{
  "docId": "019dd923-5e88-73ef-bd5d-0856e1444fb9",
  "docSlug": "8757f1b44ef7f176",
  "documentTitle": "2024 Air Street Capital The State of AI Report 2024",
  "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": 43,
  "pageCount": 213,
  "prevPage": 42,
  "nextPage": 44,
  "slideType": "industry_trends",
  "function": "summarize",
  "density": "dense",
  "nDataPoints": 0,
  "notes": "Part of the State of AI 2024 report.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "photo"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-0856e1444fb9/43",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-0856e1444fb9",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-0856e1444fb9.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-0856e1444fb9#slide-43",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Both Google DeepMind and OpenAI have given us sneak previews of highly powerful text-to-video diffusion models. But access remains heavily gated and neither has supplied much technical detail.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c3-7407-8d11-650823c55524",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.33,
        "w": 0.33,
        "x": 0.64,
        "y": 0.495
      },
      "kind": "image",
      "text": "Image of mammoths in snow generated by AI",
      "attrs": null,
      "subkind": "photo",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "194e5ed3-08c5-4e36-a089-dac50da4a168",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.6,
        "w": 0.6,
        "x": 0.02,
        "y": 0.31
      },
      "kind": "list",
      "text": "OpenAI's Sora is able to generate videos up to a minute long, while maintaining 3D consistency, object permanence, and high resolution. It uses spacetime patches, similar to the tokens used in transformer models, but for visual content, to learn efficiently from a vast dataset of videos.\nSora was also trained on visual data in its native size and aspect ratio, removing the usual cropping and resizing that reduces quality.\nGoogle DeepMind's Veo combines text and optional image prompts with a noisy compressed video input, processing them through encoders and a latent diffusion model to create a unique compressed video representation.\nThe system then decodes this representation into a final high-resolution video.\nAlso in the fight are Runway's Gen-3 Alpha, Luma's Dream Machine, and Kling by Kuaishou.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "d1fc0c90-7863-4754-b025-ba92a517b38f",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.95,
        "x": 0.02,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "Both Google DeepMind and OpenAI have given us sneak previews of highly powerful text-to-video diffusion models. But access remains heavily gated and neither has supplied much technical detail.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "115e1ec9-2dd2-4b5d-bd80-639bad956036",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.6,
        "x": 0.02,
        "y": 0.14
      },
      "kind": "title",
      "text": "...leading the big labs to release their own gated text-to-video efforts",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "fc193785-6b6e-4ae8-a636-9e230c86c6b7",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Action Titles",
      "slug": "action-titles",
      "agent": "Architect",
      "layer": "slide",
      "matchId": "019dd95a-1055-74e0-a958-db7f48221bee",
      "evidence": "Title '...leading the big labs to release their own gated text-to-video efforts'",
      "confidence": 80
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 153,
      "from": 9,
      "beatId": "019dd95a-0682-776c-8e35-5f2398b8d1d0",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Exec summary + Research + Industry sections inventory the year",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 85,
      "from": 12,
      "beatId": "019dd95a-0682-776c-8e35-6d9d8ee06d0a",
      "arcName": "Voyage and Return",
      "arcSlug": "voyage-return",
      "beatName": "The Unknown",
      "beatSlug": "voyage-return-the-unknown",
      "evidence": "Research section explores frontier model uncharted territory",
      "position": 2,
      "confidence": 55,
      "parentBeatName": "Complication",
      "parentBeatSlug": "complication"
    }
  ],
  "loops": [
    {
      "to": 44,
      "from": 40,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3e-3752140f3c2d",
      "evidence": "VLMs, diffusion image, Stable Video Diffusion, gated text-to-video, Movie Gen audio",
      "position": 6,
      "objective": "Map progress in multimodal vision/video/audio generation",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 76,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}