{
  "docId": "019dd923-5e88-73ef-bd5d-01cf0d8a8fbc",
  "docSlug": "cc8183ec02431b7a",
  "documentTitle": "2022 Air Street Capital The State of AI Report 2022",
  "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": 28,
  "pageCount": 114,
  "prevPage": 27,
  "nextPage": 29,
  "slideType": "industry_trends",
  "function": "illustrate_case",
  "density": "overcrowded",
  "nDataPoints": 18,
  "notes": "The chart compares truthfulness and informativeness across different model sizes and prompting strategies, highlighting WebGPT's performance.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "bar_chart_grouped"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc/28",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc#slide-28",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Importantly, the use of increasing amounts of human demonstration data significantly increased the truthfulness and informativeness of answers.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a21a-077a1cb80e24",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.4,
        "w": 0.3,
        "x": 0.68,
        "y": 0.4
      },
      "kind": "chart",
      "text": "Performance comparison of GPT-3 and WebGPT models on truthfulness and informativeness metrics.",
      "attrs": null,
      "subkind": "bar-grouped",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "ef973f97-88f7-40f3-a12f-6cf2c5e3a6a0",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.03,
        "w": 0.3,
        "x": 0.68,
        "y": 0.39
      },
      "kind": "legend",
      "text": "Truthful (%) vs Truthful and informative (%)",
      "attrs": null,
      "subkind": null,
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "0ec0fade-4ea7-44dc-9a67-8505932d9f91",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.55,
        "w": 0.65,
        "x": 0.03,
        "y": 0.31
      },
      "kind": "list",
      "text": "OpenAI's WebGPT was the first model to demonstrate this convincingly by fine-tuning GPT-3 to interact with a search engine to provide answers grounded with references. This merely required collecting data of humans doing this task and converting the interaction data into text that the model could consume for training by standard supervised learning. Importantly, the use of increasing amounts of human demonstration data significantly increased the truthfulness and informativeness of answers (right panel, white bars for WebGPT), a significant advance from when we covered truthfulness evaluation in our 2021 report (slide 44).\nAdept, a new AGI company, is commercializing this paradigm. The company trains large transformer models to interact with websites, software applications and APIs (see more at adept.ai/act) in order to drive workflow productivity.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "1f9e1be4-6434-4148-843b-f8eac8ac94a7",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Truthfulness and informativeness",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a21a-09241c44ac2c",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.65,
        "x": 0.03,
        "y": 0.21
      },
      "kind": "paragraph",
      "text": "Language models can learn to use tools such as search engines and calculators, simply by making available text interfaces to these tools and training on a very small number of human demonstrations.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "05a669ec-05fa-4514-8e44-8998d6ed1a43",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.04,
        "w": 0.5,
        "x": 0.03,
        "y": 0.14
      },
      "kind": "title",
      "text": "Teach a machine to fish: tool use as the next frontier?",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "173cb38e-e501-430f-a8c4-bea5f4a47979",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 80,
      "from": 5,
      "beatId": "019dd95a-0682-776c-8e35-2557e3799e96",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Definitions, exec summary, Research and Industry sections inventory the state of AI.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 93,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-35abf646aa8b",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Research, Industry, Politics sections accumulate signals.",
      "position": 2,
      "confidence": 55,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 29,
      "from": 22,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-a20c665fe3de",
      "evidence": "Codex, Transformer alternatives, Minerva, BIG-Bench, Chinchilla, emergence, tool-use, compute eras stack as evidence.",
      "position": 3,
      "objective": "Survey LLM research frontier — transformers, scaling, emergence, compute eras",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 78,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}