mlsanityData to 7 Oct 2026

Which Economic Tasks are Performed with AI? Evidence from Millions of Claude Conversations

arXiv 2503.04761 ↗First posted 11 Feb 2025cs.CY cs.AI cs.CLcited by 88

Not on the map. Why?

Abstract (arXiv)

Despite widespread speculation about artificial intelligence's impact on the future of work, we lack systematic empirical evidence about how these systems are actually being used for different tasks. Here, we present a novel framework for measuring AI usage patterns across the economy. We leverage a recent privacy-preserving system to analyze over four million Claude.ai conversations through the lens of tasks and occupations in the U.S. Department of Labor's O*NET Database. Our analysis reveals that AI usage primarily concentrates in software development and writing tasks, which together account for nearly half of all total usage. However, usage of AI extends more broadly across the economy, with approximately 36% of occupations using AI for at least a quarter of their associated tasks. We also analyze how AI is being used for tasks, finding 57% of usage suggests augmentation of human capabilities (e.g., learning or iterating on an output) while 43% suggests automation (e.g., fulfilling a request with minimal human involvement). While our data and methods face important limitations and only paint a picture of AI usage on a single platform, they provide an automated, granular approach for tracking AI's evolving role in the economy and identifying leading indicators of future impact as these technologies continue to advance.

Key facts. Cited by 88 papers in the map (data through 7 Oct 2026). Builds on 3 of its 34 references that are in the map. First posted on arXiv 11 Feb 2025.

Builds on

3 of 34 references.

References · 3

  1. 2023The Impact of AI on Developer Productivity: Evidence from GitHub Copilotcited by 264
  2. 2023GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Modelscited by 174
  3. 2024Scaling Laws for Economic Productivity: Experimental Evidence in LLM-Assisted Translationcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 44; 2026: 44.

88 papers.

Sort:

By area:

  1. 2025LLMs Get Lost In Multi-Turn Conversationcited by 400
  2. 2025CWM: An Open-Weights LLM for Research on Code Generation with World Modelscited by 87
  3. 2025ML-Master: Towards AI-for-AI via Integration of Exploration and Reasoningcited by 57
  4. 2025The Widespread Adoption of Large Language Model-Assisted Writing Across Societycited by 50
  5. 2025REAL: Benchmarking Autonomous Agents on Deterministic Simulations of Real Websitescited by 43
  6. 2025The Homogenizing Effect of Large Language Models on Human Expression and Thoughtcited by 35
  7. 2025OptimalThinkingBench: Evaluating Over and Underthinking in LLMscited by 35
  8. 2025Future of Work with AI Agents: Auditing Automation and Augmentation Potential across the U.S. Workforcecited by 33
  9. 2025What Prompts Don't Say: Understanding and Managing Underspecification in LLM Promptscited by 32
  10. 2025Simple Prompt Injection Attacks Can Leak Personal Data Observed by LLM Agents During Task Executioncited by 25
  11. 2025AI-Slop to AI-Polish? Aligning Language Models through Edit-Based Writing Rewards and Test-time Computationcited by 24
  12. 2025Prototypical Human-AI Collaboration Behaviors from LLM-Assisted Writing in the Wildcited by 24
38 more
  1. 2026How AI Impacts Skill Formationcited by 20
  2. 2025Reality Check: A New Evaluation Ecosystem Is Necessary to Understand AI's Real World Effectscited by 18
  3. 2025Evaluating LLM Metrics Through Real-World Capabilitiescited by 17
  4. 2025How Do AI Agents Do Human Work? Comparing AI and Human Workflows Across Diverse Occupationscited by 17
  5. 2025The AI Gap: How Socioeconomic Status Affects Language Technology Interactionscited by 15
  6. 2025Intelligence per Watt: Measuring Intelligence Efficiency of Local AIcited by 13
  7. 2025Working with AI: Measuring the Applicability of Generative AI to Occupationscited by 12
  8. 2025When Two LLMs Debate, Both Think They'll Wincited by 10
  9. 2026How Well Does Agent Development Reflect Real-World Work?cited by 10
  10. 2025Improving Generative Ad Text on Facebook using Reinforcement Learningcited by 9
  11. 2026LLMs Corrupt Your Documents When You Delegatecited by 9
  12. 2025Translation in the Hands of Many:Centering Lay Users in Machine Translation Interactionscited by 8
  13. 2025EconEvals: Benchmarks and Litmus Tests for Economic Decision-Making by LLM Agentscited by 8
  14. 2025Readers Prefer Outputs of AI Trained on Copyrighted Books over Expert Human Writerscited by 8
  15. 2026AgentIF-OneDay: A Task-level Instruction-Following Benchmark for General AI Agents in Daily Scenarioscited by 8
  16. 2026The Hot Mess of AI: How Does Misalignment Scale With Model Intelligence and Task Complexity?cited by 8
  17. 2025The Pluralistic Moral Gap: Understanding Judgment and Value Differences between Humans and Large Language Modelscited by 7
  18. 2026How are AI agents used? Evidence from 177,000 MCP toolscited by 7
  19. 2025REALM: A Dataset of Real-World LLM Use Casescited by 6
  20. 2025Measuring AI "Slop" in Textcited by 6
  21. 2026Persona Vectors in Games: Measuring and Steering Strategies via Activation Vectorscited by 6
  22. 2026LLM Spirals of Delusion: A Benchmarking Audit Study of AI Chatbot Interfacescited by 6
  23. 2025What happens when generative AI models train recursively on each others' outputs?cited by 5
  24. 2025PersonaFuse: A Personality Activation-Driven Framework for Enhancing Human-LLM Interactionscited by 5
  25. 2025Urania: Differentially Private Insights into AI Usecited by 4
  26. 2025Death of the Novel(ty): Beyond n-Gram Novelty as a Metric for Textual Creativitycited by 4
  27. 2026Agentic Search in the Wild: Intents and Trajectory Dynamics from 14M+ Real Search Requestscited by 4
  28. 2026Investigating the Effects of LLM Use on Critical Thinking Under Time Constraints: Access Timing and Time Availabilitycited by 4
  29. 2025Access Controls Will Solve the Dual-Use Dilemmacited by 3
  30. 2025Show or Tell? Modeling the evolution of request-making in Human-LLM conversationscited by 3
  31. 2025EVALUESTEER: Measuring Reward Model Steerability Towards Values and Preferencescited by 3
  32. 2025TheMCPCompany: Creating General-purpose Agents with Task-specific Toolscited by 3
  33. 2026Designing Benchmarks for Knowledge Workcited by 3
  34. 2025Generative AI Practices, Literacy, and Divides: An Empirical Analysis in the Italian Contextcited by 2
  35. 2026The Facade of Truth: Uncovering and Mitigating LLM Susceptibility to Deceptive Evidencecited by 2
  36. 2026Understanding Usage and Engagement in AI-Powered Scientific Research Tools: The Asta Interaction Datasetcited by 2
  37. 2026LLMpedia: A Transparent Framework to Materialize an LLM's Encyclopedic Knowledge at Scalecited by 2
  38. 2026AI Fiction in the Wildcited by 2

50 most cited of 88 All 88 (JSON)