AI's impact on new graduates remains unclear as employment data shows little decline

Despite predictions that artificial intelligence could devastate entry-level job markets, fresh economic evidence suggests the threat may be overstated—at least for now. A new working paper from researchers at Munich's CESifo challenges concerns raised just weeks earlier, finding no widespread evidence that recent college graduates face reduced hiring or unemployment spikes tied to AI adoption.

Conflicting signals on AI's employment effects#

The CESifo analysis, titled "The Early Impacts of AI on Employment Among Recent College Graduates," stands in direct contrast to a Stanford University study released the previous month. While Stanford researchers documented weak entry-level hiring in occupations vulnerable to AI disruption, the CESifo team concluded plainly that "there is no evidence of any significant, widespread displacement or reduction in hiring of recent college graduates in absolute or relative levels."

Economists Robert Fairlie and Jane Wu focused specifically on this cohort because recent graduates represent the canary in the coal mine—companies are more likely to cut hiring for entry-level positions than to lay off experienced staff when automation becomes viable.

Why 2026 looks different#

There are legitimate reasons to think this particular year might mark a turning point. The researchers note a sharp uptick in Census survey responses showing firms actively replacing substantial employee workloads with AI systems. Meanwhile, corporate AI spending per worker and ChatGPT Enterprise usage both surged over the past 12 months—signals that adoption is accelerating.

High-profile executives have publicly warned of disruption ahead. Venture capitalist Marc Andreessen remarked that until late 2025, "AI literally was not actually good enough to do any of the jobs that they're actually cutting." BlackRock CEO Larry Fink was more ominous in March, suggesting that "when this year's college graduates enter the workforce, we could see the highest unemployment rate among them in years—even without a recession."

Measuring the unmeasurable#

To test whether such concerns materialized, CESifo researchers examined detailed labor statistics from the , tracking unemployment rates among recent bachelor's degree holders aged 22 to 25 who weren't pursuing graduate studies. Because hiring patterns predictably fluctuate seasonally when new graduates flood the market each summer, the team analyzed year-over-year trends stretching back to 2022—the dual inflection point when pandemic-era unemployment normalized and ChatGPT launched.