{
  "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": 12,
  "pageCount": 163,
  "prevPage": 11,
  "nextPage": 13,
  "slideType": "case_study",
  "function": "illustrate_case",
  "density": "dense",
  "nDataPoints": 20,
  "notes": "The chart shows a comparison of GPT-4, GPT-4 (no vision), and GPT-3.5 across various academic and professional exams.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "paragraph",
    "bar_chart_grouped"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-06b04d219fea/12",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-06b04d219fea#slide-12",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "GPT-4 is, as of the release of this report, the uncontested most generally capable AI model.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a220-a9e17267bc90",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.5,
        "w": 0.3,
        "x": 0.68,
        "y": 0.42
      },
      "kind": "chart",
      "text": "Exam results (ordered by GPT-3.5 performance)",
      "attrs": null,
      "subkind": "bar-grouped",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "e337c8db-64b1-42bb-88bc-29ce500f3fed",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.4,
        "w": 0.65,
        "x": 0.02,
        "y": 0.48
      },
      "kind": "list",
      "text": "OpenAI did a comprehensive evaluation of GPT-4 not only on classical NLP benchmarks, but also on exams designed to evaluate humans (e.g. Bar exam, GRE, Leetcode).\nGPT-4 is the best model across the board. It solves some tasks that GPT-3.5 was unable to, like the Uniform Bar Exam where GPT-4 scores 90% compared to 10% for GPT-3.5. On most tasks, the added vision component had only a minor impact, but it helped tremendously on others.\nOpenAI reports that although GPT-4 still suffers from hallucinations, it is factually correct 40% more often than the previous-best ChatGPT model on an adversarial truthfulness dataset (generated to fool AI models).",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "ab9b1c1f-4f83-4f59-8578-2303409b7ba7",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Estimated percentile: 90%",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c4-7719-a220-ad17345042ff",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.25,
        "w": 0.65,
        "x": 0.02,
        "y": 0.22
      },
      "kind": "paragraph",
      "text": "GPT-4 is OpenAI's latest Large Language Model. In contrast with text-only GPT-3 and follow-ups, GPT-4 is multimodal: it was trained on both text and images; it can among other capabilities generate text based on images. At 8,192 tokens when it was released, it had already exceeded the previous-best GPT-3.5 in possible input size. It is, of course, trained using RLHF. Equipped with these advances, GPT-4 is, as of the release of this report, the uncontested most generally capable AI model.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "c13a2060-7d09-4ddb-93bf-0ac67a092df3",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.05,
        "w": 0.6,
        "x": 0.02,
        "y": 0.14
      },
      "kind": "title",
      "text": "GPT-4 is out and it crushes every other LLM, and many humans",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "64d5646f-ba90-47f0-957d-8f4e73fb4330",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Inductive Reasoning",
      "slug": "inductive-reasoning",
      "agent": "Architect",
      "layer": "loop",
      "matchId": "019dd95a-1055-74e0-a957-06ef7ed8c2eb",
      "evidence": "Opens a multi-slide pattern_hunter loop on GPT-4/RLHF dominance.",
      "confidence": 70
    },
    {
      "name": "Action Titles",
      "slug": "action-titles",
      "agent": "Architect",
      "layer": "slide",
      "matchId": "019dd95a-1055-74e0-a956-ff04dd35c74a",
      "evidence": "Title 'GPT-4 is out and it crushes every other LLM, and many humans' states the verdict.",
      "confidence": 92
    },
    {
      "name": "Annotation",
      "slug": "annotation",
      "agent": "Designer",
      "layer": "slide",
      "matchId": "019dd95a-1055-74e0-a957-029d35355dd8",
      "evidence": "Callout quotes the headline finding to anchor the chart.",
      "confidence": 70
    }
  ],
  "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": 21,
      "from": 12,
      "name": "Pattern Hunter",
      "slug": "02-pattern-hunter",
      "bestFor": "Time-pressed audiences, building consensus, when data is strong",
      "matchId": "019dd95a-07fe-70ce-8d3d-e61d00ffc6e4",
      "evidence": "Sequential GPT-4, RLHF, LLaMa, popularity-mention slides converge on a single conclusion.",
      "position": 2,
      "objective": "Pile up evidence that GPT-4 and RLHF dominate the LLM landscape",
      "structure": "Evidence A -> Evidence B -> Evidence C -> Pattern/Conclusion",
      "confidence": 82,
      "description": "Group multiple pieces of evidence that together point to a pattern or conclusion"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}