Fed officials warn AI's economic costs may arrive faster than benefits

Don't count on AI to solve America's inflation problem: That's the message from several Federal Reserve officials who warn that the promise of an AI-fueled productivity boom might not justify cheaper money.Why it matters: How AI shapes inflation and productivity will be a defining question for the Fed under the leadership of Kevin Warsh, who has staked out a case that the technology's supply-side benefits justify keeping rates low.Some Fed officials say they see clearer evidence of AI-related investment boosting demand for labor, equipment and infrastructure than they do of widespread productivity gains.The upshot: Inflation risks look more immediate than any AI-related productivity benefits, especially as inflation remains stubbornly above the Fed's target.What they're saying: "I believe it would be risky to rely on the prospect of higher productivity growth in the future to solve our inflation problem today," St. Louis Fed president Alberto Musalem said in a speech last week."AI shows great promise as a transformative technology, but the risks of a miscalculation about its impact on productivity and inflation are too great," Musalem said."[A]t present, I believe we should keep our guard up against persistent above-target inflation today, rather than base monetary policy on the hope that we will have higher productivity growth tomorrow."The big picture: Warsh has argued that AI will be a "significant disinflationary force, increasing productivity and bolstering American competitiveness," as he wrote in a Wall Street Journal op-ed late last year.The theory is that if AI helps workers and businesses produce more with the same resources, the economy can grow faster without generating inflation, giving the Fed more room to lower interest rates.But policymakers want evidence that the productivity gains are here to stay.By the numbers: Productivity started to take off before most companies had adopted AI, making it difficult to know how much to credit AI for the productivity lift.Over the past three years, productivity has averaged about 2.4% annually, far stronger than the 1.5% rate seen during the 2010s, according to the Bureau of Labor Statistics.Between the lines: Internet-fueled productivity gains in the 1990s were visible "everywhere except in the statistics," San Francisco Fed president Mary Daly told Neil at the Reagan Economic Forum on Friday. "We've got the productivity surge a little bit earlier this time. But what's problematic is it's hard for economists or anyone to link it directly back to the AI investments. In fact, if you talk to companies, they say they haven't seen the productivity yet," Daly said."I'm bullish, but I want to see some more evidence that this is actually picking up durable, sustained gains in productivity — but I see all the green shoots there." The intrigue: A new World Economic Forum survey shows economists think most sectors won't see notable AI-driven productivity gains for another two years, a longer timeline than they anticipated at the start of 2026.Companies and investors in recent weeks have begun to publicly question whether the enormous costs of deploying AI are translating into output and efficiency gains.What to watch: Fed governor Lisa Cook pointed to signs that AI investment demand is pushing prices higher for chips, high-tech equipment and software, as well as for construction labor, electricity and water. That comes alongside price pressures from the Iran war and tariffs."[Y]et another shock to prices could be layered on from the heightened investment demand due to AI," Cook said in a speech last week, noting that companies have announced roughly $1.5 trillion in data center investment plans."Those figures suggest that substantial AI-related investment remains in the pipeline from data centers alone. Effects of this demand on prices are apparent."