position": 18, "objective": "Argue national AI strategies create urgent geopolitical window", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 143, "type": "analysis", "title": "The slide uses a photo of a government official (Matt Hancock) to contextualize the policy announcement.", "function": "illustrate_case", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "balanced", "densityScore": 36, "componentCount": 4, "textChars": 508, "nDataPoints": 5, "notes": "The slide uses a photo of a government official (Matt Hancock) to contextualize the policy announcement.", "elementsJson": [ "headline_text", "bullet_list", "photo" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:11.813+00", "seconds": 4.688, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/143", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-143", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.612, "w": 0.394, "x": 0.58, "y": 0.288 }, "kind": "image", "text": "Photo of Matt Hancock speaking at a podium", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.813+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b3bf6442-1a80-4fff-bfcf-f366efafc2aa", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.48, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "U.K. AI Sector Deal (announced April 2018)\nGovernment committed to train 8000 computer science teachers and fund 1000 AI-related PhDs by 2025\n£603 million in newly allocated government funding and £300 million in matched private sector funding\nInvestment of £93 million in robotics and AI in extreme environments challenge (for use in industries like nuclear energy and space)", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.813+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f659f4dc-1845-4a09-826a-4aa34ec426d1", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Government funding: £603 million", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:32+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-b645caf165ab", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.6, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "AI Nationalism: flurry of National AI strategies announced", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.813+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "09d7ab0d-06c3-4d23-acfb-6a66ebceae72", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-4fdee9adf969", "evidence": "£603M UK strategy figure.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 143, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "3e9cc291-6a72-44a2-92ea-999cf1141c1e", "evidence": "The slide contains a list/bullet component with text.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.316262+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 143, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 143, "from": 139, "loop": { "name": "15_why_now", "slug": "15-why-now", "status": "active", "bestFor": "Sales pitches, fundraising, requesting immediate budget approval", "canonId": "019dd956-6c00-70cd-bd42-6bacc4af726f", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Context (Trends) -> The Trigger Event -> The Window of Opportunity", "description": "Create temporal urgency by proving that the window of opportunity is opening or closing", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8b5e319977cd", "evidence": "Five contiguous slides with $150B China, €1.5B France, $2B Korea, €20B EU, £603M UK strategies.", "position": 18, "objective": "Argue national AI strategies create urgent geopolitical window", "confidence": 75, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 144, "type": "setup", "title": "The table compares China and USA across multiple metrics, highlighting US dominance in most areas except for mobile users (data).", "function": "compare_peers", "rawType": "comparison_table", "block": null, "metadata": { "slideType": "comparison_table", "slideTypeCanon": { "name": "Comparison table", "slug": "comparison_table", "status": "active", "canonId": "019de52d-0a06-73f8-8813-fe022fab1de2", "version": 1, "description": null }, "function": "compare_peers", "functionCanon": { "name": "Compare peers", "slug": "compare_peers", "status": "active", "canonId": "019de52d-1453-72ba-a367-9ba3d34c1630", "version": 1, "description": null }, "density": "sparse", "densityScore": 17, "componentCount": 3, "textChars": 130, "nDataPoints": 18, "notes": "The table compares China and USA across multiple metrics, highlighting US dominance in most areas except for mobile users (data).", "elementsJson": [ "comparison_table" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:09.173+00", "seconds": 1.907, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/144", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-144", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "metric", "text": "Market share: 50%", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-82cf54abd751", "parentComponentId": null }, { "bbox": { "h": 0.68, "w": 0.68, "x": 0.16, "y": 0.218 }, "kind": "table", "text": "Main Driver in AI | Proxy Measure(s) | China | USA", "attrs": null, "canon": { "kind": { "name": "Table", "slug": "table", "status": "active", "canonId": "019de52c-f9c8-7148-8119-eaac9594cd28", "version": 1, "description": "Tabular data layout." }, "tool": null, "subkind": { "name": "Data table", "slug": "data", "status": "active", "canonId": "019de52d-02d5-75b8-8abf-88e92bf63569", "version": 1, "description": null }, "framework": null }, "subkind": "data", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.173+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "a7adefa7-ad5b-4a5c-a3fe-0038136475d2", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "For now, US leads China on almost every measure other than data", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.173+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "08800e82-07ac-4445-a332-274b2b075f50", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "MECE Principle", "slug": "mece-principle", "status": "active", "bestFor": null, "canonId": "019dd956-7c60-71a8-8758-c79d914489c9", "version": 1, "bodyDocId": null, "description": "Mutually Exclusive, Collectively Exhaustive - ensuring complete, non-overlapping structure", "familyLabel": null, "categoryName": "Block", "categorySlug": "block" }, "agent": "Architect", "layer": "block", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-58f86578b64c", "evidence": "Driver categories (Hardware/Data/Research/Commercial) structure the table.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 144, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Contrast Pairs", "slug": "contrast-pairs", "status": "active", "bestFor": null, "canonId": "019dd956-9b77-70d3-88b1-f34e26f60860", "version": 1, "bodyDocId": null, "description": "Creating tension through opposites: Before↔After, Problem↔Solution, Pain↔Relief", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Storyteller", "layer": "loop", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-549506a89ada", "evidence": "Side-by-side US/China table — explicit contrast pair.", "pageRefs": null, "priority": "Core", "whenToUse": null, "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 144, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-53f908119754", "evidence": "Title 'For now, US leads China on almost every measure other than data'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 90, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 144, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [ { "matchId": "019dd95a-1ca5-70bb-bba0-6c779bc4c430", "evidence": "Side-by-side table across Hardware, Data, Research, Commercial drivers.", "framework": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:34.826634+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 144, "frameworkName": "US vs China AI Capability Benchmark" } ], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 149, "from": 144, "loop": { "name": "04_tale_two_worlds", "slug": "04-tale-two-worlds", "status": "active", "bestFor": "Competitive analysis, benchmarking, case for change", "canonId": "019dd956-674f-750b-b15d-f082e2717f4e", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Current State -> Desired State / Benchmark -> The Gap & Implication", "description": "Show the gap between two states to drive urgency or highlight opportunity", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8fdb0c2bbdc3", "evidence": "Side-by-side US/China table on every dimension -> CFIUS scrutiny -> China import gap -> chip industry size.", "position": 19, "objective": "Compare US vs China AI capability head to head", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 145, "type": "analysis", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "dense", "densityScore": 56, "componentCount": 7, "textChars": 435, "nDataPoints": 8, "notes": null, "elementsJson": [ "headline_text", "action_title", "bar_chart_vertical", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:09.818+00", "seconds": 2.312, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/145", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-145", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "From 2008 to 2015, CFIUS investigations into foreign acquisitions nearly tripled.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-afd34111b62f", "parentComponentId": null }, { "bbox": { "h": 0.4, "w": 0.48, "x": 0.23, "y": 0.45 }, "kind": "chart", "text": "Bar chart showing investigations from 2008 to 2015", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.818+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "65ebd982-7ef2-40f0-b731-ec909134b1c7", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "CFIUS investigations: 65", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-b3f978fe99e4", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.5, "x": 0.23, "y": 0.35 }, "kind": "paragraph", "text": "From 2008 to 2015, CFIUS investigations into foreign acquisitions nearly tripled.", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.818+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "b727204f-cdd9-449a-b040-c905a4844759", "parentComponentId": null }, { "bbox": { "h": 0.02, "w": 0.2, "x": 0.23, "y": 0.88 }, "kind": "source-note", "text": "SOURCE: Annual Report to Congress, CFIUS", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.818+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9243614f-343f-49b4-a290-857ab840ebd9", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.13 }, "kind": "title", "text": "America increasingly using the Committee on Foreign Investment in the United States (CFIUS) to scrutinise foreign acquisitions", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.818+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "821a4ce1-1b84-4c4b-8d46-eac434e7273f", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.3, "x": 0.23, "y": 0.3 }, "kind": "title", "text": "Foreign investment investigations", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Subtitle", "slug": "subtitle", "status": "active", "canonId": "019de52c-fb09-739d-900c-815b0d9ac526", "version": 1, "description": null }, "framework": null }, "subkind": "subtitle", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:09.818+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d2c1ff0f-f512-4e53-ac1f-c77f129c0b79", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-5f5053365ad2", "evidence": "Title 'America increasingly using CFIUS to scrutinise foreign acquisitions'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 145, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 149, "from": 144, "loop": { "name": "04_tale_two_worlds", "slug": "04-tale-two-worlds", "status": "active", "bestFor": "Competitive analysis, benchmarking, case for change", "canonId": "019dd956-674f-750b-b15d-f082e2717f4e", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Current State -> Desired State / Benchmark -> The Gap & Implication", "description": "Show the gap between two states to drive urgency or highlight opportunity", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8fdb0c2bbdc3", "evidence": "Side-by-side US/China table on every dimension -> CFIUS scrutiny -> China import gap -> chip industry size.", "position": 19, "objective": "Compare US vs China AI capability head to head", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 146, "type": "evidence", "title": "The slide uses two photo-based case studies to illustrate political intervention in the semiconductor industry.", "function": "illustrate_case", "rawType": "case_study", "block": null, "metadata": { "slideType": "case_study", "slideTypeCanon": { "name": "Case study", "slug": "case_study", "status": "active", "canonId": "019de52d-0b95-7344-8090-926f3b8e92c8", "version": 1, "description": null }, "function": "illustrate_case", "functionCanon": { "name": "Illustrate case", "slug": "illustrate_case", "status": "active", "canonId": "019de52d-13b4-772c-bb73-1873612a7463", "version": 1, "description": null }, "density": "dense", "densityScore": 54, "componentCount": 7, "textChars": 550, "nDataPoints": 0, "notes": "The slide uses two photo-based case studies to illustrate political intervention in the semiconductor industry.", "elementsJson": [ "headline_text", "photo", "photo" ], "confidence": 0.95, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:13.421+00", "seconds": 5.911, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/146", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-146", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.5, "w": 0.4, "x": 0.5, "y": 0.37 }, "kind": "image", "text": "Lattice Semiconductor building", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.421+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "de900197-d189-4e8a-bec1-5370ce977179", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.05, "y": 0.37 }, "kind": "image", "text": "Qualcomm building", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Photo", "slug": "photo", "status": "active", "canonId": "019de52c-fd37-757f-b4a8-ca26d4888875", "version": 1, "description": null }, "framework": null }, "subkind": "photo", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.421+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e074b946-62e2-4fed-b262-193cfa7ad2bb", "parentComponentId": null }, { "bbox": null, "kind": "quote", "text": "Lattice Semiconductor offices in San Jose, Calif., in 2007. President Trump prevented the acquisition of Lattice by an investor group with ties to Beijing. Eugene Zelenko", "attrs": null, "canon": { "kind": { "name": "Quote", "slug": "quote", "status": "active", "canonId": "019de52c-f8fb-751e-8bc6-d58e2d59562a", "version": 1, "description": "Pull quote or testimonial." }, "tool": { "name": "Authority citation", "slug": "authority-citation", "status": "active", "bestFor": null, "canonId": "019dd9e1-49ae-7557-89ad-3fca70469b45", "version": 1, "bodyDocId": null, "description": "Quoting an authority figure, study, or precedent to lend weight.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-f8404798b96e", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.4, "x": 0.5, "y": 0.89 }, "kind": "source-note", "text": "Lattice Semiconductor offices in San Jose, Calif., in 2007. President Trump prevented the acquisition of Lattice by an investor group with ties to Beijing. Eugene Zelenko", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.421+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "22e4d910-5ec4-4d44-ab57-e7cbf8048a4e", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "CFIUS used to block two key semiconductor acquisitions in the last year", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.421+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "8d6b969e-5af1-436d-9231-d7d76d5094d7", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.3, "x": 0.05, "y": 0.25 }, "kind": "title", "text": "Trump Blocks Broadcom's Bid for Qualcomm", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Subtitle", "slug": "subtitle", "status": "active", "canonId": "019de52c-fb09-739d-900c-815b0d9ac526", "version": 1, "description": null }, "framework": null }, "subkind": "subtitle", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.421+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "22df0c22-dd87-4fc7-ba49-419f931baaa6", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.3, "x": 0.5, "y": 0.25 }, "kind": "title", "text": "Trump Blocks China-Backed Bid to Buy U.S. Chip Maker", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Subtitle", "slug": "subtitle", "status": "active", "canonId": "019de52c-fb09-739d-900c-815b0d9ac526", "version": 1, "description": null }, "framework": null }, "subkind": "subtitle", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.421+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fda274fc-787e-4c23-bf0e-2f71aa1c9f27", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 149, "from": 144, "loop": { "name": "04_tale_two_worlds", "slug": "04-tale-two-worlds", "status": "active", "bestFor": "Competitive analysis, benchmarking, case for change", "canonId": "019dd956-674f-750b-b15d-f082e2717f4e", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Current State -> Desired State / Benchmark -> The Gap & Implication", "description": "Show the gap between two states to drive urgency or highlight opportunity", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8fdb0c2bbdc3", "evidence": "Side-by-side US/China table on every dimension -> CFIUS scrutiny -> China import gap -> chip industry size.", "position": 19, "objective": "Compare US vs China AI capability head to head", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 147, "type": "analysis", "title": "The chart uses a combination of bar charts (for absolute dollar values) and line charts (for percentage of total imports).", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 32, "componentCount": 5, "textChars": 235, "nDataPoints": 24, "notes": "The chart uses a combination of bar charts (for absolute dollar values) and line charts (for percentage of total imports).", "elementsJson": [ "bar_chart_grouped", "line_chart" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:10.445+00", "seconds": 2.503, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/147", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-147", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Semiconductors top oil as China's No. 1 import", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-130173fe27b1", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.65, "x": 0.18, "y": 0.25 }, "kind": "chart", "text": "Semiconductors top oil as China's No. 1 import", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Grouped bar chart", "slug": "bar-grouped", "status": "active", "canonId": "019de52c-ff3b-751e-a8ff-f21ff11097b1", "version": 1, "description": null }, "framework": null }, "subkind": "bar-grouped", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.445+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e95a2c56-aad0-4712-b1d6-96e4d48cae26", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Import value in billions of dollars: $260 billion", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-140008eaf4ac", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.15, "x": 0.18, "y": 0.85 }, "kind": "source-note", "text": "Sources: Natixis, CEIC", "attrs": null, "canon": { "kind": { "name": "Source note", "slug": "source-note", "status": "active", "canonId": "23484952-494b-4a5e-b1ad-550d0d70e948", "version": 1, "description": "Attribution / source / caveat placed at the slide footer. Numbered footnotes use attrs.numbered." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.445+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "434f9631-6bbf-44c2-b710-b30395c4dd7a", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Why: China's annual imports of semiconductors have risen to $260 billion", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.445+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "44f42b44-30cd-437a-8a10-ea9b1c50437d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-63ae8fd05bb4", "evidence": "Title 'China's annual imports of semiconductors have risen to $260 billion'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 88, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 147, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 149, "from": 144, "loop": { "name": "04_tale_two_worlds", "slug": "04-tale-two-worlds", "status": "active", "bestFor": "Competitive analysis, benchmarking, case for change", "canonId": "019dd956-674f-750b-b15d-f082e2717f4e", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Current State -> Desired State / Benchmark -> The Gap & Implication", "description": "Show the gap between two states to drive urgency or highlight opportunity", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8fdb0c2bbdc3", "evidence": "Side-by-side US/China table on every dimension -> CFIUS scrutiny -> China import gap -> chip industry size.", "position": 19, "objective": "Compare US vs China AI capability head to head", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 148, "type": "tension", "title": "The slide uses a red bar and a callout to emphasize the disparity between China and the U.S. in 2017.", "function": "frame_problem", "rawType": "problem_statement", "block": null, "metadata": { "slideType": "problem_statement", "slideTypeCanon": { "name": "Problem statement", "slug": "problem_statement", "status": "active", "canonId": "019de52d-0879-768d-bba8-8da8a5a121d4", "version": 1, "description": null }, "function": "frame_problem", "functionCanon": { "name": "Frame problem", "slug": "frame_problem", "status": "active", "canonId": "019de52d-13db-753e-ad70-e9b2d0dab398", "version": 1, "description": null }, "density": "balanced", "densityScore": 39, "componentCount": 6, "textChars": 261, "nDataPoints": 6, "notes": "The slide uses a red bar and a callout to emphasize the disparity between China and the U.S. in 2017.", "elementsJson": [ "bar_chart_horizontal", "callout_box" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:10.474+00", "seconds": 2.392, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/148", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-148", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.1, "w": 0.3, "x": 0.42, "y": 0.65 }, "kind": "callout", "text": "China's chip industry still roughly nine times smaller than that of the U.S.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": null, "subkind": { "name": "Primary callout", "slug": "primary", "status": "active", "canonId": "019de52c-fbd1-70eb-a9ce-8997f8674c1d", "version": 1, "description": null }, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.474+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7325742b-eece-4195-b9d1-9ff62ad74b26", "parentComponentId": null }, { "bbox": { "h": 0.6, "w": 0.6, "x": 0.19, "y": 0.25 }, "kind": "chart", "text": "Chip industry revenue by country/region (in billions of dollars)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Horizontal bar chart", "slug": "bar-horizontal", "status": "active", "canonId": "019de52c-fec4-7221-986b-1cb5f5e84b1f", "version": 1, "description": null }, "framework": null }, "subkind": "bar-horizontal", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.474+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4d663bd9-9a5d-45fa-8ee6-3da8046388e3", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "Chip industry revenue: 24.7", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-319a5a31bcd1", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.75, "y": 0.82 }, "kind": "paragraph", "text": "Source: IEK", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.474+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "00ec18bb-155d-4dbe-b6a2-f8ecdf19f8ad", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.1, "x": 0.2, "y": 0.82 }, "kind": "paragraph", "text": "As of 2017", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.474+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0e20ab3d-c665-4bbc-bee4-9e7ff75a8687", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.8, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Why: China's semiconductor industry is small compared to that of the U.S.", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.474+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e967d410-222b-46ac-a49b-95ffac6ac846", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-66bf07ca8af2", "evidence": "Title 'China's semiconductor industry is small compared to that of the U.S.'", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 148, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 149, "from": 144, "loop": { "name": "04_tale_two_worlds", "slug": "04-tale-two-worlds", "status": "active", "bestFor": "Competitive analysis, benchmarking, case for change", "canonId": "019dd956-674f-750b-b15d-f082e2717f4e", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Current State -> Desired State / Benchmark -> The Gap & Implication", "description": "Show the gap between two states to drive urgency or highlight opportunity", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8fdb0c2bbdc3", "evidence": "Side-by-side US/China table on every dimension -> CFIUS scrutiny -> China import gap -> chip industry size.", "position": 19, "objective": "Compare US vs China AI capability head to head", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 149, "type": "analysis", "title": "The slide uses a downward arrow annotation to highlight the impact of U.S. policy on M&A activity.", "function": "analyze_data", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "analyze_data", "functionCanon": { "name": "Analyze data", "slug": "analyze_data", "status": "active", "canonId": "019de52d-1590-7413-8c82-cc218cff40b4", "version": 1, "description": null }, "density": "balanced", "densityScore": 38, "componentCount": 6, "textChars": 239, "nDataPoints": 4, "notes": "The slide uses a downward arrow annotation to highlight the impact of U.S. policy on M&A activity.", "elementsJson": [ "bar_chart_vertical", "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:11.021+00", "seconds": 2.848, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/149", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-149", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "Increased U.S. protectionism", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-cb3956c249f8", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.4, "x": 0.28, "y": 0.28 }, "kind": "chart", "text": "Cross-border semiconductor M&As by Chinese companies (in billions of dollars)", "attrs": null, "canon": { "kind": { "name": "Chart", "slug": "chart", "status": "active", "canonId": "019de52c-f9a0-76ef-adc6-0b859109d1f8", "version": 1, "description": "Quantitative chart." }, "tool": null, "subkind": { "name": "Vertical bar chart", "slug": "bar-vertical", "status": "active", "canonId": "019de52c-fe9d-70f2-b3b7-a9ae294f5917", "version": 1, "description": null }, "framework": null }, "subkind": "bar-vertical", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.021+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0bf66232-3af9-4057-8176-e4fe68d82998", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.2, "x": 0.28, "y": 0.9 }, "kind": "disclaimer", "text": "Sources: Natixis, Bloomberg", "attrs": null, "canon": { "kind": { "name": "Disclaimer", "slug": "disclaimer", "status": "active", "canonId": "91173477-d91f-4024-bbaf-d4f93e3f82dc", "version": 1, "description": "Legal/cautionary text that occupies most of the body of its slide." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.021+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0f89f4a9-629f-4d3f-8672-8f5cbd07cdbf", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.15, "x": 0.58, "y": 0.65 }, "kind": "legend", "text": "Increased U.S. protectionism", "attrs": null, "canon": { "kind": { "name": "Legend", "slug": "legend", "status": "active", "canonId": "31361933-3d22-4c70-bbad-0801e9fffc0c", "version": 1, "description": "Color/symbol decoder placed adjacent to a chart or diagram." }, "tool": null, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.021+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "26d9617f-6d11-4eb3-bb52-cacebddb77b1", "parentComponentId": null }, { "bbox": null, "kind": "metric", "text": "M&A value", "attrs": null, "canon": { "kind": { "name": "Metric", "slug": "metric", "status": "active", "canonId": "019de52c-f923-73bf-8081-d0b1a8c5264d", "version": 1, "description": "Big-number or KPI value." }, "tool": { "name": "Quantification", "slug": "quantification", "status": "active", "bestFor": null, "canonId": "019dd9e1-47ab-72ce-a079-aab1f5299e88", "version": 1, "bodyDocId": null, "description": "Translating a claim into numbers, charts, or tables.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": "primary", "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a818-cf49e767101d", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.7, "x": 0.05, "y": 0.15 }, "kind": "title", "text": "Why: China has been actively acquiring foreign semiconductor companies", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.021+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "9f1d5707-6be7-480f-9bff-cd2193974f7d", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Chart Selection Guide", "slug": "chart-selection-guide", "status": "active", "bestFor": null, "canonId": "019dd956-b1a0-700b-ba17-0e16c4d5bcae", "version": 1, "bodyDocId": null, "description": "Choosing the right chart type for your data and message", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Designer", "layer": "slide", "agents": [ "Designer" ], "matchId": "980b92b3-ea06-45f0-8a45-ae88357d2ea5", "evidence": "chart/bar-vertical: Cross-border semiconductor M&As by Chinese companies (in billions of dollars)", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.124149+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 149, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 149, "from": 144, "loop": { "name": "04_tale_two_worlds", "slug": "04-tale-two-worlds", "status": "active", "bestFor": "Competitive analysis, benchmarking, case for change", "canonId": "019dd956-674f-750b-b15d-f082e2717f4e", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "Current State -> Desired State / Benchmark -> The Gap & Implication", "description": "Show the gap between two states to drive urgency or highlight opportunity", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-8fdb0c2bbdc3", "evidence": "Side-by-side US/China table on every dimension -> CFIUS scrutiny -> China import gap -> chip industry size.", "position": 19, "objective": "Compare US vs China AI capability head to head", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 150, "type": "analysis", "title": "The slide uses a sequence of images showing Theresa May to illustrate the progression of synthetic video manipulation.", "function": "frame_problem", "rawType": "industry_trends", "block": null, "metadata": { "slideType": "industry_trends", "slideTypeCanon": { "name": "Industry trends", "slug": "industry_trends", "status": "active", "canonId": "019de52d-0b1d-7705-a0e0-c663199cd1fa", "version": 1, "description": null }, "function": "frame_problem", "functionCanon": { "name": "Frame problem", "slug": "frame_problem", "status": "active", "canonId": "019de52d-13db-753e-ad70-e9b2d0dab398", "version": 1, "description": null }, "density": "balanced", "densityScore": 35, "componentCount": 5, "textChars": 458, "nDataPoints": 0, "notes": "The slide uses a sequence of images showing Theresa May to illustrate the progression of synthetic video manipulation.", "elementsJson": [ "screenshot", "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:13.158+00", "seconds": 4.982, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/150", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-150", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "This has very significant implications for enabling those who are engaged in producing disinformation and propaganda.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:31+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a819-57e5775bca5b", "parentComponentId": null }, { "bbox": { "h": 0.5, "w": 0.9, "x": 0.05, "y": 0.25 }, "kind": "image", "text": "Sequence of four images showing source video and three stages of editing: Expression, Pose, and Expression + Pose + Blinking.", "attrs": null, "canon": { "kind": { "name": "Image", "slug": "image", "status": "active", "canonId": "019de52c-f978-701f-a720-51d029d769e7", "version": 1, "description": "Photo, illustration, screenshot or icon." }, "tool": null, "subkind": { "name": "Illustration", "slug": "illustration", "status": "active", "canonId": "019de52c-fe76-7772-a5ac-c0b7294d7615", "version": 1, "description": null }, "framework": null }, "subkind": "illustration", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.158+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "22aaf10f-ac66-4548-8113-ec4a87ccad6b", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.9, "x": 0.05, "y": 0.8 }, "kind": "list", "text": "This has very significant implications for enabling those who are engaged in producing disinformation and propaganda.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.158+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "7e56b06a-cc31-44a2-9c9c-f936289664e4", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.82, "y": 0.92 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.158+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "28361b84-cfbd-4387-9686-17462d14d988", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.9, "x": 0.05, "y": 0.14 }, "kind": "title", "text": "Fake views: Generating synthetic video is getting cheaper, easier and more realistic", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:13.158+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "05a724c7-31dd-41fb-a67d-51c30867f62a", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-6bd1ec0b98ad", "evidence": "Title 'Fake views: Generating synthetic video is getting cheaper, easier and more realistic'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 85, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 150, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Emotional Appeal", "slug": "emotional-appeal", "status": "active", "bestFor": null, "canonId": "019dd956-bde9-7668-8be7-0097678fec59", "version": 1, "bodyDocId": null, "description": "Connecting to audience feelings rather than just logic", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-6cd5adb00ef3", "evidence": "Disinformation/propaganda framing.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 150, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 150, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 70, "name": "The Implications (So What)", "beatId": "019dd95a-0682-776c-8e34-b35a0f5ce04f", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Implications (So What)", "slug": "triple-take-the-implications-so-what", "status": "active", "canonId": "019dd9b8-09c7-75cd-8c7b-ae0e23cca2b8", "version": 1, "description": null }, "evidence": "Industry applications and political/labor consequences across verticals and geographies.", "position": 2, "isPrimaryArc": true, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 150, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 109, "name": "Core Insight", "beatId": "019dd95a-0682-776c-8e34-c5963b5d1d13", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Core Insight", "slug": "onion-core-insight", "status": "active", "canonId": "019dd9b8-0d41-766a-8ad4-4ba6c6e72446", "version": 1, "description": null }, "evidence": "Politics: labor market, public attitudes, AI nationalism.", "position": 4, "isPrimaryArc": false, "parentBeatType": { "name": "Turn", "slug": "turn", "status": "active", "canonId": "019dd9b8-0396-702b-ac2f-9901f8dd8b54", "version": 1, "description": "The pivot — solution proposed, catalyst introduced, decision made. Common (15/20 arcs) but not universal: Triple-take, AIDA, PAS, Voyage-return and Pixar substitute it with a strong reflection or development beat." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 151, "type": "setup", "function": "transition", "rawType": "section_divider", "block": null, "metadata": { "slideType": "section_divider", "slideTypeCanon": { "name": "Section divider", "slug": "section_divider", "status": "active", "canonId": "019de52d-066a-7402-9554-6ae504999041", "version": 1, "description": null }, "function": "transition", "functionCanon": { "name": "Transition", "slug": "transition", "status": "active", "canonId": "019de52d-1365-7149-92aa-66df859e75fd", "version": 1, "description": null }, "density": "sparse", "densityScore": 14, "componentCount": 4, "textChars": 127, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:15.155+00", "seconds": 6.65, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/151", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-151", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.03, "w": 0.08, "x": 0.9, "y": 0.95 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:15.155+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1293975d-4118-4739-9b49-f0dfde1bbe38", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.06, "x": 0.92, "y": 0.03 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:15.155+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "99e23317-69bb-41d5-9783-cc484b5c0bf0", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.75, "x": 0.02, "y": 0.03 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:15.155+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ba9b5c3b-28f4-4678-99c8-9484c94ef972", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.31, "x": 0.345, "y": 0.45 }, "kind": "title", "text": "Section 5: Predictions", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:15.155+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1d61646e-3e1d-46b8-bccc-0ab75949cb90", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "a4270a2d-6749-457c-b444-cab81c5af940", "evidence": "Section 5: Predictions", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:38.24304+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 151, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 156, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 151, "name": "The Action (Now What)", "beatId": "019dd95a-0682-776c-8e34-b4730e3b2f0d", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Action (Now What)", "slug": "triple-take-the-action-now-what", "status": "active", "canonId": "019dd9b8-0a5e-735a-bf53-3a1ba629b7aa", "version": 1, "description": null }, "evidence": "Section 5 '8 predictions for the next 12 months' and conclusion.", "position": 3, "isPrimaryArc": true, "parentBeatType": { "name": "Resolution", "slug": "resolution", "status": "active", "canonId": "019dd9b8-045d-703b-8dd7-e92aba3ba91b", "version": 1, "description": "How the deck ends. Required in 19/20 arcs. The exception (Sparkline) ends with new bliss which is functionally a resolution." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 156, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 151, "name": "Implications", "beatId": "019dd95a-0682-776c-8e34-cb622c4afe44", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Implications", "slug": "onion-implications", "status": "active", "canonId": "019dd9b8-0dda-7730-893d-deaac7e9c295", "version": 1, "description": null }, "evidence": "Predictions for next 12 months.", "position": 5, "isPrimaryArc": false, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 152, "type": "setup", "title": "The slide uses a numbered list format to present forward-looking statements.", "function": "summarize", "rawType": "initiative_list", "block": null, "metadata": { "slideType": "initiative_list", "slideTypeCanon": { "name": "Initiative list", "slug": "initiative_list", "status": "active", "canonId": "019de52d-0caf-770e-9b91-b99239121f50", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "balanced", "densityScore": 43, "componentCount": 3, "textChars": 1055, "nDataPoints": 2, "notes": "The slide uses a numbered list format to present forward-looking statements.", "elementsJson": [ "numbered_list" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:15.789+00", "seconds": 7.227, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/152", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-152", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.65, "w": 0.95, "x": 0.02, "y": 0.22 }, "kind": "list", "text": "1. A lab located in China makes a significant research breakthrough.\n2. DeepMind has a breakthrough result successfully applying RL to learn how to play Starcraft.\n3. Deep learning continues to dominate the discussion without major alternatives appearing.\n4. The first therapeutic drug discovered using machine learning produces positive results in trials.\n5. Chinese and American headquartered technology companies make acquisitions of machine learning companies based in Europe totalling over $5b.\n6. The government of an OECD country blocks the acquisition of a leading machine learning company (defined as valuation >$100m) by a US or Chinese headquartered technology company.\n7. Access to Taiwanese and South Korean semiconductor companies becomes an explicit part of the trade war between America and China.\n8. A major research institution “goes dark” by refraining from publishing key work in the open due to geopolitical concerns.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Numbered list", "slug": "numbered", "status": "active", "canonId": "019de52c-fcc0-7718-a284-a4250c805df2", "version": 1, "description": null }, "framework": null }, "subkind": "numbered", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:15.789+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "fb61634f-0eef-45ab-a8bf-2f6d65c6c443", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.75, "x": 0.02, "y": 0.03 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:15.789+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "bc2bd605-7a66-41ed-a1ee-4d36be21c4c5", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.4, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "8 predictions for the next 12 months", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:15.789+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "98d90eff-3ff6-4b0f-8c30-e39a1aa7d964", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Chunking", "slug": "chunking", "status": "active", "bestFor": null, "canonId": "019dd956-abab-72ce-b64c-2924cc83cf39", "version": 1, "bodyDocId": null, "description": "Information grouped into chunks is remembered better - structure content in digestible units", "familyLabel": null, "categoryName": "Loop", "categorySlug": "loop" }, "agent": "Architect", "layer": "loop", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-762d8a53b101", "evidence": "8 numbered predictions chunked into discrete bullets.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 75, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 152, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Action Titles", "slug": "action-titles", "status": "active", "bestFor": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "canonId": "019dd956-ac70-700b-a608-ba58e78f5dcc", "version": 1, "bodyDocId": "816dad3c-9983-41c1-a86e-978ebd582841", "description": "The McKinsey-bred discipline of writing every slide title as a complete declarative sentence with a verb and an insight, not a topic label. Each title is a sub-claim that ladders up to the deck's governing thought; read in sequence, the titles reconstruct the executive summary.", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Architect", "layer": "slide", "agents": [ "Architect" ], "matchId": "019dd95a-0fd5-7148-8ecc-737871a8f4f5", "evidence": "Title '8 predictions for the next 12 months'.", "pageRefs": null, "priority": "Core", "whenToUse": "Any decision-grade deliverable — strategy recommendation, board update, investment memo, M&A review, post-mortem — where the reader is being asked to agree, decide, or act and will skim the title bar at speed.", "confidence": 80, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 152, "whyItWorks": "Solves two failure modes at once. (1) The buried claim: action titles place the conclusion on the page before any analysis is read, so a busy reader doesn't reverse-engineer it from the chart. (2) The spineless deck: because each title is a claim, slides have to ladder up to the governing thought; logical gaps become visible as topic-shaped titles in a sequence that no longer reads as a story.", "antipattern": "Topic labels disguised as titles — Volume by quarter, Key findings, Pricing strategy, Margins have been impacted. No verb, or a passive verb with no agent, or a fact (Revenue grew 12%) without the so-what. Also: titles that overreach the chart's evidence, multi-claim X-grew-but-Y-fell welded titles, and mechanical repetition of the same X-drove-Y template across the deck.", "cardinality": null, "narrativePurpose": "Forces the deck to carry its argument in the title bar so a senior reader can extract the recommendation without opening a single slide; converts a binder of topics into a navigable pyramid where every slide is a node defending the apex." }, { "tool": { "name": "Closing Techniques", "slug": "closing-techniques", "status": "active", "bestFor": null, "canonId": "019dd956-af13-774b-a4fd-7db0a3a88d51", "version": 1, "bodyDocId": null, "description": "Strong endings: Callback, challenge, vision, clear CTA, memorable line", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-7917deca52c5", "evidence": "Forward-looking close to deck.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 65, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 152, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "List presentation", "slug": "list-presentation", "status": "active", "bestFor": null, "canonId": "019dd9e1-4ef7-777d-ba68-9074c2b3bcf1", "version": 1, "bodyDocId": null, "description": "Bulleted, numbered, or checklist enumerations.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "bbd0278b-42fa-4cd3-ba91-6789e363871d", "evidence": "list/numbered: 1. A lab located in China makes a significant research breakthrough.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.7, "extraction": { "at": "2026-07-16 21:56:38.323547+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 152, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 156, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 151, "name": "The Action (Now What)", "beatId": "019dd95a-0682-776c-8e34-b4730e3b2f0d", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Action (Now What)", "slug": "triple-take-the-action-now-what", "status": "active", "canonId": "019dd9b8-0a5e-735a-bf53-3a1ba629b7aa", "version": 1, "description": null }, "evidence": "Section 5 '8 predictions for the next 12 months' and conclusion.", "position": 3, "isPrimaryArc": true, "parentBeatType": { "name": "Resolution", "slug": "resolution", "status": "active", "canonId": "019dd9b8-045d-703b-8dd7-e92aba3ba91b", "version": 1, "description": "How the deck ends. Required in 19/20 arcs. The exception (Sparkline) ends with new bliss which is functionally a resolution." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 156, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 151, "name": "Implications", "beatId": "019dd95a-0682-776c-8e34-cb622c4afe44", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Implications", "slug": "onion-implications", "status": "active", "canonId": "019dd9b8-0dda-7730-893d-deaac7e9c295", "version": 1, "description": null }, "evidence": "Predictions for next 12 months.", "position": 5, "isPrimaryArc": false, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [ { "to": 152, "from": 152, "loop": { "name": "35_scenario_fork", "slug": "35-scenario-fork", "status": "active", "bestFor": "Strategic planning, risk analysis, board presentations", "canonId": "019dd956-7430-71ea-ac7d-1630063fb962", "version": 1, "bodyDocId": "019df22a-2420-77be-bc41-ded96d08cb21", "structure": "The Decision Point -> Scenario A (If X) -> Scenario B (If Y) -> Scenario C (If Z) -> Recommendation", "description": "Present multiple futures based on different assumptions or decisions", "familyLabel": null }, "matchId": "019dd95a-07fe-70ce-8d3c-90b4058f80f5", "evidence": "Numbered list of 8 distinct 'if X happens' predictions for next 12 months.", "position": 20, "objective": "Project eight forward scenarios for the year ahead", "confidence": 70, "extraction": { "at": "2026-04-29 13:07:29.547081+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null } } ], "locked": true }, { "page": 153, "type": "setup", "function": "transition", "rawType": "section_divider", "block": null, "metadata": { "slideType": "section_divider", "slideTypeCanon": { "name": "Section divider", "slug": "section_divider", "status": "active", "canonId": "019de52d-066a-7402-9554-6ae504999041", "version": 1, "description": null }, "function": "transition", "functionCanon": { "name": "Transition", "slug": "transition", "status": "active", "canonId": "019de52d-1365-7149-92aa-66df859e75fd", "version": 1, "description": null }, "density": "sparse", "densityScore": 14, "componentCount": 4, "textChars": 126, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:10.728+00", "seconds": 2.121, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/153", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-153", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.03, "w": 0.65, "x": 0.02, "y": 0.02 }, "kind": "paragraph", "text": "Introduction | Research | Talent | Industry | Politics | Predictions | Conclusion", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.728+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6dd96eba-6e38-490e-aa07-96ddd9f0c7a4", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.07, "x": 0.91, "y": 0.95 }, "kind": "paragraph", "text": "stateof.ai 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.728+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "ccf29eec-53d1-4a77-b416-563ec13aa0e1", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.06, "x": 0.92, "y": 0.02 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.728+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f6bbcb26-af1d-4fba-95d2-1b93fc6b8d68", "parentComponentId": null }, { "bbox": { "h": 0.035, "w": 0.306, "x": 0.347, "y": 0.458 }, "kind": "title", "text": "Section 6: Conclusion", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.728+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "12442a9a-2795-4401-82f9-f45131339f92", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Section divider", "slug": "section-divider", "status": "active", "bestFor": null, "canonId": "019dd9e1-4c66-707d-8d1f-6e14547134c7", "version": 1, "bodyDocId": null, "description": "A transitional slide marking the start of a new section.", "familyLabel": null, "categoryName": null, "categorySlug": null }, "agent": null, "layer": "slide", "agents": null, "matchId": "27fb5db9-f574-4ad3-bfa3-0b4448c07ae3", "evidence": "Section 6: Conclusion", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:38.401609+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 153, "whyItWorks": null, "antipattern": null, "cardinality": "common", "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 156, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 151, "name": "The Action (Now What)", "beatId": "019dd95a-0682-776c-8e34-b4730e3b2f0d", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Action (Now What)", "slug": "triple-take-the-action-now-what", "status": "active", "canonId": "019dd9b8-0a5e-735a-bf53-3a1ba629b7aa", "version": 1, "description": null }, "evidence": "Section 5 '8 predictions for the next 12 months' and conclusion.", "position": 3, "isPrimaryArc": true, "parentBeatType": { "name": "Resolution", "slug": "resolution", "status": "active", "canonId": "019dd9b8-045d-703b-8dd7-e92aba3ba91b", "version": 1, "description": "How the deck ends. Required in 19/20 arcs. The exception (Sparkline) ends with new bliss which is functionally a resolution." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 156, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 151, "name": "Implications", "beatId": "019dd95a-0682-776c-8e34-cb622c4afe44", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Implications", "slug": "onion-implications", "status": "active", "canonId": "019dd9b8-0dda-7730-893d-deaac7e9c295", "version": 1, "description": null }, "evidence": "Predictions for next 12 months.", "position": 5, "isPrimaryArc": false, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 154, "type": "setup", "function": "summarize", "rawType": "closing_ask", "block": null, "metadata": { "slideType": "closing_ask", "slideTypeCanon": { "name": "Closing ask", "slug": "closing_ask", "status": "active", "canonId": "019de52d-0aa4-750f-8aa1-8d622dd99f5c", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "balanced", "densityScore": 36, "componentCount": 3, "textChars": 914, "nDataPoints": 0, "notes": null, "elementsJson": [ "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:10.946+00", "seconds": 2.214, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/154", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-154", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "We would appreciate any and all feedback on how we could improve this report further.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:36+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a81b-aeb9753d7530", "parentComponentId": null }, { "bbox": { "h": 0.65, "w": 0.96, "x": 0.02, "y": 0.2 }, "kind": "paragraph", "text": "Congratulations on making it to the end! Thanks for reading. In this report, we set out to capture a snapshot of the exponential progress in the field of machine learning, with a focus on developments in the past 12 months. We believe that AI will be a force multiplier on technological progress in our world, and that wider understanding of the field is critical if we are to navigate such a huge transition. We tried to compile a snapshot of all the things that caught our attention in the last year across the range of machine learning research, commercialisation, talent and the emerging politics of AI. Thanks to Mary Meeker for the inspiration. We would appreciate any and all feedback on how we could improve this report further. Thanks again for reading! Nathan Benaich (@nathanbenaich) and Ian Hogarth (@soundboy)", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.946+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "f8dbdf74-de3c-4de8-9d6a-616904139584", "parentComponentId": null }, { "bbox": { "h": 0.04, "w": 0.1, "x": 0.02, "y": 0.14 }, "kind": "title", "text": "Thanks!", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:10.946+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "41c35b5b-00a4-42e0-8ea2-cbb0d4cb1b80", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Closing Techniques", "slug": "closing-techniques", "status": "active", "bestFor": null, "canonId": "019dd956-af13-774b-a4fd-7db0a3a88d51", "version": 1, "bodyDocId": null, "description": "Strong endings: Callback, challenge, vision, clear CTA, memorable line", "familyLabel": null, "categoryName": "Slide", "categorySlug": "slide" }, "agent": "Storyteller", "layer": "slide", "agents": [ "Storyteller" ], "matchId": "019dd95a-0fd5-7148-8ecc-7fb92e7ae0ce", "evidence": "Thanks slide invites feedback — explicit CTA.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 70, "extraction": { "at": "2026-04-29 13:07:31.544884+00", "model": "claude-legacy", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 154, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null }, { "tool": { "name": "Storytelling: the closing ask", "slug": "closing-ask", "status": "active", "bestFor": null, "canonId": "019df269-24e0-70ab-8dd9-983832c1c665", "version": 1, "bodyDocId": "019df269-24bc-7554-b88d-245127606b8d", "description": "The single most important slide in the deck — after the cover. The final", "familyLabel": "narrative-block", "categoryName": "Narrative & Transformation", "categorySlug": "narrative-transformation" }, "agent": "storyteller", "layer": "slide", "agents": [ "storyteller" ], "matchId": "09768c91-5896-492b-8b2e-eec9533a8129", "evidence": "callout: We would appreciate any and all feedback on how we could improve this report further.", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:38.483467+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 154, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 156, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 151, "name": "The Action (Now What)", "beatId": "019dd95a-0682-776c-8e34-b4730e3b2f0d", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Action (Now What)", "slug": "triple-take-the-action-now-what", "status": "active", "canonId": "019dd9b8-0a5e-735a-bf53-3a1ba629b7aa", "version": 1, "description": null }, "evidence": "Section 5 '8 predictions for the next 12 months' and conclusion.", "position": 3, "isPrimaryArc": true, "parentBeatType": { "name": "Resolution", "slug": "resolution", "status": "active", "canonId": "019dd9b8-045d-703b-8dd7-e92aba3ba91b", "version": 1, "description": "How the deck ends. Required in 19/20 arcs. The exception (Sparkline) ends with new bliss which is functionally a resolution." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 156, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 151, "name": "Implications", "beatId": "019dd95a-0682-776c-8e34-cb622c4afe44", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Implications", "slug": "onion-implications", "status": "active", "canonId": "019dd9b8-0dda-7730-893d-deaac7e9c295", "version": 1, "description": null }, "evidence": "Predictions for next 12 months.", "position": 5, "isPrimaryArc": false, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 155, "type": "setup", "function": "summarize", "rawType": "appendix_disclosure", "block": null, "metadata": { "slideType": "appendix_disclosure", "slideTypeCanon": { "name": "Appendix disclosure", "slug": "appendix_disclosure", "status": "active", "canonId": "019de52d-07b2-7549-8e8b-932a19f6e616", "version": 1, "description": null }, "function": "summarize", "functionCanon": { "name": "Summarize", "slug": "summarize", "status": "active", "canonId": "019de52d-1541-704f-b86f-bf7b75056373", "version": 1, "description": null }, "density": "balanced", "densityScore": 43, "componentCount": 5, "textChars": 699, "nDataPoints": 0, "notes": null, "elementsJson": [ "paragraph" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:11.218+00", "seconds": 2.433, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/155", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-155", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": null, "kind": "callout", "text": "The authors declare a number of conflicts of interest as a result of being investors and/or advisors, personally or via funds, in a number of private and public companies whose work is cited in this report.", "attrs": null, "canon": { "kind": { "name": "Callout", "slug": "callout", "status": "active", "canonId": "019de52c-f8d3-761a-8953-7d37080de5e2", "version": 1, "description": "Highlighted callout box." }, "tool": { "name": "Visual emphasis", "slug": "visual-emphasis", "status": "active", "bestFor": null, "canonId": "019dd9e1-474b-736b-a8eb-a794dd88241c", "version": 1, "bodyDocId": null, "description": "Drawing the eye to one element above the rest of the slide.", "familyLabel": null }, "subkind": null, "framework": null }, "subkind": null, "groupRole": null, "confidence": null, "extraction": { "at": "2026-04-25 12:01:36+00", "model": "or:google/gemini-3.1-flash-lite-preview", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "019dd952-41ac-7180-a81a-d986546087e2", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.96, "x": 0.02, "y": 0.61 }, "kind": "list", "text": "Public companies: Alphabet, NVIDIA, Facebook, Microsoft, Intel, Baidu, Amazon, and Alibaba.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "6afd621e-47ac-4b08-95ab-229e094bc2e1", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.96, "x": 0.02, "y": 0.42 }, "kind": "list", "text": "Startups: GTN.ai, TwentyBN, Kheiron Medical, Accelerated Dynamics, Avidbots, Optimal Labs, Ravelin, Tractable, LabGenius, and Mapillary.", "attrs": null, "canon": { "kind": { "name": "List", "slug": "list", "status": "active", "canonId": "019de52c-f94b-779b-a53b-fd4904a5f4d2", "version": 1, "description": "Bullet or numbered list." }, "tool": null, "subkind": { "name": "Bullet list", "slug": "bullet", "status": "active", "canonId": "019de52c-fc98-7169-a083-5cab805076ca", "version": 1, "description": null }, "framework": null }, "subkind": "bullet", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "97ff011d-4ef7-49bd-bfdb-f11fcc6edcfc", "parentComponentId": null }, { "bbox": { "h": 0.08, "w": 0.96, "x": 0.02, "y": 0.22 }, "kind": "paragraph", "text": "The authors declare a number of conflicts of interest as a result of being investors and/or advisors, personally or via funds, in a number of private and public companies whose work is cited in this report. This concerns the following companies:", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "e6096b18-06fa-43f1-bf99-369a4016c7ca", "parentComponentId": null }, { "bbox": { "h": 0.03, "w": 0.23, "x": 0.02, "y": 0.15 }, "kind": "title", "text": "Conflicts of interest", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.218+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0373a65b-45d1-4aca-9442-9e6144ff165a", "parentComponentId": null } ], "metrics": [], "tools": [], "frameworks": [], "arcBeats": [ { "to": 156, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 151, "name": "The Action (Now What)", "beatId": "019dd95a-0682-776c-8e34-b4730e3b2f0d", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Action (Now What)", "slug": "triple-take-the-action-now-what", "status": "active", "canonId": "019dd9b8-0a5e-735a-bf53-3a1ba629b7aa", "version": 1, "description": null }, "evidence": "Section 5 '8 predictions for the next 12 months' and conclusion.", "position": 3, "isPrimaryArc": true, "parentBeatType": { "name": "Resolution", "slug": "resolution", "status": "active", "canonId": "019dd9b8-045d-703b-8dd7-e92aba3ba91b", "version": 1, "description": "How the deck ends. Required in 19/20 arcs. The exception (Sparkline) ends with new bliss which is functionally a resolution." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 156, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 151, "name": "Implications", "beatId": "019dd95a-0682-776c-8e34-cb622c4afe44", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Implications", "slug": "onion-implications", "status": "active", "canonId": "019dd9b8-0dda-7730-893d-deaac7e9c295", "version": 1, "description": null }, "evidence": "Predictions for next 12 months.", "position": 5, "isPrimaryArc": false, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true }, { "page": 156, "type": "setup", "function": "front_matter", "rawType": "cover", "block": null, "metadata": { "slideType": "cover", "slideTypeCanon": { "name": "Cover", "slug": "cover", "status": "active", "canonId": "019de52d-05f2-729d-9f93-e0504f0a2410", "version": 1, "description": null }, "function": "front_matter", "functionCanon": { "name": "Front matter", "slug": "front_matter", "status": "active", "canonId": "019de52d-133d-73f0-9e26-201eda96b098", "version": 1, "description": null }, "density": "balanced", "densityScore": 26, "componentCount": 6, "textChars": 68, "nDataPoints": 0, "notes": null, "elementsJson": [ "headline_text" ], "confidence": 1, "extraction": { "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "at": "2026-05-02 17:44:11.589+00", "seconds": 2.763, "promptVersion": "v2-canon-conditioned" } }, "links": { "slide": "/slides/019dd923-5e88-73ef-bd5c-f341d4394195/156", "deck": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195", "deckAnchor": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195#slide-156", "deckJson": "/decks/019dd923-5e88-73ef-bd5c-f341d4394195.json" }, "components": [ { "bbox": { "h": 0.05, "w": 0.2, "x": 0.4, "y": 0.48 }, "kind": "paragraph", "text": "June 29, 2018", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "0091e2f4-f9f4-4a41-8756-56230bc2a8a7", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.1, "x": 0.02, "y": 0.02 }, "kind": "paragraph", "text": "stateof.ai", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "1181dd02-b34f-4f84-b4fa-3a2e1e02cf16", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.1, "x": 0.85, "y": 0.9 }, "kind": "paragraph", "text": "Ian Hogarth", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "2f772481-c777-47d9-b2e9-551ebb66bff3", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.15, "x": 0.05, "y": 0.9 }, "kind": "paragraph", "text": "Nathan Benaich", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "4900a2c0-7cfc-4c97-be7b-a1cc49d49c07", "parentComponentId": null }, { "bbox": { "h": 0.05, "w": 0.08, "x": 0.9, "y": 0.02 }, "kind": "paragraph", "text": "#AIreport", "attrs": null, "canon": { "kind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "5f2affc0-941d-4d07-959d-4bfc1930cea4", "version": 1, "description": "Single prose block in the body of a slide." }, "tool": null, "subkind": { "name": "Paragraph", "slug": "paragraph", "status": "active", "canonId": "019de52c-fb80-75af-aa5c-a7a4bfac2433", "version": 1, "description": null }, "framework": null }, "subkind": "paragraph", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "d41bb302-0d4f-4785-833d-392d79bf444d", "parentComponentId": null }, { "bbox": { "h": 0.1, "w": 0.3, "x": 0.35, "y": 0.35 }, "kind": "title", "text": "State of AI", "attrs": null, "canon": { "kind": { "name": "Title", "slug": "title", "status": "active", "canonId": "019de52c-f7dd-76d9-b848-b1f12bd85cf4", "version": 1, "description": "Slide title or headline." }, "tool": null, "subkind": { "name": "Headline", "slug": "headline", "status": "active", "canonId": "019de52c-fae1-7463-9a7c-d60668eced0d", "version": 1, "description": null }, "framework": null }, "subkind": "headline", "groupRole": null, "confidence": null, "extraction": { "at": "2026-05-02 17:44:11.589+00", "model": "google/gemini-3.1-flash-lite-preview-20260303", "runId": null, "seconds": null, "promptVersion": null }, "componentId": "32c93393-83c2-478d-9e9d-9fcc8e720be4", "parentComponentId": null } ], "metrics": [], "tools": [ { "tool": { "name": "Slide recipe: Cover", "slug": "cover-slide", "status": "active", "bestFor": null, "canonId": "019df269-2a08-767d-9128-e7431e6fe8a3", "version": 1, "bodyDocId": "019df269-29e5-7738-b5fa-6892e5dc8e65", "description": "The first ten seconds. Build it deliberately.", "familyLabel": "slide-recipe", "categoryName": "Slide", "categorySlug": "slide" }, "agent": "designer", "layer": "slide", "agents": [ "designer" ], "matchId": "06b05a93-8683-408e-9a39-df7027d63caa", "evidence": "State of AI", "pageRefs": null, "priority": null, "whenToUse": null, "confidence": 0.9, "extraction": { "at": "2026-07-16 21:56:38.602577+00", "model": "or:meta-llama/llama-4-scout", "runId": null, "seconds": null, "promptVersion": null }, "pageNumber": 156, "whyItWorks": null, "antipattern": null, "cardinality": null, "narrativePurpose": null } ], "frameworks": [], "arcBeats": [ { "to": 156, "arc": { "name": "The Triple Take", "slug": "triple-take", "status": "active", "bestFor": "Data-driven presentations, research findings", "canonId": "019dd956-6221-73ad-bc12-378f8eb1f018", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "The Facts (What) -> The Implications (So What) -> The Action (Now What)", "description": null, "familyLabel": null }, "from": 151, "name": "The Action (Now What)", "beatId": "019dd95a-0682-776c-8e34-b4730e3b2f0d", "matchId": "019dd95a-0542-7728-a78e-e9950bdbe8e2", "beatType": { "name": "The Action (Now What)", "slug": "triple-take-the-action-now-what", "status": "active", "canonId": "019dd9b8-0a5e-735a-bf53-3a1ba629b7aa", "version": 1, "description": null }, "evidence": "Section 5 '8 predictions for the next 12 months' and conclusion.", "position": 3, "isPrimaryArc": true, "parentBeatType": { "name": "Resolution", "slug": "resolution", "status": "active", "canonId": "019dd9b8-045d-703b-8dd7-e92aba3ba91b", "version": 1, "description": "How the deck ends. Required in 19/20 arcs. The exception (Sparkline) ends with new bliss which is functionally a resolution." }, "alignedBlockIds": null, "matchConfidence": 78 }, { "to": 156, "arc": { "name": "The Onion", "slug": "onion", "status": "active", "bestFor": "Root cause analysis, diagnostics", "canonId": "019dd956-627f-76b1-b286-0c575846d572", "version": 1, "bodyDocId": "019df22a-2356-73eb-8061-d8c81dc0aa49", "structure": "Surface Observation -> First Layer -> Deeper Layer -> Core Insight -> Implications", "description": null, "familyLabel": null }, "from": 151, "name": "Implications", "beatId": "019dd95a-0682-776c-8e34-cb622c4afe44", "matchId": "019dd95a-0542-7728-a78e-ee7d1be6c763", "beatType": { "name": "Implications", "slug": "onion-implications", "status": "active", "canonId": "019dd9b8-0dda-7730-893d-deaac7e9c295", "version": 1, "description": null }, "evidence": "Predictions for next 12 months.", "position": 5, "isPrimaryArc": false, "parentBeatType": { "name": "Reflection", "slug": "reflection", "status": "active", "canonId": "019dd9b8-0510-73fd-8d00-1c7e45f7bc98", "version": 1, "description": "The \"so what\" beat. Used by analytical arcs (Triple-take, Onion) and learning arcs (Voyage-return, Tragedy)." }, "alignedBlockIds": null, "matchConfidence": 55 } ], "loops": [], "locked": true } ], "arcBeats": [ { "from": 4, "to": 69, "label": "The Facts (What)", "description": "Definitions then research breakthroughs (transfer learning, hardware, RL) and talent supply data." }, { "from": 70, "to": 150, "label": "The Implications (So What)", "description": "Industry applications and political/labor consequences across verticals and geographies." }, { "from": 151, "to": 156, "label": "The Action (Now What)", "description": "Section 5 '8 predictions for the next 12 months' and conclusion." } ], "loops": [ { "from": 7, "to": 9, "label": "Aha Moment", "description": "Show transfer learning enables superhuman skin-cancer detection" }, { "from": 10, "to": 21, "label": "Pattern Hunter", "description": "Stack evidence that AI hardware is the binding constraint" }, { "from": 22, "to": 29, "label": "Pattern Hunter", "description": "Demonstrate hardware vendors are scaling fast" }, { "from": 30, "to": 35, "label": "Zoom In", "description": "Zoom from object detection to full scene understanding" }, { "from": 36, "to": 39, "label": "Pattern Hunter", "description": "Show RL agents now beat human experts across domains" }, { "from": 40, "to": 43, "label": "Zoom In", "description": "Diagnose ML bias by isolating its sources" }, { "from": 44, "to": 50, "label": "Iceberg", "description": "Surface hidden risks beneath model accuracy" }, { "from": 51, "to": 55, "label": "Pattern Hunter", "description": "Catalogue frontier research trends (AutoML, federated)" }, { "from": 57, "to": 64, "label": "Pattern Hunter", "description": "Prove AI talent is scarce and concentrated at Google" }, { "from": 65, "to": 67, "label": "Cost Of Inaction", "description": "Quantify escalating talent compensation arms race" }, { "from": 71, "to": 73, "label": "Pattern Hunter", "description": "Show GAFAMBAT capture AI infrastructure layer" }, { "from": 74, "to": 80, "label": "Day In Life", "description": "Show pharma/healthcare struggle and AI intervention" }, { "from": 81, "to": 85, "label": "Zoom In", "description": "Zoom into surveillance/privacy as the dark side of AI" }, { "from": 86, "to": 108, "label": "Pattern Hunter", "description": "Catalogue vertical AI deployment evidence" }, { "from": 110, "to": 122, "label": "Pattern Hunter", "description": "Aggregate survey data on public attitudes to AI/automation" }, { "from": 123, "to": 129, "label": "Pattern Hunter", "description": "Stack indicators of structural shift in US labor market" }, { "from": 130, "to": 138, "label": "Cost Of Inaction", "description": "Frame inaction risk and policy responses to automation" }, { "from": 139, "to": 143, "label": "Why Now", "description": "Argue national AI strategies create urgent geopolitical window" }, { "from": 144, "to": 149, "label": "Tale Two Worlds", "description": "Compare US vs China AI capability head to head" }, { "from": 152, "to": 152, "label": "Scenario Fork", "description": "Project eight forward scenarios for the year ahead" } ] }