Stanford Study: Entry-Level Employment Gap for Young Workers in AI-Exposed Jobs Widens to 19%

A revised Stanford Digital Economy Lab study using ADP payroll data finds that employment for workers ages 22–25 in highly AI-exposed occupations now stands 19% below where it would be if it had kept pace with less-exposed peers — up from 15% in July 2025. The gap is driven by reduced hiring, not layoffs, and is concentrated in roles where AI automates rather than complements human work.

Key Takeaways

A revised Stanford Digital Economy Lab study using ADP payroll data finds that employment for workers ages 22–25 in highly AI-exposed occupations now stands 19% below where it would be if it had kept pace with less-exposed peers — up from 15% in July 2025.
The gap is driven by reduced hiring, not layoffs, and is concentrated in roles where AI automates rather than complements human work.

AI synthesis3 sources9

A major updated study from Stanford's Digital Economy Lab reveals that the employment gap for young workers in AI-exposed occupations has widened significantly, reaching 19% by mid-2026 — up from 15% just a year earlier.

The research, titled "Canaries in the Coal Mine? Six Facts About the Recent Employment Effects of Artificial Intelligence," uses payroll data from ADP to track how employment has evolved since the release of ChatGPT. Led by economists Erik Brynjolfsson, Bharat Chandar, and Ruyu Chen, the study documents six key findings.

The headline finding: employment among workers ages 22–25 in highly AI-exposed occupations now stands about 19% below where it would be if it had kept pace with similarly aged workers in less-exposed occupations. In absolute terms, employment in the most exposed occupations fell about 11% between November 2022 and June 2026, while employment in the least exposed grew about 10%.

Crucially, the study finds no widespread, economy-wide job displacement. Experienced workers show no comparable gap, and the adjustment appears to operate primarily through reduced hiring of young workers rather than increased layoffs.

A key insight from the revised paper is the distinction between codified and tacit knowledge. Employment has declined among young workers in occupations relying on codified knowledge — formal, standardized, documented knowledge teachable through textbooks and procedures. Meanwhile, employment has increased among experienced workers in roles relying on tacit knowledge acquired through practice and mentorship.

This pattern is consistent with a world where generative AI excels at reproducing and applying knowledge encoded in text and digital information, while experience-based knowledge remains harder to replicate.

The declines are concentrated specifically in occupations where AI usage tends to automate human tasks. In occupations where AI is used more to complement workers, employment is flat or rising, particularly among experienced workers. So far, the adjustment shows up primarily in employment rather than base pay.

The researchers caution that while the patterns are suggestive, the data alone cannot establish causation. However, the divergence continued widening through mid-2026, well after interest rates peaked, and the concentration in automating occupations with a clear age gradient is not naturally explained by interest rates, education, or remote work alone.

The Stanford Digital Economy Lab has launched the AI Economic Indicators dashboard, which will update key results monthly to track whether these patterns accelerate, stabilize, or reverse.