{
  "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": 104,
  "pageCount": 114,
  "prevPage": 103,
  "nextPage": 105,
  "slideType": "diagnosis",
  "function": "diagnose",
  "density": "balanced",
  "nDataPoints": 0,
  "notes": "The slide uses the CoinRun experiment to illustrate a specific AI safety failure mode.",
  "elementsJson": [
    "headline_text",
    "paragraph",
    "bullet_list",
    "screenshot"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc/104",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc#slide-104",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "One concern of using RL agents is that they may learn strong skills while having failed to learn the right goals, and for this failure to only exhibit at test-time under distribution shifts.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a21f-0a7096d49c8d",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.4,
        "w": 0.52,
        "x": 0.45,
        "y": 0.4
      },
      "kind": "image",
      "text": "Vanilla CoinRun (Train) and Randomized coin position (Test) screenshots",
      "attrs": null,
      "subkind": "screenshot",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "789f3355-fe30-4e69-a038-017a8d14dbed",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.42,
        "x": 0.02,
        "y": 0.38
      },
      "kind": "list",
      "text": "Agents were trained on the CoinRun video game task, in which a reward is obtained and the level completes when reaching a coin at the end of a stage.\nAt test-time, the coin is randomly placed within the stage instead. Agents maintained their capabilities to navigate and traverse obstacles, but ignore the coin and instead run to the end of the level, demonstrating a failure to learn the correct goal.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "b466667c-e73a-45b0-b29c-7c1fb863b7bd",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.12,
        "w": 0.96,
        "x": 0.02,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "One concern of using RL agents is that they may learn strong skills while having failed to learn the right goals, and for this failure to only exhibit at test-time under distribution shifts. This issue was empirically demonstrated for the first time in a paper presented at ICML this year.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "05ebdbbc-a434-41ee-a2eb-b3da026de83b",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.8,
        "x": 0.02,
        "y": 0.14
      },
      "kind": "title",
      "text": "Goal misgeneralization – agents can learn the right skills but the wrong objective",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "b2d0772d-8848-47d2-bbbc-66542cbec96d",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Problem Statement Canvas",
      "slug": "problem-statement-canvas",
      "agent": "Architect",
      "layer": "slide",
      "matchId": "e8081971-d9d0-4afd-a67b-d5e8809f5644",
      "evidence": "The slide presents a problem statement for goal misgeneralization in RL agents, a key concern for AI safety.",
      "confidence": 0.6
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 106,
      "from": 81,
      "beatId": "019dd95a-0682-776c-8e35-28fc5bded874",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Implications (So What)",
      "beatSlug": "triple-take-the-implications-so-what",
      "evidence": "Politics and Safety sections interpret what the facts mean for governance and risk.",
      "position": 2,
      "confidence": 78,
      "parentBeatName": "Reflection",
      "parentBeatSlug": "reflection"
    },
    {
      "to": 106,
      "from": 94,
      "beatId": "019dd95a-0682-776c-8e35-381428a808af",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Climax",
      "beatSlug": "mountain-climax",
      "evidence": "Safety section confronts existential risk and alignment urgency.",
      "position": 3,
      "confidence": 55,
      "parentBeatName": "Turn",
      "parentBeatSlug": "turn"
    }
  ],
  "loops": [
    {
      "to": 106,
      "from": 100,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-db1f31ffae02",
      "evidence": "RLHF, red-teaming, mechanistic interpretability, goal misgeneralization, moral behavior, Conjecture lab.",
      "position": 17,
      "objective": "Catalogue alignment-research techniques and example labs",
      "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
}