Fed Governor Barr warns AI could destabilize financial system, fuel rapid bank runs
Key Takeaways
- •Fed Governor Michael Barr warned at a conference in Milan that AI's rapid adoption in banking and markets could threaten the stability of the US financial system.
- •Barr said AI tools could push traders using similar models to buy or sell simultaneously, enable automated programs to manipulate prices, and let AI agents withdraw deposits faster than banks can react, echoing classic bank-run dynamics.
- •He flagged circular financing, where AI companies raise funds from investors who then sell them services, and rising borrowing among large tech firms, equipment suppliers, and smaller AI startups as links that could spread and amplify problems if AI growth disappoints.
- •Barr cautioned that AI could amplify bias in consumer lending, assist hackers attacking financial systems, and noted that developers struggle to keep systems accurate, lawful, explainable, and consistent.
- •Roughly 87% of respondents to the Federal Reserve's November 13 Financial Stability Report survey identified AI as a major risk to the US financial system.

Federal Reserve Governor Michael Barr has cautioned that the rapid spread of artificial intelligence across banking and markets could threaten the stability of the US financial system. In a speech in Milan, he said AI could prompt traders to buy or sell at the same time, allow automated trading programs to manipulate prices, and cause depositors to rush to withdraw money from banks far more quickly than institutions can manage.
Barr delivered the warning at the Seventh Annual Association for Computing Machinery International Conference in Milan, Italy. As a sitting governor of the US central bank — the body that sets and enforces rules for American banks — his remarks carry weight beyond a routine policy speech. He did not discuss interest rates or the outlook for the economy, focusing instead on the growing use of AI in banking, trading, and everyday financial services.
Circular financing risks in the AI boom
The governor flagged the dangers of circular financing, in which AI companies raise money from investors and then spend it on services provided by those same investors. Such arrangements can make the companies involved appear more successful than they actually are.
Barr also raised concerns about how much companies are borrowing to build AI systems, pointing to deepening financial ties among large technology companies, equipment suppliers, and smaller AI firms. If AI companies grow more slowly or make less money than expected, he warned, those financial links could cause problems to spread across the system and become worse.
How an AI agent could trigger a bank run
Barr also warned about AI tools that manage money on behalf of customers. He noted that banks may generally find it hard to verify whether a customer has given an AI tool permission to move funds, and that even with permission, such tools can remain risky. They could, he said, move money out of bank accounts so quickly that banks would have little time to react, potentially causing large withdrawals at once that outpace the safety measures banks currently rely on. The scenario reflects one of banking's oldest fragilities: the speed at which depositors can act in unison.
He additionally cautioned that hackers could use AI to attack financial systems, and that AI tools that stop following their rules could find and take advantage of weaknesses in those systems.
Herding, manipulation, and biased lending
Financial markets could face problems if too many traders use similar AI tools, Barr warned. Such tools could push traders to buy or sell at the same time, producing sharp price changes. He also said AI trading programs could learn to manipulate prices on their own.
"We need to reduce the risk that AI amplifies bias in consumer lending," Barr stated during his speech. He also warned of "collusion and market manipulation, significant market swings, and even trading strategies that create systemic risk."
Barr added that even AI developers are struggling to keep their systems under control and to ensure they deliver accurate results and follow the law. They also face challenges in explaining how these systems work and in making sure they produce consistent results.
AI a major risk for 87% of survey respondents
Barr's speech came shortly after the Federal Reserve released its Financial Stability Report on November 13. According to BigGo Finance, about 87% of those surveyed said AI was a major risk to the US financial system. Respondents cited concerns over cyberattacks, complex financial deals involving AI companies, and the possibility that slower AI profit growth could hurt the wider stock market. The breadth of that response underscores how far AI-related concerns now reach across the financial system.
Barr spoke just hours before Fed Chair Kevin Warsh was scheduled to open a separate conference hosted by the San Francisco Fed, where AI developers, business leaders, and researchers were expected to discuss how AI could affect jobs, productivity, and the economy — placing the central bank's engagement with AI on display across two separate events in a single day.
The warnings highlight growing concerns that AI is spreading through the financial industry faster than regulators expected. Regulators, for their part, continue to work on ways to monitor the technology and manage its risks. For banks, technology firms, and market participants, the near-term signals to watch are the outcomes of the San Francisco Fed conference and any follow-up measures from regulators.