AI Hits Young Workers Hardest, Stanford Study Finds
A new study from Stanford University reveals that generative artificial intelligence is having an uneven impact on the U.S. job market, with early-career employees facing the harshest consequences. Workers aged 22 to 25 have seen employment in AI-exposed occupations fall by about 13 percent, according to research published online on August 26. More experienced workers, however, remain largely shielded, particularly those with tacit knowledge—the subtle, experience-based skills that AI struggles to replicate.
The findings, authored by Stanford’s Erik Brynjolfsson, Bharat Chandar, and Ruyu Chen, are based on payroll data from ADP, the largest provider in the United States. They show that job declines are concentrated in areas where AI can directly automate tasks, such as software development and customer service, while positions that depend more heavily on human judgment or personal interaction remain stable. “In contrast, employment for workers in less exposed fields and more experienced workers in the same occupations has remained stable or continued to grow,” the researchers noted.
OpenAI chief executive Sam Altman echoed this dynamic during a Federal Reserve Conference last month. Asked which fields could experience significant displacement, he remarked, “Some areas … are totally gone,” pointing to customer support roles already being replaced by AI bots. At the same time, he highlighted that programmers have become “ten times more productive” thanks to AI, with salaries in Silicon Valley rising accordingly. Still, he cautioned against overreliance on AI in critical settings, saying he would never entrust ChatGPT with medical treatment without a human doctor involved. Looking ahead, Altman predicted that robotics could bring an even larger disruption within three to seven years.
The Stanford team observed that the decline in entry-level employment began in late 2022, coinciding with the rapid rollout of generative AI tools. Chandar noted in a post on X that software developers between the ages of 22 and 25 have experienced a roughly 20 percent employment drop since their 2022 peak, while older workers in the same field continue to gain ground. At the same time, he emphasized that not all job losses stem from AI alone. Broader economic headwinds are also shaping the market. A report from outplacement firm Challenger, Gray & Christmas found that U.S. companies announced more than 62,000 job cuts in July, a 140 percent increase from the previous year, with AI cited as the reason for more than 10,000 of those cuts.
Still, not everyone sees a catastrophic outlook. Goldman Sachs researchers recently projected that widespread AI adoption could displace 6 to 7 percent of the U.S. workforce, but argued that the effect would likely be temporary as new opportunities emerge. “While these trends could broaden as adoption increases, we remain skeptical that AI will lead to large employment reductions over the next decade,” said Joseph Briggs of Goldman Sachs Research. The report pointed out that around 60 percent of today’s jobs did not exist in 1940, underlining how innovation consistently reshapes the labor force.
According to Goldman, the occupations most vulnerable to AI include programmers, accountants, auditors, legal and administrative assistants, customer service representatives, telemarketers, proofreaders, copy editors, and credit analysts. Roles least threatened by automation range from air traffic controllers and chief executives to radiologists, pharmacists, residential advisors, photographers, and clergy members.
