Young workers paying price of AI boom, and the data keeps getting worse

Artificial intelligence is shrinking job opportunities for young workers at an accelerating rate, according to new data from Stanford economist Erik Brynjolfsson and ADP Research. The findings, published through the newly launched Canaries Dashboard, draw on payroll records covering roughly one in six American workers across more than 730 occupations.

The top-line numbers look calm. Employment in the most AI-exposed occupations fell just 0.2% year over year as of April 2026. But the picture changes sharply when sorted by age.

A widening gap by career stage

• Workers aged 22 to 25 in highly AI-exposed roles: down 3.8% per year
• Workers aged 31 to 34 in the same roles: down 1.7% per year
• Workers aged 35 to 40 in the same roles: growing at 2% per year

The reason, Brynjolfsson argues, is structural. AI first absorbs routine tasks: summarising, formatting, scheduling, and retrieving information. These are precisely the tasks typically assigned to people early in their careers. Senior workers hold job-specific knowledge that is harder to automate. Junior workers have not yet built that buffer.

ADP chief economist Nela Richardson frames the distinction as augmentation versus automation. Where AI amplifies what humans do, employment holds up. Where it replaces tasks outright, contraction follows. Entry-level workers sit squarely in the second category.

Brynjolfsson tested the finding against competing explanations: interest rate sensitivity, tech-sector overhiring, remote work effects. He removed the entire tech industry from the dataset. The pattern held each time. “Whatever it is,” he told Fortune, “it’s not going away.”

The trend has not reversed. Since its first measurement, the early-career decline has grown by roughly half a percentage point per month. The dashboard, a partnership between Brynjolfsson’s Stanford Digital Economy Lab and ADP Research, will update continuously.

Meanwhile, a separate initiative is taking shape on the policy side. A coalition of employers including Amazon, Bank of America, OpenAI, and Anthropic has raised 500 million dollars for a nonprofit called Raise Us, led by former commerce secretary Gina Raimondo. The group will fund pilot programmes in four US states, focusing on retraining, wage insurance for workers who accept lower-paying roles, and technical support for companies that want to redeploy staff rather than cut them.

Whether retraining can keep pace with displacement remains an open question. As one policy researcher told the New York Times: “What we think is resilient now might not be resilient later.”

Sources

• ‘It’s not going away’: The Stanford economist who called the AI entry-level jobs crisis early has the receipts – Fortune
• Big Companies Aim to Ease A.I. Transition for American Workers – The New York Times