
Fed Chair Warsh Says Staff Are Modeling an AI Crash Scenario as Gensler Warns Bubble May Burst Before 2028
Former SEC Chair Gary Gensler warned that the AI infrastructure boom is already a bubble and could reprice before November 2028. Fed Chair Kevin Warsh said staff are modeling AI scenarios that include a crash-like downturn, while the Fed's latest Financial Stability Report found 87% of surveyed market contacts now view AI as a financial stability risk.
Gensler, now an MIT professor and former SEC chair, told CNN that the AI build-out is already a bubble and could cause a recession before the November 2028 election. He cited estimates that as much as $800 billion a year is being spent on AI infrastructure and compared the cycle to railroads, 1920s electricity and the 1990s internet. His comments came days before Anthropic was due to begin trading in what would be the largest IPO on record.
The Fed's Financial Stability Report, released Friday, said 87% of market contacts surveyed named AI as a risk to the financial system, up from 50% in the spring, making it the most cited risk. Contacts pointed to stretched valuations, earnings disappointment, cyberattacks and complex financing. Gensler also pointed to circular financing and off-balance-sheet deals, saying it is hard to see how much risk lenders are carrying.
Warsh, opening the Fed's Productivity and Jobs Task Force Conference in San Francisco, said staff are modeling three AI scenarios: broad gains, labor disruption, or a downturn in which expected returns fall and equity valuations decline, causing a slowdown and higher unemployment. He called the scenarios illustrative, not forecasts, and said productivity growth has not yet broken out structurally. The Fed's role, he said, is contingency planning, not advocacy.
The debate centers on who bears losses if AI spending slows. Gensler said major companies would likely survive, while debt-heavy neocloud firms renting GPU capacity could face bankruptcies. CoreWeave guided to a $1.1 billion quarterly interest bill as debt reached $47 billion, and AI power stocks fell after SoftBank's write-down. Musk said debt markets had shut off access for companies without substantial non-AI revenue, adding to concern about financing risks.
It matters because lenders, public-market investors and workers are exposed to a possible repricing of AI spending, with debt-heavy GPU-rental firms and power suppliers most vulnerable if financing conditions tighten or earnings disappoint.
Sources (1)
- [1]Original source(https://www.zerohedge.com/economics/gensler-warns-ai-bubble-will-burst-2028-warsh-reveals-fed-modeling-crash)