U.S. Federal Reserve officials recognize that artificial intelligence is set to transform large parts of the economy, yet they disagree sharply over the timing and scale of those changes and the implications for labor markets and inflation.
The debate intensified after fintech firm Block said it would reduce its workforce by about 40%, equal to roughly 4,000 jobs, attributing the cuts to shifts in how work is performed because of AI. That announcement underscored how quickly companies can alter staffing as they adopt new technologies, and it has fed into a broader conversation inside the Fed about whether such developments will ultimately ease or amplify inflationary pressures.
Traditionally, rising layoffs are seen as a signal that central bankers can loosen policy because weaker labor markets reduce demand and ease price pressures. But some Fed officials and outside economists now argue the dynamic may be different with AI: while unemployment could rise, the gains from increased capital returns and higher pay for those benefiting from automation may sustain inflationary forces. Displaced workers could also take longer to regain employment, complicating the usual policy response.
Adam Posen, president of the Peterson Institute for International Economics, summarized this view in remarks about inflation: AI-related changes represent a "positive, real shock" that mainly shows up as higher real incomes for some households rather than disinflation. Posen pointed to stock-market gains lifting wealth and to large-scale investment in computing and infrastructure that can strain local electricity supply and construction resources - developments that, he said, are likely to add to price pressures.
On the other side of the debate is a view that AI will be disinflationary in the near term by boosting productivity. That outlook is embodied by Kevin Warsh, recently nominated to lead the Fed, who has argued that AI's productivity advances warrant lower interest rates. In a November opinion piece he wrote that AI is "a significant disinflationary force, increasing productivity and bolstering American competitiveness," and he suggested monetary policy could and should reflect that reality. Warsh compares his forward-looking posture to earlier episodes when central bankers factored technological improvements into policy decisions.
But several Fed officials have pushed back on giving AI credit for rapid productivity gains. They emphasize that some of the efficiency improvements observed recently may be continuations of adjustments that began during pandemic-related labor shortages rather than a sudden AI effect. Moreover, officials are cautious about assuming broad-based job creation will offset displacement quickly, as happened with other historical technology transitions.
The emerging split over AI's effects extends into interpretations of how unemployment interacts with inflation. Within the Fed's framework is a concept of a long-run "natural" rate of unemployment, currently estimated near 4.2%. Officials worry that if AI raises productivity while causing more churn in the labor market, a higher unemployment rate could coexist with robust growth and still not relieve inflationary pressure. As Fed Governor Lisa Cook observed, an increase in unemployment during a productivity-driven expansion may not signal increased slack that monetary easing could safely address without reigniting inflation.
Not all analysts agree. Krishna Guha, vice chair at Evercore ISI, has argued that AI may weaken worker bargaining power and lead employees to accept lower wage increases, producing downward pressure on inflation. That perspective reaches similar policy conclusions to Warsh on reducing rates, though it rests on different mechanics - labor-market adjustments rather than an immediate productivity windfall.
Adding to policymakers' unease is the character of the AI transition itself. Unlike earlier waves of automation that largely affected manual or production jobs, AI appears capable of performing many white-collar tasks such as coding and data analysis. Block's chief executive Jack Dorsey attributed his company's cuts to AI-enabled changes in how teams are organized, saying AI "paired with smaller and flatter teams, are enabling a new way of working which fundamentally changes what it means to build and run a company." Such shifts suggest the disruption could reach employment sectors that have been central targets of educational and workforce development efforts.
Independent research has documented the breadth of tasks AI could alter. A 2024 study by Brookings Institution analysts estimated that more than 30% of U.S. workers could see half of their job tasks disrupted, and the article notes that those percentages have likely increased since the paper's release. Thought experiments and warnings about a potential jobs apocalypse have even prompted swift market reactions - for example, a study by Citrini Research spurred a brief but notable stock selloff, illustrating how sensitive investors are to prospects of rapid labor-market change.
Recognizing the strategic importance of the topic, Fed attention to AI has surged. An AI-driven count of the central bank's research articles and policymaker speeches on AI, machine learning and related subjects found virtually none before the release of ChatGPT in late 2022. The number rose to five in 2023, about 17 last year, and already 14 this year - a clear acceleration. Minutes from the Fed's January meeting recorded an extensive discussion of productivity and AI, and at least five policymakers addressed the matter publicly last month.
Despite growing engagement, most Fed officials are not yet prepared to treat AI as justification for cutting interest rates. They see productivity improvements but remain reluctant to attribute them primarily to AI rather than to other recent efficiency gains. Many policymakers also anticipate AI may create a structurally higher unemployment rate that lower interest rates would not easily remedy without risking higher inflation.
The debate reflects deeper uncertainties about how AI will reshape resource demand and returns on capital. Posen highlighted large capital investment associated with AI deployment, which can squeeze local construction and power markets and thus generate additional cost pressures. At the same time, wealth gains among households - partly from stock market gains tied to AI optimism - may sustain consumption and demand, supporting inflation.
Richmond Fed President Tom Barkin captured the uncertainty succinctly: there are numerous forecasts about AI's rollout, efficacy, energy needs, and labor-market consequences, and the only certainty is that many of these forecasts will be wrong. Whether they err on the side of optimism or pessimism will become clear only as developments unfold, he said.
The policymaking challenge is to navigate several moving parts: task-level disruption across occupations, shifts in bargaining power, differential gains across workers and firms, investment-driven resource constraints, and changing productivity dynamics. These forces point in different directions for inflation and employment, leaving the Fed to weigh outcomes and consider whether conventional demand-side tools will remain effective if AI-driven dislocation persists.
Amid this uncertainty, market participants and observers are seeking ways to assess corporate and investment opportunities in the AI era. The original article included a description of ProPicks AI, a tool that evaluates companies using artificial intelligence across many financial metrics to generate stock ideas. That promotional content underscores how private-sector actors are also leveraging AI to inform investment decisions even as policymakers parse its macroeconomic implications.
As AI adoption expands, Fed officials and outside analysts will continue to test competing narratives: that AI represents a strong, near-term disinflationary force via productivity gains; or that it will foster structural labor-market frictions, wealth effects and resource constraints that sustain or even raise inflation. For now, the balance of views at the Fed appears cautious, and officials are preparing to monitor evolving data rather than relying on a single consensus forecast about AI's economic trajectory.