Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Some links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.

Jerome Powell said the Federal Reserve was watching AI-related hiring freezes and layoffs “very, very, very carefully,” and acknowledged that AI could affect job creation. But “deeply concerned” was a headline’s characterization, not a verified quote from Powell. In October 2025, he also said the effects were not yet clearly showing up in initial unemployment claims.

What Powell said in October 2025

At a Federal Open Market Committee press conference on October 29, Powell responded to a question about companies announcing hiring reductions and layoffs while citing artificial intelligence. He said executives were frequently mentioning AI and that the Fed was watching the issue “very, very, very carefully.” He added that AI “could have implications for job creation.”

Powell drew a distinction between company announcements and broad labor-market evidence: at that point, he did not see the reported effects clearly in initial unemployment-claims data. His remarks described a risk the Fed was monitoring, not a finding that AI had already caused widespread unemployment. Read the October 29 press-conference transcript.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Where “deeply concerned” came from

Futurism used “deeply concerned” in its November 2, 2025 headline about Powell and AI. The phrase is an editorial description, not wording established by the official October transcript. Powell’s own language supports saying he was watching the issue closely and recognized uncertainty about job creation. Futurism’s article framed the remarks around weak job creation and corporate announcements.

A June 2025 congressional exchange may also help explain the framing. Representative Bill Foster said, “I’ve been very concerned” about AI and the job market before asking Powell about a potential employment shock. Powell addressed the possibility of disruption, but did not adopt Foster’s phrase as his own declaration. Powell’s June testimony and exchange.

What Powell had said earlier—and what he did not predict

In June 2025, Powell acknowledged that AI could initially replace some workers rather than simply make them more productive. He also pointed to the historical tendency for major technologies to raise productivity and eventually create new work. He cautioned that the timing and scale of AI’s effects were unknown and that productivity improvements could take longer to appear than expected.

That was not a forecast of imminent mass unemployment. Powell did not specify a number of jobs AI would eliminate, predict a recession, or say employment would permanently collapse. His position allowed for short-term displacement and longer-term gains without treating either outcome as certain.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

How the focus shifted from June to October

In June, lawmakers asked about a possible future “job shock,” including public predictions by AI-industry leaders about entry-level white-collar work. By October, Powell was responding to reported company decisions: hiring pauses, layoff announcements, and executives citing AI. The October comments were more observational and data-focused than predictive.

The FOMC’s October 29 statement separately said job gains had slowed and downside risks to employment had increased. That broader assessment did not attribute the labor-market slowdown to AI. See the October FOMC statement.

What the labor-market indicators can—and cannot—show

A company announcement is not the same thing as an observed increase in unemployment. A hiring freeze limits future additions; a layoff announcement may precede actual separations; and workers who lose jobs may file for benefits only later. Claims, payroll employment, unemployment, job openings, labor-force participation, and the rate at which workers find jobs measure different parts of the adjustment.

  • Announcements and hiring plans: Show what employers say they intend to do, not necessarily how many workers have already left.
  • Initial claims: Track people newly filing for unemployment benefits; they can lag an announcement and do not capture every separation.
  • Payrolls and unemployment: Help show changes in employment and joblessness across the economy, but can obscure differences by age, occupation, industry, and region.
  • Openings and job-finding: Can weaken before a wave of layoffs appears, making it harder for new entrants and displaced workers to find jobs.

In December 2025, Powell said AI was probably part of the weak-hiring story, but “not a big part” yet. He noted that companies cited AI in layoff and hiring decisions even though unemployment claims had not risen proportionately. Read the December press-conference transcript.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Other forces were also affecting hiring. In September 2025, Vice Chair Philip Jefferson said labor supply and labor demand had both slowed; he cited average payroll growth of 29,000 jobs per month over the preceding three months, while unemployment remained relatively low. That figure describes that specific three-month period, not a current or ongoing monthly rate. Jefferson’s September remarks.

So a company’s reference to AI is evidence that management sees it as relevant to a decision, not proof that AI alone caused the cut. Cost reduction, restructuring, post-pandemic adjustments, weaker demand, and changing business strategies may also matter. The available comments do not establish AI as the main cause of economy-wide weak hiring.

AI can replace tasks, augment workers, or do both

Exposure is better understood at the task level than as a prediction that entire occupations will vanish. Automation substitutes for human labor on a task; augmentation helps a worker do it faster or better. Many jobs combine tasks that AI can handle with work requiring judgment, accountability, interpersonal interaction, or context. A firm can also produce more with fewer workers per unit of output, so rising productivity does not guarantee that employment in a particular role will rise.

Potentially exposed work includes repetitive research, drafting, coding, data processing, customer service, and administrative tasks. Entry-level and less-experienced workers may be especially vulnerable if routine tasks that once helped them gain experience are automated. Jefferson said effects could vary across occupations and industries and cited research suggesting younger and less-experienced workers may face greater pressure. Jefferson’s November discussion of AI and the economy.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Replacement is not the only possibility. Federal Reserve Vice Chair for Supervision Michael Barr described ways AI might support workers in customer service, professional writing, and software engineering, including by giving less-experienced employees tools that raise productivity. He also acknowledged the potential for task and job displacement. Barr’s February 2025 remarks.

Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

Why the Fed cares about AI

The Fed’s monetary-policy goals are maximum employment and stable prices. AI could complicate both: weaker hiring or layoffs may reduce labor demand, while productivity gains could expand the economy’s capacity and lower costs. Investment in AI may also support growth. Effects on wages could differ between workers whose tasks are readily automated and those whose skills complement the technology.

These forces point in different directions. If employment weakens while productive capacity rises, the Fed must assess the combined effects on demand, wages, and inflation rather than respond to an AI headline alone. Powell has said the Fed would continue pursuing its existing mandate; it does not directly retrain displaced workers or decide which occupations survive. Worker transitions are more directly shaped by employers, educators, Congress, and the private sector. Powell’s testimony on the Fed’s role.

The long-run outcome is not settled

Past technological change has destroyed some jobs, altered others, and created new industries and occupations; productivity gains can raise output and incomes. That history is a reason not to assume that every automated task becomes a permanently lost job. It is not a guarantee that AI’s transition will be painless or follow the same pattern.

Free tools Windows power users keep installed

One-click scans. No signup required.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

In December 2025, Powell said previous technology waves eventually produced more work and higher productivity, while explicitly allowing that AI might be different. The central uncertainty remains whether new tasks and demand will emerge quickly enough to offset disruption, and how uneven the transition will be across workers and industries.

What to take from Powell’s remarks

  • Powell said the Fed was watching reported AI-related hiring cuts closely and that AI could affect job creation.
  • He did not say AI had already caused mass unemployment or predict a specific employment shock.
  • In October 2025, he said initial claims did not yet clearly reflect the reported effects; in December, he described AI as part of weak hiring, but not a major part at that time.
  • Company explanations are not enough to establish AI as the cause of aggregate labor-market weakness.
  • The near-term risk of displacement and the possibility of longer-run productivity and job creation can both be real; the eventual balance is unknown.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.