Methodology

How we analyze and measure AI's impact on occupations.

Data Sources

Our analysis is based on peer-reviewed research from Anthropic, OpenAI, Google DeepMind, government labor statistics, and academic publications. We continuously monitor new research to update our assessments.

Analysis Framework

We decompose each occupation into individual tasks and assess the degree to which current and near-future AI systems can automate each task. This task-level approach provides more nuanced insights than whole-occupation estimates.

Key References

All data points are linked to their original sources. We provide full citation information for transparency and independent verification.

References

All data sources and research papers cited in our analysis.

12 references

  1. [1]Report

    Anthropic Research Team

    “Labor market impacts of AI: A new measure and early evidence”

    Anthropic, 2026.

    Introduces 'observed exposure' metric combining theoretical LLM capabilities with real-world Claude usage data. Finds Computer Programmers at 75% coverage, while actual adoption remains far below theoretical capacity.

  2. [2]Report

    Ruth Appel, Maxim Massenkoff, Peter McCrory, Miles McCain, Ryan Heller, Tyler Neylon, Alex Tamkin

    “Anthropic Economic Index report: economic primitives”

    Anthropic, 2026.

    Defines five economic primitives for AI task classification: complexity, skills, use case, autonomy, and success rate. 34% of Claude.ai usage in Computer & Math occupations.

  3. [3]Working Paper

    Andrew Johnston, Christos Makridis

    “The Labor Market Effects of Generative AI: A Difference-in-Differences Analysis”

    SSRN, 2025.

    Applies difference-in-differences methodology to measure generative AI's labor market effects across occupations.

  4. [4]Paper

    Erik Brynjolfsson, Bharat Chandar, Ruyu Chen

    “Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence”

    Stanford Digital Economy Lab, 2025.

    Young software developers (22-25) see ~20% employment decline from 2022 peak. 13% decline for early-career workers in AI-exposed occupations. Uses ADP payroll microdata.

  5. [5]Report

    Kunal Handa, Alex Tamkin, Miles McCain, Saffron Huang, Esin Durmus, Sarah Heck, Jared Mueller, Jerry Hong, Stuart Ritchie, Tim Belonax, Kevin K. Troy, Dario Amodei, Jared Kaplan, Jack Clark, Deep Ganguli

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

    Anthropic, 2025.

    Analyzes millions of Claude conversations to map AI usage to O*NET occupational tasks. 36% of occupations have significant AI task coverage.

  6. [6]Working Paper

    Menaka Hampole, Dimitris Papanikolaou, Lawrence DW Schmidt, Bryan Seegmiller

    “Artificial Intelligence and the Labor Market”

    National Bureau of Economic Research, 2025.

    Instruments for firm-level AI adoption using historical university hiring networks. Firms with AI-related hiring history face lower adoption costs.

  7. [7]Article

    Sarah Eckhardt, Nathan Goldschlag

    “AI and Jobs: The Final Word (Until the Next One)”

    Economic Innovation Group (EIG), 2025.

    Finds AI effect on jobs 'invisible' by conventional metrics. Highly exposed workers show 0.30pp unemployment increase vs. 0.94pp for less exposed. Only ~9% of businesses report using AI.

  8. [8]Dataset

    U.S. Bureau of Labor Statistics

    “Employment Projections: 2024-2034”

    U.S. Bureau of Labor Statistics, 2024.

    Projects 5.2M new jobs 2024-2034 (+3.1% total). Computer & Math +10.1%. Retail declining. AI impacts incorporated for first time in BLS projections.

  9. [9]Paper

    Xiang Hui, Oren Reshef, Luofeng Zhou

    “The Short-Term Effects of Generative Artificial Intelligence on Employment”

    Organization Science, 2024.

    Studies effects of generative AI on freelance platforms. Finds immediate negative impact on earnings and employment for workers in AI-exposed tasks.

  10. [10]Paper

    Tyna Eloundou, Sam Manning, Pamela Mishkin, Daniel Rock

    “GPTs are GPTs: An early look at the labor market impact potential of large language models”

    arXiv, 2023.

    80% of U.S. workforce could have 10%+ tasks affected by LLMs. 19% may see 50%+ tasks impacted. Introduces beta task exposure metric (0, 0.5, 1).

  11. [11]Paper

    Daron Acemoglu, David Autor, Jonathon Hazell, Pasciano Restrepo

    “Artificial Intelligence and Jobs: Evidence from Online Vacancies”

    Journal of Labor Economics, 2022. DOI: 10.1086/718327

    Analyzes AI's impact on job postings using vacancy data. Finds AI adoption displaces some tasks while creating demand for new AI-complementary skills.

  12. [12]Paper

    Daron Acemoglu, Pasciano Restrepo

    “Robots and Jobs: Evidence from US Labor Markets”

    Journal of Political Economy, 2020. DOI: 10.1086/705716

    Estimates that one additional robot per thousand workers reduces employment-to-population ratio by 0.2pp and wages by 0.42%.