RWP 26-06, August 2026
We show that AI technologies are oriented toward jobs and workers that typically exhibit greater volatility in labor market outcomes over the business cycle. Occupations currently most exposed to AI are those that have historically exhibited (i) greater volatility in employment levels over business cycles, (ii) higher job-switching rates by workers, (iii) higher job-finding rates, and (iv) a lower likelihood for workers to exit the labor force following a job loss. The sectors of the U.S. economy that produce AI technologies have also historically exhibited high volatility in productivity. We then quantify how technology-led structural changes in economic activities contribute to aggregate business cycle volatility. Growth in the production of AI-based technologies in recent years increased the volatility of U.S. output by 2.8 percent, roughly 3.5 times what resulted from the late 1990s’ IT boom. If the orientation of AI toward occupations that exhibit higher variability in labor market outcomes results in a higher aggregate labor supply elasticity, the effects on aggregate volatility are even greater.
JEL classifications: E32, E37, J62, J63, O33
Article Citation
Munoz Henao, Juan David, and Nicholas Sly. 2026. “Is the AI Boom Volatility-Biased Technological Change?” Federal Reserve Bank of Kansas City, Research Working Paper no. 26-06, August. Available at External Linkhttps://doi.org/10.18651/RWP2026-06
The views expressed are those of the authors and do not necessarily reflect the positions of the Federal Reserve Bank of Kansas City or the Federal Reserve System.
Authors
Juan David Munoz Henao
Research Associate II
