AI didn't cause the tech jobs bust, but it explains why there's no recovery

The analysis compares seven years of employment, vacancies, layoffs and youth labor-market data across the United States, Europe, India and China. In the US, software-development postings peaked in February 2022, before the Federal Reserve began raising rates and nine months before ChatGPT launched. By ChatGPT’s release, postings had already fallen 36%; payroll employment declined later, consistent with the normal lag between job demand and hiring. Information-sector employment fell 12.1% from its November 2022 peak through September 2026, even as the broader US economy added about 4.8 million jobs. The timing points to rising interest rates and the unwinding of the hiring boom - not AI - as the cause of the initial tech downturn.

AI, the analysis argues, helps explain why hiring has not recovered and who is missing out. Although US software postings have risen from their 2025 low, 71% of the rebound has been in senior roles, while junior hiring remains weak; 56% of active tech postings now request AI skills. The pattern suggests that firms are seeking experienced workers able to use AI while offering fewer first- and second-job opportunities. Youth unemployment is also high in several measured markets: 19% among urban 15-to-29-year-olds in India, 18.9% among China’s 16-to-24-year-olds, and 14.5% among US teenagers. The evidence does not establish that AI alone caused these changes, but it supports a distinction between the financial forces that triggered the bust and the changing hiring patterns that may be prolonging it.