Work · updated Oct 2, 2026
Is AI showing up in jobs yet?
Only in places, so far. The numbers below compare AI-exposed work with less-exposed work, and countries with each other, over the same period. That is as close as public data gets to the question. Rates, the business cycle, and the post-pandemic hiring hangover move all of these too, so read them as context, not proof.
Eight economies, side by side
Changes are since November 2022, the month ChatGPT launched. Pick an economy for its full page.
| Economy | Unemployment | Youth unemployment | Postings since ChatGPT: exposed / less exposed | Firms using AI | AI usage index | Productivity, y/y |
|---|---|---|---|---|---|---|
| United States | 4.2% (+0.6 pt) | 9.9% (+1.8 pt) | -38.7% / -29.4% | 5.7% (2021) | 3.9× | +2.2% (Q2 2026) |
| United Kingdom | 4.9% (+1 pt) | 15.8% (+4.1 pt) | -59.7% / -60.2% | no data | 3.4× | +1.5% (2025) |
| Germany | 3.8% (+0.7 pt) | 7.2% (+1.3 pt) | -50.2% / -17.5% | 26% (2025) | 2.4× | +1.1% (2025) |
| France | 8.2% (+1.1 pt) | 21.4% (+4.6 pt) | -56.8% / -48.1% | 18.2% (2025) | 4.0× | +1% (2025) |
| Canada | 6.6% (+1.6 pt) | 13.4% (+3.2 pt) | -39.7% / -40.5% | 8.3% (2023) | 4.1× | +1.6% (2025) |
| Japan | 2.5% (±0) | 3.6% (−1.3 pt) | no data | 9.2% (2023) | 1.9× | +1.2% (2025) |
| European Union (27) | 6.1% (−0.1 pt) | 15.4% (+0.9 pt) | no data | 19.9% (2025) | – | +1.7% (2025) |
| OECD average | 5% (+0.1 pt) | 11.4% (+0.8 pt) | no data | no data | – | +1.4% (2025) |
Source: Unemployment: BLS (US), OECD harmonised rates, Eurostat (EU). Postings: Indeed Hiring Lab. Firms using AI: OECD business ICT surveys (10+ employees; years differ, grey means old). AI usage index: Anthropic Economic Index (2026-05), Claude usage relative to working-age population (1.0 = proportional). Productivity: BLS output per hour (US), World Bank output per worker. · US Bureau of Labor Statistics · OECD Data Explorer · Eurostat · Indeed Hiring Lab job postings tracker · ICT Access and Usage by Businesses · Anthropic Economic Index · World Development Indicators
Job postings: exposed versus less-exposed work
Indeed postings indexed to November 2022 = 100. If AI were cutting demand for desk work, the purple line should fall further than the green one.
How the groups are built
AI-exposed: Software Development, Data & Analytics, IT Systems & Solutions, Media & Communications, Marketing, Accounting, Banking & Finance, Insurance, Customer Service, Administrative Assistance, Human Resources, Legal.
Less exposed: Nursing, Personal Care & Home Health, Construction, Food Preparation & Service, Driving, Cleaning & Sanitation, Loading & Stocking, Installation & Maintenance, Childcare, Production & Manufacturing, Dental, Therapy.
The split follows where measured AI usage concentrates (Anthropic Economic Index) and where the ILO's exposure index is highest. Each category is rebased to its November 2022 level, then the categories in a group are averaged equally.
Caveat: tech and finance hiring boomed in 2021–22 and cooled with interest rates, so part of any gap is the hiring cycle, not AI.
Source: Job postings: Indeed Hiring Lab (CC BY 4.0), seasonally adjusted daily indices averaged by month. · Indeed Hiring Lab job postings tracker
What the careful studies find
These studies do the causal work this page cannot: matched comparisons, payroll records, and exposure measures. Their answers so far: sharp effects at specific links, slow change overall.
Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of AI
Brynjolfsson, Chandar & Chen (2025)
Using ADP payroll records, early-career workers (ages 22–25) in the most AI-exposed occupations saw about a 13% relative decline in employment since late 2022, while older workers in the same jobs held steady. Firms adjusted through hiring, not pay.
Evaluating the Impact of AI on the Labor Market: Current State of Affairs
The Budget Lab at Yale (2025, updated monthly)
No discernible disruption to the broader labor market since ChatGPT's release: the mix of occupations is shifting at about the pace of earlier technology waves, and exposure measures do not yet predict who loses work.
Generative AI and Jobs: A Refined Global Index of Occupational Exposure
Gmyrek et al., ILO Working Paper 140 (2025)
About one in four workers worldwide is in an occupation with some exposure to generative AI, and 3.3% are in the most exposed group. Exposure covers 34% of jobs in high-income countries versus 11% in low-income ones. The likeliest effect is changing jobs, not eliminating them.
Gen-AI: Artificial Intelligence and the Future of Work
Cazzaniga et al., IMF Staff Discussion Note (2024)
Almost 40% of global employment is exposed to AI, rising to about 60% in advanced economies. Roughly half of exposed jobs may gain from AI; for the other half, AI may take over key tasks and lower demand for labor.
Country pages: United States · United Kingdom · Germany · France · Canada · Japan · European Union (27) · OECD average