The Federal Reserve's latest Financial Stability Report, released on Friday 13 November, put artificial intelligence at the top of its list of near-term risks to the US financial system — a first, and by a wide margin.
In the Fed's autumn survey of market contacts, 87% named AI as a salient risk, up from 50% in the spring of 2026 and roughly 30% in the autumn of 2025. The report's authors describe that as the fastest single-cycle escalation recorded in the near-term risk category.
The concern is not that AI models will misbehave, but that the money surrounding them will. Respondents pointed to stretched equity valuations in AI-related stocks and to fast-growing corporate debt tied to data-centre construction, alongside persistent uncertainty about whether the huge capital expenditure will generate returns on the timeline investors expect. The report flags a possible sharp correction in AI-related equities as the mechanism to watch.
The timing is pointed. The report lands in the same week that Anthropic set the price range for what would be the largest listing in history — $70 to $78 a share, valuing the company at up to $1.4 trillion when it starts trading on Nasdaq on 19 November — and one week after a broad selloff in technology shares. Nvidia, the company at the centre of the AI build-out, is meanwhile without its long-serving chief executive and under a Department of Justice antitrust probe.
For anyone who tracks AI as a technology story, the Fed report is a useful corrective. It is a reminder that the most direct risk from the AI boom is not the model, but the balance sheet behind it — and that the people who price risk for a living are now watching that balance sheet more closely than almost anything else.

