Federal Reserve Vice Chair for Supervision Michelle W. Bowman said Nov. 10 that disappointing artificial intelligence returns could trigger a large correction in equity valuations, and that complex AI financing structures make counterparty risk difficult to assess.
She described a correction as a risk if investment returns fall short, not a prediction. Lower share prices could curb consumer spending, while a sharp rise in private long-term interest rates could put pressure on housing and small-business investment, she said.
Bowman said borrowing connected to AI is poised to exceed $800 billion in 2026, including corporate bonds, private credit and project finance. Potential payments tied to long-term leases and contracts, together with residual-value guarantees, could be more than three times that debt.
Financing ties run in both directions. Hyperscalers and chip designers have invested in smaller businesses that buy their products, which Bowman said gives the system the appearance of circular financing. She also said smaller, riskier companies can use off-balance-sheet obligations taken on by larger firms as collateral when securing funding for data centers and accelerated-computing equipment. The connected arrangements make counterparty risk — whether firms in a financial relationship can meet their obligations — difficult to assess, she said.
The financing also touches mainstream institutions. Bowman identified the largest U.S. bank holding companies, life insurers, pension funds, private-credit firms and mutual funds among those with some exposure. Most of those organizations, she said, are well-capitalized and have stable funding.
The commercial case for AI has yet to catch up with the spending, in Bowman's view: revenue remains insufficient to cover the combined cost of operating existing models and data centers while developing new ones. If returns lag costs, lower valuations and more expensive borrowing could prompt companies to scale back investment; tighter credit could also slow AI infrastructure expansion and weigh on economic activity and inflation.
Bowman said AI-related capital spending and consumer outlays have supported U.S. growth, comparing the pace of real business investment with the 1990s technology boom. The sustainability of AI investment could shape her policy views. EconoTimes also covered Bowman's core inflation outlook from the Nov. 10 remarks. Before the Fed's December meeting, officials are due to receive the October personal consumption expenditures (PCE) inflation report and November employment report.
