Young workers face AI replacement in US workplaces: study

SACRAMENTO: A comprehensive study has revealed that artificial intelligence (AI) is beginning to displace entry-level employees in the United States, with workers aged 22-25 experiencing a 13% decline in employment in AI-exposed occupations since late 2022.

The study, published on Tuesday, was conducted by Stanford University economists Erik Brynjolfsson, Bharat Chandar and Ruyu Chen using data from Automatic Data Processing, the largest payroll processing firm in the United States. It tracked employment patterns across tens of thousands of companies through July 2025.

Their findings provided the first large-scale evidence that generative AI tools like ChatGPT are reshaping the job market.

"Since the widespread adoption of generative AI, early-career workers in the most AI-exposed occupations have experienced a 13% relative decline in employment even after controlling for firm-level shocks," the study said, noting that employment adjustments occurred primarily through job cuts rather than wage reductions.

Young software developers and customer service representatives bore the brunt of these changes. Employment for software developers aged 22-25 plummeted nearly 20% from its late 2022 peak, while older workers in the same occupations maintained stable or growing employment levels, according to the study.

The authors said young workers in AI-exposed jobs may be the first warning sign of wider changes in the broader workforce.

In the study, the researchers analysed monthly payroll records covering 3.5-5 million workers, linking job classifications to established measures of AI exposure developed by previous academic studies. They found that occupations where AI automates tasks showed the steepest employment declines, while jobs where AI augments human capabilities experienced continued growth.

The timing aligned with the November 2022 launch of ChatGPT, which sparked rapid adoption of AI tools across US workplaces. By mid-2025, 46% of US survey respondents reported using large language models at work, according to research cited in the study.

However, the study showed that AI impact varied significantly by occupation type. Health aides, nursing assistants, and other less AI-exposed roles showed employment growth for young workers. The researchers suggested that AI primarily replaces codified knowledge taught in schools rather than the tacit knowledge gained through work experience.

By RSS/Xinhua