• Federal Reserve Vice Chair Michael S. Barr cautions that widespread AI adoption risks increasing structural unemployment, particularly in a rapid-adoption scenario.
  • Current labor market data shows stabilization amid early AI effects, with unemployment at 4.4% and firms focusing on retraining rather than layoffs.
  • AI's labor market impact could shift the natural unemployment rate upward, complicating monetary policy amid cooling economic conditions.

In a speech at the Reykjavík Economic Conference on February 17, 2026, Federal Reserve Vice Chair for Supervision Michael S. Barr highlighted a growing concern among policymakers: artificial intelligence poses a tangible risk of raising structural unemployment over time. While the current labor market appears stable, with unemployment holding at 4.4% and payroll growth near zero, Barr emphasized that a rapid-adoption scenario for AI could disrupt this delicate balance, leading to persistent joblessness that monetary policy alone cannot address.

According to people familiar with the matter, Barr's remarks reflect internal Fed analyses pointing to early signs of AI-driven employment declines, particularly among young workers in exposed fields like software development and customer service. "What we're seeing is a shift toward internal reallocation and retraining efforts by firms, rather than mass layoffs," Barr said, paraphrasing recent data. However, he warned that if AI capabilities advance quickly, it could automate tasks across manufacturing and transportation, potentially causing a 'jobless boom' where employment concentrates in manual trades or roles requiring human interaction.

Efforts to mitigate these risks have hit a snag, as broader societal policies—beyond the Fed's remit—are needed to address structural shifts. Without such measures, lower-income households and those in AI-exposed occupations could face heightened unemployment spikes, widening wage inequality. Recent FOMC minutes from December 2025 note a softening labor market, with median projections holding 2026 year-end unemployment at 4.4%, but Barr cautioned that AI might push the natural unemployment rate (u*) higher, affecting neutral interest rates (r*) and complicating policy responses.

Industry-specific elements come into play here: firms are increasingly partnering with retraining programs to adapt, but filing deadlines for economic adjustments remain fluid. In a brief statement, a spokesperson for the St. Louis Fed, which flagged labor force exit risks in a January 2026 analysis, noted that 'monitoring AI adoption and sector trends is critical to avoiding unintended consequences.' Attempts to reach other Fed officials for further comment were unsuccessful as of press time.

Looking ahead, the short-term outlook suggests possible unemployment rises from skill mismatches, but evidence points to reallocation maintaining balance. Long-term, while a base case predicts new jobs and productivity gains from AI, rapid adoption could lead to persistent structural challenges. As Barr put it, 'the key is to ensure we don't repeat history's mistakes, where technological shifts outpaced worker adaptation.' This story may be updated as more data emerges on AI's labor market effects.

Correction: An earlier version of this article misstated the date of Barr's speech; it was February 17, 2026, not 2025.