{
  "docId": "019dd923-5e88-73ef-bd5d-06b04d219fea",
  "docSlug": "dd91c78f6570bf29",
  "documentTitle": "2023 Air Street Capital The State of AI Report 2023",
  "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": 57,
  "pageCount": 163,
  "prevPage": 56,
  "nextPage": 58,
  "slideType": "industry_trends",
  "function": "summarize",
  "density": "overcrowded",
  "nDataPoints": 10,
  "notes": "The slide compares traditional numerical weather prediction methods with new deep learning/physics-informed models.",
  "elementsJson": [
    "headline_text",
    "paragraph",
    "bullet_list",
    "bar_chart_horizontal",
    "map",
    "line_chart",
    "logo_grid"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-06b04d219fea/57",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea#slide-57",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "For both problems, learned methods and physics-informed models that incorporate relevant priors are able to deliver performance improvements preferred by professional meteorologists.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a225-88c18089694d",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.2,
        "w": 0.3,
        "x": 0.15,
        "y": 0.75
      },
      "kind": "chart",
      "text": "NowcastNet performance comparison charts",
      "attrs": null,
      "subkind": "bar-horizontal",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "81c91e00-35d0-4a49-9a62-c79adac7f3a3",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.2,
        "w": 0.4,
        "x": 0.55,
        "y": 0.75
      },
      "kind": "chart",
      "text": "Pangu-Weather cyclone tracking and error charts",
      "attrs": null,
      "subkind": "line",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "ff84aa7e-ee31-497a-abd0-0635cf135861",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.46,
        "x": 0.52,
        "y": 0.45
      },
      "kind": "list",
      "text": "Pangu-Weather is a 3D deep learning model with Earth-specific priors trained on 39 years of global data that can generate medium-range global weather for. The system can be used for more accurate early-stage cyclone tracking vs status quo.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "4947f336-1f0b-4c0f-a665-dcb82b6992ec",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.46,
        "x": 0.02,
        "y": 0.45
      },
      "kind": "list",
      "text": "NowcastNet is a nonlinear model that uses physical first principles and statistical-learning methods, unified under a deep generative model framework. Evaluated by 62 professional meteorologists from across China, the model ranks 1st in 71% of cases against leading methods.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "805796fa-9e92-4072-b2ff-30305fd8d170",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Model preference rate: 71%",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a225-8fe4142fac18",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.18,
        "w": 0.96,
        "x": 0.02,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "Skilful short term precipitation predictions (nowcasting) today are blurry, prone to dissipation and are slow. Medium-range global weather forecasts using the accurate numerical weather prediction method is computationally expensive. For both problems, learned methods and physics-informed models that incorporate relevant priors are able to deliver performance improvements preferred by professional meteorologists. New benchmark datasets such as Google's WeatherBench 2 help data-driven weather model development.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "8cb6e174-ef0f-4750-a583-69bb7639780d",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.7,
        "x": 0.02,
        "y": 0.14
      },
      "kind": "title",
      "text": "More accurate weather predictions, in the now(casts) and the longer ranges",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "997a87fa-b324-4752-a776-874261d91fe1",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Chart Selection Guide",
      "slug": "chart-selection-guide",
      "agent": "Designer",
      "layer": "slide",
      "matchId": "f6d55b9c-96b4-4393-9099-840707c63772",
      "evidence": "chart/line: Pangu-Weather cyclone tracking and error charts",
      "confidence": 0.7
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 120,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-41afd44ef59f",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Facts (What)",
      "beatSlug": "triple-take-the-facts-what",
      "evidence": "Research + Industry sections inventory model, compute, funding facts.",
      "position": 1,
      "confidence": 78,
      "parentBeatName": "Setup",
      "parentBeatSlug": "setup"
    },
    {
      "to": 120,
      "from": 11,
      "beatId": "019dd95a-0682-776c-8e35-523bfb7f96e6",
      "arcName": "The Mountain",
      "arcSlug": "mountain",
      "beatName": "Rising Action",
      "beatSlug": "mountain-rising-action",
      "evidence": "Escalating capabilities, compute concentration and capital flows.",
      "position": 2,
      "confidence": 45,
      "parentBeatName": "Development",
      "parentBeatSlug": "development"
    }
  ],
  "loops": [
    {
      "to": 58,
      "from": 50,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-f70f6435b722",
      "evidence": "Text-to-video, instruction editing, Gaussian splatting, NeRFs, depth, SAM, DINOv2, weather, music.",
      "position": 6,
      "objective": "Generative media and visual foundation models proliferate",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 72,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}