{
  "docId": "019dd923-5e88-73ef-bd5d-0d0f98caffe1",
  "docSlug": "5df6bd1b0447b5f6",
  "documentTitle": "2025 Air Street Capital The State of AI Report 2025",
  "authorId": "AirStreetCapital",
  "authorName": "Air Street Capital",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "vc_research",
  "sourceTypeLabel": "VC research",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.777,
  "pageNumber": 105,
  "pageCount": 313,
  "prevPage": 104,
  "nextPage": 106,
  "slideType": "industry_trends",
  "function": "summarize",
  "density": "dense",
  "nDataPoints": 1,
  "notes": "The slide discusses the trade-off between technical capability and user experience during a product rollout.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "scatter_plot"
  ],
  "metadataConfidence": 0.95,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-0d0f98caffe1/105",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1#slide-105",
  "components": [
    {
      "bbox": {
        "h": 0.12,
        "w": 0.95,
        "x": 0.02,
        "y": 0.21
      },
      "kind": "callout",
      "text": "GPT-5 leads the intelligence-per-dollar frontier with best-in-class benchmarks at 12x lower cost than Claude, but user backlash over the sudden removal of GPT-4o and o3 and concerns/teething problems about opaque router decisions that users aren’t used to overshadowed its technical achievements.",
      "attrs": null,
      "subkind": "primary",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "9d357f9f-0131-4a4b-8ad6-1ce0a7399fa6",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "callout",
      "text": "GPT-5 leads the intelligence-per-dollar frontier with best-in-class benchmarks at 12x lower cost than Claude, but user backlash over the sudden removal of GPT-4o and o3 and concerns/teething problems about opaque router decisions that users aren't used to overshadowed its technical achievements.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-41ac-7180-a824-be6474eca612",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.5,
        "w": 0.3,
        "x": 0.65,
        "y": 0.35
      },
      "kind": "chart",
      "text": "Plot of model pricing vs LMSys Elo (Aug 2025)",
      "attrs": null,
      "subkind": "scatter",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "cab3ae9b-869f-4584-815f-1cf63aa8c614",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.6,
        "w": 0.6,
        "x": 0.02,
        "y": 0.35
      },
      "kind": "list",
      "text": "GPT-5 is ostensibly a brilliant model: it swept the leaderboard on LMArena and has a 400K context window in the API. OpenAI now dominates the intelligence per dollar frontier for the first time.\nThe rollout wasn’t ideal: Altman held an emergency Reddit AMA to address the abrupt removal of previous models and viral ‘chart crimes’.\nThe removal of GPT-4o and o3’s familiar personality upset users, which was ironic given the same launch introduced custom personas.\nWe’ve previously hypothesised that model companies would dynamically route queries to right-sized models for latency and cost reasons. GPT-5 is the first major chat system to introduce this using a router as the consumer endpoint. Users can opt for faster responses by hitting “skip” if the model invokes its more capable version. In practice, people will take time to acclimate to this UX: the perceived opacity of model selection has led to a flurry of complaints.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "f80837bd-c071-42e7-83a3-9c252228d24f",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Intelligence-per-dollar: 12x",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-41ac-7180-a824-c17d00c596a7",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.6,
        "x": 0.02,
        "y": 0.13
      },
      "kind": "title",
      "text": "The duality of GPT-5: today’s best model was clouded by the worst launch",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "3f6d70b4-b6eb-49c4-92b5-2a004671e93b",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Contrast Pairs",
      "slug": "contrast-pairs",
      "agent": "Storyteller",
      "layer": "loop",
      "matchId": "019dd95a-1055-74e0-a95c-546b630e8f27",
      "evidence": "Best model vs. worst launch as deliberate contrast.",
      "confidence": 80
    },
    {
      "name": "Action Titles",
      "slug": "action-titles",
      "agent": "Architect",
      "layer": "slide",
      "matchId": "019dd95a-1055-74e0-a95c-530c4434f00d",
      "evidence": "'The duality of GPT-5: today's best model was clouded by the worst launch'.",
      "confidence": 85
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 189,
      "from": 12,
      "beatId": "019dd95a-0682-776c-8e35-9f380673831f",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Sections 1-2 lay out research findings and industry data with charts and case studies.",
      "position": 1,
      "confidence": 70,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 189,
      "from": 90,
      "beatId": "019dd95a-0682-776c-8e35-b06421e14afb",
      "arcName": "The Onion",
      "arcSlug": "onion",
      "beatName": "Deeper Layer",
      "beatSlug": "onion-deeper-layer",
      "evidence": "Industry section exposes economics, infra, geopolitics beneath research.",
      "position": 3,
      "confidence": 45,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 112,
      "from": 105,
      "name": "Funnel Analysis",
      "slug": "32-funnel-analysis",
      "bestFor": "Sales optimization, customer journey, process efficiency",
      "matchId": "019dd95a-07fe-70ce-8d3e-a3f81a44ee54",
      "evidence": "GPT-5 economics → vibe coding costs → Cursor unit economics → 'When do we see profitable models?' (p.112).",
      "position": 6,
      "objective": "Trace AI economics from revenue through margins to profitability",
      "structure": "Top of Funnel -> Stage 1 Conversion -> Stage 2 Conversion -> Bottom of Funnel -> The Leak",
      "confidence": 65,
      "description": "Show progressive filtering through stages to identify where drop-off occurs"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}