Macro · San Francisco
Fed chair Warsh outlines AI scenarios, from faster growth to job losses and stock bust
Fed chair Kevin Warsh told a San Francisco conference that staff are modelling AI contingencies, ranging from faster growth to a slowdown if the technology's returns disappoint.
AI text based on reports by The New York Times, europa press, Bloomberg Business and other publishers. Published . How it is made
For membersThe first 30 seconds for everyone. Synthetic voice (AI).
In brief
Kevin Warsh described upside growth, labour-market disruption and an economic slowdown if AI returns disappoint, calling them illustrative rather than forecasts.
The Fed's financial stability report, released on Friday, 13 November, shows 87% of respondents citing AI as a risk, up from 50% six months earlier.
The Fed estimates the largest US banks hold nearly $1 trillion in exposure, while companies borrowed more than $680 billion in the first 10 months of 2026.
Several colleagues on the rate-setting committee back more hikes, including at the December meeting, after inflation accelerated in October.
- 13 NovFed releases its semiannual financial stability report, citing AI as the most frequently cited risk
- 16 NovWarsh opens a one-day AI and productivity conference at the San Francisco Fed
- December 2026Next Fed rate-setting meeting, where some colleagues back further rate hikes
Scenarios drawn up at the Fed
Kevin Warsh said on Monday, 16 November, that Fed staff have started modelling contingencies for how artificial intelligence could affect the economy. He spoke at the opening of a one-day conference on AI and productivity at the Federal Reserve Bank of San Francisco, and he stressed that the scenarios are illustrative rather than forecasts. Warsh calls the technology "SI," or super-intelligence, the term preferred by President Donald Trump. The conference centred on a task force he set up to study how productivity gains could affect the labour market, one of five groups he has created on issues central to monetary policy.
He framed the work as preparation rather than prediction, saying:
“Our job is to prepare for whatever comes next.”
Three possible outcomes
In the most favourable case, Warsh said, AI could lift productivity and spread the gains across workers and owners of capital, with growth staying persistently above its long-run average and unemployment holding near full employment. A gloomier case involves substantial labour-market disruption, in which jobs are displaced rather than complemented by the technology, which he said would bring higher growth but rising unemployment. A third risk concerns returns: if AI investments disappoint after heavy capital spending and stock market gains, equity valuations could fall, slowing the economy and pushing unemployment up more sharply. Warsh added that the economy would eventually recover, but without the productivity gains. He also stressed the human role in these systems, saying:
“Machines that can amass or even exceed human knowledge are still machines.”
Financial stability warnings
The Fed's semiannual financial stability report, released on Friday, 13 November, lists AI as the most frequently cited risk among surveyed market contacts. Some 87% of respondents now name AI as a risk to financial stability, up from 50% six months earlier, and a broad sell-off in stocks and other risky assets ranks second at 52%. The Fed estimates the largest US banks hold nearly $1 trillion of financial exposure to technology companies, data centre developers and other lenders tied to the AI build-out. Companies borrowed more than $680 billion for data centres and other advanced computing infrastructure in the first 10 months of 2026, roughly five times the total raised across 2023. Other obligations have grown faster still, with seven large technology firms, among them Nvidia, Microsoft and Oracle, committing over $3 trillion through future leases, contracts and guarantees.
- AI risk, six months earlier
- 50 %
- AI risk, latest report
- 87 %
- Broad stock sell-off, latest report
- 52 %
Data from the article text
Rates and inflation
Warsh did not comment on the Fed's next interest rate steps, nor on recent inflation data showing that prices accelerated in October. That data led several colleagues on the rate-setting committee to back further rate hikes, including at the December meeting. Warsh has long argued that AI could raise efficiency enough to let the economy grow without worsening inflation. Other officials increasingly say the buildout of AI infrastructure is fanning prices, because demand has surged for computer chips, electronic components, electricity and copper. Those cost pressures have pushed up the price of smartphones, computers and other consumer electronics, and smartphone prices rose in October at the fastest pace on record, according to government data.
Ask about this text
Sources and context
- The New York TimesFed Chairman Highlights Economic Uncertainty of A.I. Boom
- europa pressWarsh (Fed) admits that the Fed is preparing responses if AI destroys...Warsh (Fed) admite que la Fed prepara respuestas si la IA destruye... · original in Spanish
- Bloomberg BusinessFed's Warsh Says Staff Are Gaming Out Contingencies on AI Impact
- AP NEWSWarsh grapples with AI's potential economic impact, steers clear of its inflation risk
- Kevin Warsh
- Chair of the US Federal Reserve who outlined AI economic scenarios at the San Francisco Fed conference
- Donald Trump
- US President whose preferred term for advanced AI, super-intelligence, Warsh uses
- Places
- San Francisco, United States
Pollar is free and carries no ads in the text. Its members keep it going.
Support Pollar