{
  "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": 59,
  "pageCount": 313,
  "prevPage": 58,
  "nextPage": 60,
  "slideType": "case_study",
  "function": "illustrate_case",
  "density": "dense",
  "nDataPoints": 4,
  "notes": "The slide highlights the shift from screening existing materials to generative design using diffusion models.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "process_diagram",
    "line_chart"
  ],
  "metadataConfidence": 0.95,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-0d0f98caffe1/59",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1#slide-59",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "MatterGen generates materials that are 2x more likely to be stable and novel, and 10x closer to equilibrium energy minima vs. SOTA.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-41ac-7180-a820-4407e27d0170",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.2,
        "w": 0.25,
        "x": 0.72,
        "y": 0.7
      },
      "kind": "chart",
      "text": "Chart comparing MatterGen performance vs screening in finding stable structures.",
      "attrs": null,
      "subkind": "line",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "93b1de1b-1de3-49b4-963f-a546e88d8392",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.3,
        "w": 0.2,
        "x": 0.74,
        "y": 0.35
      },
      "kind": "chart",
      "text": "Diagram showing the diffusion process for crystal generation (chemistry, symmetry, property control).",
      "attrs": null,
      "subkind": "process",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "6b043dea-27d3-410c-a0c3-fb3a21e0f436",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.5,
        "w": 0.65,
        "x": 0.02,
        "y": 0.35
      },
      "kind": "list",
      "text": "MatterGen is a diffusion model that learns to refine lattices, element types, and atomic positions into stable crystal structures by independently randomizes atom types, coordinates, and lattices, followed by iteratively denoising them back toward physically plausible structures.\nIt’s trained on ~600k stable crystal structures of compounds (Materials Project and Alexandria) with adapter modules for chemistry, symmetry, and property control that are fine-tuned into the model.\nMatterGen generates materials that are 2x more likely to be stable and novel, and 10x closer to equilibrium energy minima vs. SOTA.\nIt achieved multi-property inverse design (e.g. combining band gap and magnetism) and saw its first lab synthesis succeed, with measured values within ~20% of AI predictions.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "7b287ea6-94a9-4d5e-b8c5-5ddb58d2534e",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Number of SUN structures found: 106",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-41ac-7180-a820-49b052aca04a",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.1,
        "w": 0.65,
        "x": 0.02,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "Where UMA shows that scaling data and parameters yields universal predictive models, MatterGen takes the next leap: using diffusion to directly generate new inorganic crystals with targeted properties, rather than screening existing ones.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "f724a558-8899-4a7f-9491-2bd9e1a237d9",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.5,
        "x": 0.02,
        "y": 0.14
      },
      "kind": "title",
      "text": "From property predictions to material generation",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "3e98b4ef-4512-4ce5-976d-cb214bb6b533",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "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": 89,
      "from": 12,
      "beatId": "019dd95a-0682-776c-8e35-ac68557d958b",
      "arcName": "The Onion",
      "arcSlug": "onion",
      "beatName": "First Layer",
      "beatSlug": "onion-first-layer",
      "evidence": "Research section unpacks technical layer beneath the headlines.",
      "position": 2,
      "confidence": 45,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 67,
      "from": 55,
      "name": "Zoom In",
      "slug": "06-zoom-in",
      "bestFor": "Technical deep-dives, case studies, detailed analysis",
      "matchId": "019dd95a-07fe-70ce-8d3e-9d32ca243ef1",
      "evidence": "Sequence of case studies: AlphaEvolve, UMA, MatterGen, AMIE, AlphaFold reproductions all illustrate AI-as-scientist.",
      "position": 5,
      "objective": "Survey AI's rise as scientific collaborator across biology/chemistry/medicine",
      "structure": "The Big Picture -> Key Area of Focus -> Specific Detail -> Implication",
      "confidence": 75,
      "description": "Start broad, then progressively focus on specific details that prove your point"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}