{
  "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": 82,
  "pageCount": 163,
  "prevPage": 81,
  "nextPage": 83,
  "slideType": "industry_trends",
  "function": "establish_context",
  "density": "dense",
  "nDataPoints": 4,
  "notes": "The chart illustrates the 'Current Export Restrictions' as a threshold boundary on a 2D plane of Computational Performance vs Interconnect Bandwidth.",
  "elementsJson": [
    "headline_text",
    "paragraph",
    "bullet_list",
    "scatter_plot"
  ],
  "metadataConfidence": 0.95,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-06b04d219fea/82",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea#slide-82",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "NVIDIA (and its competitors) developed chips which fly right below the export list thresholds.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a229-0d9d0cc00bda",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.45,
        "w": 0.3,
        "x": 0.65,
        "y": 0.4
      },
      "kind": "chart",
      "text": "Computational Performance vs Interconnect Bandwidth chart showing chip positioning relative to export restrictions.",
      "attrs": null,
      "subkind": "scatter",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "8b186e56-3f78-418a-bdc2-cd338d526604",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.4,
        "w": 0.6,
        "x": 0.05,
        "y": 0.4
      },
      "kind": "list",
      "text": "In late August 2022, NVIDIA's A100 and H100 – their most powerful chips for AI applications – were added to the US Commerce Department's export control list and became out of reach for Chinese companies. By November, NVIDIA had already started advertising the A800 and H800 chips, which it designed to be below the performance threshold set by the US ban.\nIntel did the same with a new version of their Habana Gaudi 2 chip, and AMD expressed a similar intent.\nAs a result, the likes of ByteDance and Baidu have ordered >$1B worth of A800/H800 NVIDIA GPUs. There has also been reports of increasing A100/H100 GPU traffic in China, but on a much smaller scale.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "85e438e2-4f79-4835-94e9-bf397ef7fe24",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Computational Performance: $1B",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a229-1272ef339a03",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.15,
        "w": 0.6,
        "x": 0.05,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "According to NVIDIA's CFO, China historically accounted for 20-25% of NVIDIA's revenue from data centre-related products (Financial Times). As a result, as the US commerce department became increasingly aggressive with export controls of AI chips, NVIDIA (and its competitors) developed chips which fly right below the export list thresholds.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "e50081e2-888d-4919-afe0-d47645c83f21",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.6,
        "x": 0.05,
        "y": 0.14
      },
      "kind": "title",
      "text": "NVIDIA, Intel and AMD make Chinese-export controls proof chips",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "a0db9ace-48a1-4366-90b6-6f81e668529f",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "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": 83,
      "from": 70,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-ff781be07723",
      "evidence": "Earnings, GPU clusters, A100/H100 indices, 19x research-paper share, 5-year LTV, hyperscaler capex all point one way.",
      "position": 8,
      "objective": "Establish NVIDIA's chokehold on AI compute via converging evidence",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 85,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}