{
  "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": 46,
  "pageCount": 114,
  "prevPage": 45,
  "nextPage": 47,
  "slideType": "case_study",
  "function": "illustrate_case",
  "density": "dense",
  "nDataPoints": 2,
  "notes": "The slide highlights the dual-use nature of AI in pharmaceutical research, specifically referencing the MegaSyn model.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "scatter_plot",
    "histogram",
    "other",
    "logo_grid"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc/46",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-01cf0d8a8fbc#slide-46",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Researchers from Collaborations Pharmaceuticals and King’s College London showed that machine learning models designed for therapeutic use can be easily repurposed to generate biochemical weapons.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a21a-5c6f3c2de144",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.25,
        "x": 0.72,
        "y": 0.65
      },
      "kind": "chart",
      "text": "Distribution of predicted LD50 values",
      "attrs": null,
      "subkind": "histogram",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "cb90972b-5af1-4911-9405-70d33f2a5910",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.25,
        "x": 0.72,
        "y": 0.35
      },
      "kind": "chart",
      "text": "t-SNE plot showing generated compounds vs LD50 dataset",
      "attrs": null,
      "subkind": "scatter",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "232f942a-5b47-4de4-a5d8-6ba06706c471",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.15,
        "w": 0.15,
        "x": 0.82,
        "y": 0.7
      },
      "kind": "image",
      "text": "Chemical structure of VX",
      "attrs": null,
      "subkind": "illustration",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "59f61d66-3e4a-4a20-9753-267b1b60a228",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.65,
        "x": 0.05,
        "y": 0.32
      },
      "kind": "list",
      "text": "The researchers had trained their “MegaSyn” model to maximize bioactivity and minimize toxicity. To design toxic molecules, they kept the same model, but now simply training it to maximize both bioactivity and toxicity. They used a public database of drug-like molecules.\nThey directed the model towards generation of the nerve agent VX, known to be one of the most toxic chemical warfare agents.\nHowever, as is the case with regular drug discovery, finding molecules with a high predicted toxicity doesn’t mean it is easy to make them. But as drug discovery with AI in the loop is being dramatically improved, we can imagine best practices in drug discovery diffusing into building cheap biochemical weapons.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "4300f255-241d-4a63-9343-b1316ee9a9b2",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.65,
        "x": 0.05,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "Researchers from Collaborations Pharmaceuticals and King’s College London showed that machine learning models designed for therapeutic use can be easily repurposed to generate biochemical weapons.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "09a4e224-7b00-476d-8dc8-2791d6891f37",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.8,
        "x": 0.05,
        "y": 0.14
      },
      "kind": "title",
      "text": "Drug discovery, the flagship “AI for good” application, is not immune to misuse",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "63b7bc8d-78fa-490b-841c-9117402b89db",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Negativity Bias",
      "slug": "negativity-bias",
      "agent": "Storyteller",
      "layer": "slide",
      "matchId": "019dd95a-1055-74e0-a956-215545f072e8",
      "evidence": "Title flags 'AI for good ... not immune to misuse' as warning.",
      "confidence": 78
    }
  ],
  "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": 46,
      "from": 44,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-b2931f86a816",
      "evidence": "Antibiotic selection, mass-spec transformers, drug-discovery misuse all in biology.",
      "position": 7,
      "objective": "Cluster ML-for-biotech case studies with dual-use caveat",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 70,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}