Amid warnings of an AI-driven jobs apocalypse, there is mounting evidence to suggest that a senior professional with experience, status, and deep social network may be relatively safe from being replaced by AI. A young college graduate vying for an entry-level job, on the other hand, may face a very different reality.

New research from economists at Stanford University suggests that AI is increasingly driving the disappearance of entry-level roles for younger workers in some fields. The August 2026 paper titled ‘Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence’, provides fresh data and refined statistics that largely holds up the employment trends first identified for entry-level workers last year.

In particular, the updated research found that employment levels for workers aged 22 to 25 years in the most ‘AI-exposed’ occupations are 19 per cent below those of their peers in fields less exposed to AI disruption. The gap between the two categories was measured at 13 per cent last year.

Story continues below this ad

The researchers traced this divide back to lower hiring rates for entry-level workers in AI-impacted fields. They also found that the impact on the 22-25 age group has primarily manifested itself as fewer jobs, rather than lower wages.

The Stanford University study is another indicator of the job losses that many AI industry watchers have warned about as AI systems become more capable of replicating most human tasks. However, there is little consensus on which jobs are most vulnerable to AI, which are likely to remain safe, and which industries and economies will be hit the hardest.

Erik Brynjolfsson, the lead researcher behind the study, envisions a near future in which jobs for those employed in the pre-AI era largely persist while many jobs for the incoming working-age cohort start disappearing. “The entry-level effects we’re measuring are real, persistent and widening, and I’m more worried than I was about a labor market that keeps its overall employment level while quietly closing the on-ramp for people starting their careers,” Brynjolfsson was quoted as saying by Washington Post.

Methodology

The study’s findings are based on an analysis of a large sub-sample of anonymised, high-frequency payroll data regularly aggregated by HR management company ADP.

Story continues below this ad

The researchers said they also rated each occupation’s exposure to AI disruption using a potential labour market impact gauge established by previous research papers such as the Anthropic Economic Index, which looks at how various occupations actually use the Claude model in their everyday work. Google also published a similar report based on occupational Gemini usage last month.

Other key findings

As per the analysis, the researchers found little to no difference in relative overall employment between the jobs judged most and least affected by AI on these metrics.

But after separating out workers aged 22 to 25, they found that employment in the top 40 percent of AI-impacted jobs had fallen by about 11 percent. In the 60 percent of jobs with the least AI impact, total employment for those young workers grew by 10 percent over the same period.

In terms of which sectors are seeing this impact, the researchers highlighted jobs requiring heavily codified knowledge that can be taught through education, textbooks, or written procedures.

Story continues below this ad

On the other hand, jobs involving practice, mentorship, and repeated exposure to real situations will emerge as the roles where AI tools complement the worker rather than replace them.

This finding is supported by Anthropic’s latest Economic Index, which differentiates between queries related to tasks that are ‘automative’ or ‘augmentative’.

What can college graduates do?

While the Stanford study offers a stark picture for young college graduates, it has stated that higher education might still serve as a buffer against the employment effects identified in the research paper.

It found that occupations with a higher share of college graduates showed more “muted differences between more-exposed and less-exposed occupations” regarding AI. In jobs with few college graduates, “the least AI-exposed occupations [saw jobs] growing and the most exposed occupations [were] declining in employment.”