A new analysis from the Federal Reserve Bank of Richmond shows that job-finding rates for primary workers—those with steady employment histories—have fallen 13 percentage points since November 2022, the largest decline among worker groups studied. The drop contrasts with historical trends, where primary workers typically maintained higher job-finding rates than secondary workers.
Roles most exposed to AI—defined by task overlap with AI patent capabilities—have experienced the steepest declines in job-finding rates since ChatGPT’s public release in late 2022. Before 2023, job-finding rates for AI-exposed and non-exposed roles moved in tandem. Since then, outcomes for AI-exposed fields have worsened significantly, even as the broader U.S. economy has grown.
The analysis categorizes workers into three groups: primary workers (55% of the labor pool), secondary workers (14%), and a third group not explicitly detailed in the sources. Primary workers, historically the most resilient in job searches, now face the sharpest declines. Secondary workers, who typically experience higher unemployment rates, have seen a smaller decline of 2 percentage points in the same period.
Economists and labor analysts have noted that this shift defies past recession patterns, where primary workers’ job-finding rates fell less dramatically than those of secondary workers. The Federal Reserve’s findings suggest that AI integration may be reshaping labor market dynamics in ways not fully captured by traditional economic indicators.
The study does not attribute causality but highlights the correlation between AI exposure and declining job-finding rates. Researchers define AI-exposed roles as those where job tasks align with capabilities listed in AI patents, a metric used to assess vulnerability to automation.
While the U.S. economy has added jobs in sectors like healthcare and construction, the Federal Reserve’s analysis indicates that job losses in July underscore ongoing labor challenges. The data points to a labor market where traditional employment stability no longer guarantees quick reemployment, particularly in fields most susceptible to AI-driven changes.
Labor economists caution that these trends could signal deeper structural shifts, with potential long-term implications for workforce adaptability and wage growth in AI-exposed sectors. The Federal Reserve has not yet released a formal policy response to these findings, but the analysis has prompted discussions among policymakers about the need for targeted workforce development initiatives.
The Richmond Fed’s study is based on job-finding rate data spanning from November 2022 to September 2025, providing a three-year window to assess labor market changes. The analysis does not include projections for future trends but serves as a baseline for evaluating the impact of AI on employment stability.