A spectre of mass service-sector job losses is shaping how the Federal Reserve under Chairman Kevin Warsh is beginning to think about artificial intelligence.
Warsh drew an explicit parallel at a San Francisco conference this week between AI’s potential economic disruption and the upheaval caused when China joined the World Trade Organisation in 2001.
That so-called China shock, as economists have labelled it, cost the United States an estimated one million manufacturing jobs and 2.4 million positions in total over the years that followed.
Warsh suggested white-collar and service workers could be next, as AI takes over legal, administrative, and other knowledge-based tasks at a fraction of the human cost.
“I don’t want to hold this too tightly,” Warsh said. “But if in some ways over the last generation or generation and a half, the China shock, for lack of a better word, lowered prices of goods in the United States, the way that I heard you talking about the philosophy of this new technology is it might do something for services.”
The implications for monetary policy are significant: cheaper services driven by AI could produce a sustained disinflationary effect, potentially pointing toward lower interest rates over the longer term.
Warsh convened the gathering as part of his AI task force, one of five advisory groups he has assembled since taking the Fed’s top job, drawing expertise from outside the central bank’s traditional ranks of PhD economists.
The move reflects his longstanding criticism of the Fed as an institution too insular to grapple with fast-moving economic change, and the San Francisco event saw him moderate three of the four sessions himself.
His task force includes venture capitalist Marc Andreessen, Stanford economist Chad Jones, and Xbox chief executive Asha Sharma, though Jones was barred from speaking during a blackout period ahead of Anthropic’s imminent IPO, and Sharma did not attend.
Andreessen, a longtime friend of Warsh’s who attended Stanford with the chairman’s wife, offered a notably measured take on AI-driven disruption despite his reputation as one of the technology’s most fervent backers.
“To the extent our economy is a free market economy, then maybe you could see the, let’s say, extremely rapid displacement scenarios people are talking about,” Andreessen said. “But we’ve mostly outlawed the kind of job transition that … both the utopians the doomers talk about.”
He argued that heavy regulation in sectors such as healthcare and education would effectively shield workers in those fields from being replaced by AI tools.
The conference also touched on a growing political backlash against the technology, after voters in the 3 November midterms approved a series of state-level measures designed to slow the construction of data centres.
Gina Raimondo, a former Commerce Secretary under President Joe Biden and ex-governor of Rhode Island, used the platform to push back against that trend, calling it “a very troubling sign.”
Stopping AI expansion, she added, “would be very bad for America” and ultimately harmful to the very voters who backed those restrictions.
The urgency of the Fed’s deliberations is underscored by the pace of industry growth: AI lab Anthropic is set to price its initial public offering on Wednesday, with the company targeting a valuation of $1.4 trillion and a fundraise of as much as $64 billion.
Meanwhile, surging investment across the AI sector is already influencing near-term rate thinking, with the Fed reportedly weighing a December rate rise to cool the inflationary pressure building from technology-driven spending.
Warsh offered no definitive conclusions on Monday, and any policy shift remains several steps away, with task force members expected to deliver a formal report to the chairman within the next few months before the full Fed board debates its implications.

