AI Labor Market Tracker: September 2026
What we're reading this month
A brief look at other research on AI and the labor market that informed how we read this month's results.
Bick, A., Blandin, A., Deming, D. J., & Schumacher, T. R. (2026). What Work Does Generative AI Do? NBER Working Paper No. 35677.
Using a nationally representative survey linking genAI adoption to detailed occupations and tasks, the authors build the first task-level genAI adoption indexes and find that current adoption is widespread but shallow—genAI is used across many occupations and tasks, but within most of them fewer than half of workers have adopted it—suggesting that who adopts AI may matter as much as which tasks it can assist.
Chandar, B., & Klein Teeselink, B. (2026). How Does AI Change Labor Demand? Evidence from 41 Countries. Working paper, Stanford Digital Economy Lab, September 21, 2026.
Analyzing 1.25 billion job postings and 154 million employment records across 41 countries, the authors find that foreign affiliates of AI-adopting companies reduce the junior share of their workforce relative to comparable control affiliates—driven by growth in senior employment rather than a fall in junior employment—with senior roles shifting toward AI-exposed occupations while junior hiring shifts away from them.
Luduvice, A. V. D., & Pinheiro, R. B. (2026). AI-Augmented Capital-Skill Complementarity. Federal Reserve Bank of Cleveland Working Paper No. 26-22.
Modeling AI as a distinct capital input complementary to the high-skill/equipment composite, the authors find that rising AI usage can initially contract output by straining scarce complementary inputs even as falling AI prices support long-run growth; they also show that large-scale universal basic income funded through consumption taxes would slow AI investment, benefiting lower-skill workers but leaving high-skill workers and entrepreneurs worse off.
Imas, A., & Schaal, J. (2026). Has AI Impacted the Labor Market Yet? Substack, September 29, 2026.
Reviewing the empirical evidence through September 2026, the authors describe aggregate stability coexisting with subgroup displacement: overall employment and unemployment show little sign of AI-driven disruption, but evidence of reduced hiring for early-career workers in AI-exposed white-collar roles is real, if still contested and complicated by remote-work exposure correlating strongly with AI exposure. They note that actual AI adoption remains shallow relative to its potential—workers use AI in only about 21% of tasks where it could plausibly apply—so broader disruption may still be ahead as adoption deepens.