Austan Goolsbee, president of the Chicago Federal Reserve Bank, told Yahoo Finance at the Semafor World Economy conference that he would "take the under" on the idea that artificial intelligence will push steady-state unemployment to 15 percent within a decade. It is a bold bet, and one that sits uneasily beside a wave of tech-sector layoffs already washing over American workers in 2026.
More than 52,000 U.S. tech employees lost their jobs in just the first three months of this year, according to outplacement firm Challenger, Gray & Christmas. March alone saw 18,720 tech-industry job cuts announced, a 40 percent year-over-year increase. Oracle reportedly laid off up to 30,000 workers across the United States, Mexico, and other countries on April 1. Block slashed its workforce by 40 percent. Amazon and Meta carried out their own massive reductions.
A good portion of these layoffs, Yahoo Finance reported, were tied directly to AI adoption. That is the tension at the heart of Goolsbee's reassurance: the technology he says will "make us rich" is already making tens of thousands of workers redundant, right now, not in some distant hypothetical.
Goolsbee did not dismiss the disruption. He acknowledged the short-run pain. But he framed it as a familiar pattern in the long arc of technological change, one that ultimately creates more jobs than it destroys.
"There have been many changes to the nature of many jobs over a lot of decades. We have a pretty good sense that there can be short-run disruptions. And then in the long run, most of these big technologies are job creators on net, not job destroyers."
That is the standard economist's answer, and it has a long track record of being broadly correct, from the loom to the assembly line to the personal computer. The question is whether AI breaks the pattern or merely accelerates it.
Goolsbee was responding, at least in part, to a scenario laid out by Citrini Research in a February report. That analysis described a potential "doom loop" in which AI-driven productivity gains allow companies to replace high-income knowledge workers with autonomous AI agents, sending U.S. unemployment surging to 10 percent by 2028. It is the kind of forecast that sounds extreme until you look at the quarterly layoff data and notice the trend line.
The Fed president was unpersuaded. As he put it: "I think if AI has the improvement to productivity that everyone's describing, it's going to make us rich... So if you think that if we come back in 10 years, that the steady-state unemployment rate is going to be 15% or something, I'll take the under on that."
The labor market, for its part, has shown resilience in recent months. Weekly jobless claims recently dipped to 213,000, a sign that broad-based employment still holds even as specific sectors shed workers at speed.
Goolsbee's long-run optimism may prove right. But the short run is where real people live, pay mortgages, and feed families. And the short run is ugly.
Block's CFO, Amrita Ahuja, offered a window into the corporate mindset driving these cuts when she spoke to Yahoo Finance. She urged fellow executives to experiment with AI tools firsthand:
"You often don't get the 'aha moment' until you realize you've automated a piece of your work, something that might have taken days before can now take hours or less."
That is an honest description of what AI can do. It is also a frank admission that companies are discovering they can get the same output with far fewer employees. When a CFO tells her peers to "be curious" about the tools, she is not describing a future possibility. She is describing what already happened inside her own company, the one that cut 40 percent of its workforce.
Oracle's reported layoff of up to 30,000 workers on a single day, April 1, stands as perhaps the starkest example. The company, still guided by billionaire founder Larry Ellison, executed one of the largest single-day workforce reductions in recent memory. The cuts spanned multiple countries.
The broader picture from Challenger, Gray & Christmas puts those individual company actions in context. The 52,000-plus tech layoffs in the first quarter of 2026 are significant, though the firm noted that the year-to-date total still trails the 102,391 tech-sector job cuts recorded in 2023. That comparison offers some comfort, but 2023 was itself a historically brutal year for the industry.
A recent Boston Consulting Group analysis projected that AI could reshape more than half of U.S. jobs within three years, a forecast that makes the current wave of cuts look less like an anomaly and more like the opening act.
There is a familiar rhythm to how Washington and Wall Street process technological disruption. First, the boosters promise transformation. Then the layoffs arrive. Then an official steps forward to say the disruption is temporary and the long run will be fine. Then the people who lost their jobs are left to figure out what "the long run" means for their family's next rent check.
Goolsbee's argument, that technology is historically a net job creator, is not wrong on the merits. The industrial revolution, electrification, and the internet all eventually generated more employment than they destroyed. But each of those transitions also produced years of dislocation, political upheaval, and real suffering for workers caught in the transition.
The Fed president said plainly: "I think the nature of jobs are going to change." That is almost certainly true. The harder question, the one Goolsbee did not answer in detail, is what specific evidence or data supports his confidence that the transition will be manageable rather than catastrophic. He offered historical analogy and general optimism. He did not offer a policy roadmap.
That matters because the Federal Reserve's own posture on interest rates, inflation, and employment directly shapes how quickly or slowly the economy can absorb displaced workers. With prediction markets already pricing in elevated recession odds and macroeconomic headwinds building, the margin for error is thin.
The Fed's leadership itself is in flux. President Trump has nominated Kevin Warsh to lead the Federal Reserve, a move that could reshape the central bank's approach to employment, inflation, and the broader economic disruption that AI is accelerating. How the next Fed chair weighs the competing pressures of technological displacement and monetary policy will matter enormously for the workers Goolsbee says will be fine in the long run.
The conservative case for technological progress is strong and well-established. Free markets allocate capital to its most productive uses. Innovation drives growth. Government should not stand in the way of efficiency gains that make the whole economy wealthier.
But conservatives also understand that institutions owe honesty to the people they serve. And the honest truth is that AI is not replacing burger flippers and warehouse workers first. It is replacing analysts, coders, project managers, and mid-level professionals, the very knowledge workers who were told for decades that education and white-collar skills would insulate them from automation.
When Block's CFO celebrates the "aha moment" of realizing that days of work can now be done in hours, she is celebrating a genuine productivity gain. She is also describing the elimination of someone's livelihood. Both things are true at once. The question is whether the people making these decisions, and the officials reassuring the public, are being honest about the costs, or whether they are papering over real pain with long-run abstractions.
Goolsbee's bet against 15 percent unemployment may well pay off. But the workers at Oracle, Block, Amazon, and Meta who already lost their jobs this year are not living in a 10-year forecast. They are living in the short-run disruption that even the Fed president concedes is real.
Broader economic pressures only sharpen the stakes. Stagflation concerns have gripped U.S. markets as energy costs climb and growth slows, conditions that make it harder for displaced workers to find new footing quickly.
Goolsbee is probably right that AI will not produce 15 percent unemployment a decade from now. History is on his side, broadly speaking. But "broadly speaking" is cold comfort to the 52,000 tech workers who already got the news this quarter.
The Fed's job is not just to forecast the long run. It is to manage the economy that real Americans live in today. And today, companies are cutting workers by the tens of thousands, citing AI as the reason, while officials assure the public that it will all work out eventually.
Eventually is not a paycheck. And optimism is not a plan.