Entry-level employment falls 19% in most AI-exposed jobs, Stanford study finds

A recent Stanford University study reveals that entry-level employment has fallen by 19% in jobs most exposed to artificial intelligence (AI), particularly for young workers aged 22 to 25. The study, titled 'Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence', was published in August 2026 and offers fresh data on the employment trends first identified last year.

According to the research, employment levels for these young workers in AI-exposed occupations are 19% below those in less AI-affected fields, compared to a 13% gap last year. The study attributes this divide to lower hiring rates for entry-level workers in AI-impacted fields and primarily focuses on the decrease in job opportunities rather than lower wages.

The researchers caution that the effects on young workers are persistent and widening, raising concerns about the labor market's ability to accommodate those starting their careers. Erik Brynjolfsson, the lead researcher, expresses worry over the labor market's capacity to maintain a stable employment level while simultaneously closing opportunities for new entrants.

The study analyzed anonymized payroll data from HR management company ADP and rated occupations' exposure to AI disruption using the Anthropic Economic Index. Jobs requiring heavily codified knowledge, such as those involving practice, mentorship, and real-world exposure, are less likely to be replaced by AI, while heavily codified knowledge jobs are more susceptible to AI disruption.

The research suggests that higher education might provide a buffer against these employment effects, with occupations having a higher share of college graduates experiencing more muted differences between more exposed and less exposed occupations.