{
  "docId": "019dd923-5e88-73ef-bd5d-893216e9a633",
  "docSlug": "3fc7a1c5528471f7",
  "documentTitle": "2023 Benedict Evans 2023 AI and Everything Else",
  "authorId": "BenedictEvans",
  "authorName": "Benedict Evans",
  "documentKindSlug": "consulting-deck",
  "documentKindLabel": "Consulting deck",
  "sourceTypeSlug": "independent_analyst",
  "sourceTypeLabel": "Independent analyst",
  "presentationDate": null,
  "orientation": "landscape",
  "aspectRatio": 1.778,
  "pageNumber": 47,
  "pageCount": 92,
  "prevPage": 46,
  "nextPage": 48,
  "slideType": "industry_trends",
  "function": "analyze_data",
  "density": "balanced",
  "nDataPoints": 40,
  "notes": "The chart uses error bars to show the range of exposure within each industry sector.",
  "elementsJson": [
    "headline_text",
    "paragraph",
    "scatter_plot"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-893216e9a633/47",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-893216e9a633",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-893216e9a633.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-893216e9a633#slide-47",
  "components": [
    {
      "bbox": {
        "h": 0.65,
        "w": 0.45,
        "x": 0.52,
        "y": 0.21
      },
      "kind": "chart",
      "text": "Generative AI Exposure of Compustat Firms by NAICS 3-digit Subsector",
      "attrs": null,
      "subkind": "bar-vertical",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "37a82e7d-9304-4797-b389-6e0633aa7cdf",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Generative AI Exposure",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c8-76a3-bc5c-4f8d01ca6a85",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.2,
        "w": 0.4,
        "x": 0.06,
        "y": 0.45
      },
      "kind": "paragraph",
      "text": "Imagine doing this for 'the internet' in 1995\nOr 'mobile internet' in 2005\nWhat would you have got right, and wrong?",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "4373021b-ed48-4700-9dc7-9fc29b5ef8f7",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.03,
        "w": 0.4,
        "x": 0.06,
        "y": 0.91
      },
      "kind": "source-note",
      "text": "Source: The Labor Impact of Generative AI on Firm Values, Eisfeldt et al",
      "attrs": null,
      "subkind": null,
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "3e024411-24d6-453f-9ff3-efca140e70b8",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.1,
        "w": 0.4,
        "x": 0.06,
        "y": 0.25
      },
      "kind": "title",
      "text": "Can we predict the effects top-down?",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "ceabda7d-24fa-461d-aff4-3b9c0a58eb69",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Action Titles",
      "slug": "action-titles",
      "agent": "Architect",
      "layer": "slide",
      "matchId": "019dd95a-1b83-7130-933b-7b7e5677a315",
      "evidence": "Title is the question being investigated",
      "confidence": 70
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 52,
      "from": 36,
      "beatId": "019dd95a-07a7-7579-bd64-f4b4d0ccd18f",
      "arcName": "The Onion",
      "arcSlug": "onion",
      "beatName": "Deeper Layer",
      "beatSlug": "onion-deeper-layer",
      "evidence": "'How is this useful?' divider; use-case discovery",
      "position": 3,
      "confidence": 70,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    },
    {
      "to": 71,
      "from": 36,
      "beatId": "019dd95a-07a7-7579-bd65-058a01252d39",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Implications (So What)",
      "beatSlug": "triple-take-the-implications-so-what",
      "evidence": "Use-case discovery and deployment patterns",
      "position": 2,
      "confidence": 55,
      "parentBeatName": "Reflection",
      "parentBeatSlug": "reflection"
    }
  ],
  "loops": [
    {
      "to": 51,
      "from": 37,
      "name": "Tale Two Worlds",
      "slug": "04-tale-two-worlds",
      "bestFor": "Competitive analysis, benchmarking, case for change",
      "matchId": "019dd95a-08f8-7619-ac2e-5f1b452b0d9a",
      "evidence": "Repeats 'how is ML useful?' (2013) vs 'how is LLM useful?' (2023); occupation-by-occupation usage chart.",
      "position": 5,
      "objective": "Compare 2013 ML vs 2023 LLM use-case discovery, and 'I have'/'I don't have' that use case",
      "structure": "Current State -> Desired State / Benchmark -> The Gap & Implication",
      "confidence": 75,
      "description": "Show the gap between two states to drive urgency or highlight opportunity"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}