If AI fails to generate the expected returns, what could collapse isn’t just tech stocks—but a key growth engine of the U.S. economy. On November 10, Michelle Bowman, Vice Chair for Supervision at the Federal Reserve, issued a warning: if AI investments underperform, U.S. equities could face a sharp correction, and this adjustment might spill over into the real economy. She noted that falling stock prices would erode the wealth effect for American consumers, dampening consumer spending. Rising financing costs would strain real estate, small business investment, and the rapidly expanding construction of data centers. More notably, Bowman revealed the hidden financing risks behind the AI boom: AI-related debt is projected to exceed $800 billion this year, while other potential payment obligations—such as long-term leases, contractual commitments, and residual value guarantees—could exceed three times that amount. AI investment is no longer just tech giants betting their own cash flows. Corporate bonds, private credit, project financing, and increasingly complex off-balance-sheet arrangements are drawing more financial institutions into this capital expenditure race. The most striking takeaway from her remarks is this: AI is transitioning from a driver of U.S. economic growth to a potential source of systemic risk that the Fed must take seriously. Over the past few years, AI investment has fueled U.S. capital spending, tech stock valuations, and overall economic growth. But as more growth becomes predicated on the assumption that future returns will materialize, any doubt about that assumption could trigger a chain reaction—spreading from stock markets to credit markets and capital expenditures. Technological revolutions can be real—and investment bubbles can be real too. The two are not mutually exclusive. Yet the harshest truth of financial markets is this: they may not have the patience to wait until technological revolutions actually deliver their promised returns.