{
  "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": 234,
  "pageCount": 313,
  "prevPage": 233,
  "nextPage": 235,
  "slideType": "industry_trends",
  "function": "analyze_data",
  "density": "dense",
  "nDataPoints": 6,
  "notes": "The chart shows a scatter plot of AI models comparing speed vs cost improvement relative to unassisted human performance.",
  "elementsJson": [
    "headline_text",
    "bullet_list",
    "scatter_plot",
    "footnote"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-5e88-73ef-bd5d-0d0f98caffe1/234",
  "deckHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1",
  "deckJsonHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1.json",
  "deckAnchorHref": "/decks/019dd923-5e88-73ef-bd5d-0d0f98caffe1#slide-234",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "Across 44 professions and 1,320 tasks, models are approaching human experts in a significant subset of domains.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd952-47c2-727e-a409-7d38968477d0",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.55,
        "w": 0.3,
        "x": 0.65,
        "y": 0.3
      },
      "kind": "chart",
      "text": "GDPval: Speed and Cost Improvements from AI Assistance to Human Experts",
      "attrs": null,
      "subkind": "scatter",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "1519ffca-529d-4d8c-bae3-8efdf3625729",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.55,
        "w": 0.6,
        "x": 0.03,
        "y": 0.31
      },
      "kind": "list",
      "text": "Reasoning models outperformed GPT-4o on task win-rate by an average margin of 20.7% across 44 categories of professional work.\nClaude, which has not historically dominated other benchmarks, achieved the highest* win rates in 32 of 44 professions. The paper attributes part of this success to Claude's strengths in formatting.\nGeneral-purpose models are already demonstrating strong competence as professional assistants. Meanwhile, frontier labs and recent entrants General Reasoning and Mechanize are also rapidly building RL environments on real-world work scenarios.\nWith this hill to climb and an influx of corporate data and demos, knowledge workers may soon experience the workplace transformations that AI leaders have long predicted. The Lufthansa Group even said it expects to cut 4k administrative jobs by 2030.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "bce18df1-0902-4193-9db2-504e41f1279a",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "Task win-rate: 20.7%",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd952-47c2-727e-a409-830470fec281",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.6,
        "x": 0.03,
        "y": 0.21
      },
      "kind": "paragraph",
      "text": "OpenAI's new benchmark for economically valuable tasks, GDPval, demonstrates the steady march of AI progress. Across 44 professions and 1,320 tasks, models are approaching human experts in a significant subset of domains.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "a278c549-1ba9-43da-b1db-5a85b9358ae2",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.04,
        "w": 0.45,
        "x": 0.03,
        "y": 0.14
      },
      "kind": "title",
      "text": "GDPval: a warning shot for the job market",
      "attrs": null,
      "subkind": "headline",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "8b6af6fd-f4fe-4e79-a350-154442c160ad",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [
    {
      "name": "Action Titles",
      "slug": "action-titles",
      "agent": "Architect",
      "layer": "slide",
      "matchId": "019dd95a-1055-74e0-a95c-b0aec740e686",
      "evidence": "'GDPval: a warning shot for the job market' — title states implication.",
      "confidence": 85
    },
    {
      "name": "Data Story Arc",
      "slug": "data-story-arc",
      "agent": "Storyteller",
      "layer": "slide",
      "matchId": "e73c600a-0d8d-4e3a-9d9f-e75b67c51716",
      "evidence": "OpenAI's new benchmark for economically valuable tasks, GDPval, demonstrates the steady march of AI progress.",
      "confidence": 0.7
    }
  ],
  "frameworks": [],
  "arcBeats": [
    {
      "to": 282,
      "from": 190,
      "beatId": "019dd95a-0682-776c-8e35-a256fe16ea9c",
      "arcName": "The Triple Take",
      "arcSlug": "triple-take",
      "beatName": "The Implications (So What)",
      "beatSlug": "triple-take-the-implications-so-what",
      "evidence": "Sections 3-4 frame political and safety consequences of the technical/industry trends.",
      "position": 2,
      "confidence": 70,
      "parentBeatName": "Reflection",
      "parentBeatSlug": "reflection"
    },
    {
      "to": 282,
      "from": 190,
      "beatId": "019dd95a-0682-776c-8e35-b713af60bf88",
      "arcName": "The Onion",
      "arcSlug": "onion",
      "beatName": "Core Insight",
      "beatSlug": "onion-core-insight",
      "evidence": "Politics and Safety reveal the structural risks at the core.",
      "position": 4,
      "confidence": 45,
      "parentBeatName": "Turn",
      "parentBeatSlug": "turn"
    }
  ],
  "loops": [],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}