Regulators are asking banks to account for their use of AI, citing 'cyber risk' concerns

U.S. banking regulators are pressing banks to explain how they use artificial intelligence, warning of “cyber risk” and asking whether AI systems can be quickly shut down if they behave outside their intended scope. The Federal Reserve, the Office of the Comptroller of the Currency, and the FDIC are among the agencies seeking details, including whether banks deploy AI agents to respond to cybersecurity threats and what safeguards exist if an agent causes problems. Regulators want protections such as human review and sign-off, limits that keep systems within authorized boundaries, and an emergency disconnect. Officials also worry about cross-border AI tools entering the U.S. banking ecosystem and rapid AI-driven market disruption, with European officials similarly urging better recovery plans.