US bank regulators are questioning lenders about rogue AI agents and Chinese-built models, and British supervisors want “bare metal” fallbacks, Reuters reports

Reuters reported on 13 October that bank supervisors in the United States and Europe are stepping up their scrutiny of advanced AI, citing six people familiar with the matter. According to three of them, the Federal Reserve, the Office of the Comptroller of the Currency and the FDIC are questioning banks about whether they use Chinese-developed technology, whether they use AI agents in cyber defence, and whether they can control autonomous systems: limits, monitoring, approvals, auditability and the ability to shut them down. Reuters says the regulators are still gathering information rather than issuing rules.

In Britain, two of the sources say, supervisors are urging the largest banks to look at “bare metal” recovery plans: systems entirely separate from a bank's main technology that can take over if those systems are compromised. Reuters attributes this to supervisors in Britain; the Bank of England declined to comment.

The European Central Bank set lenders a deadline of the end of October, in a letter sent in July, to produce plans against AI-enabled cyber threats. The OCC, the Fed and the FDIC declined to comment.

Michael Woodbridge, a cybersecurity partner at Grant Thornton, told Reuters: “The adoption of AI is often outpacing the ability to put the controls in place.” No supervisor's own document was visible from here, so every point is reported.

• Reported The Fed, OCC and FDIC are questioning banks about Chinese-developed technology, AI agents in cyber defence and control of autonomous systems, three people told Reuters. Reuters, via Yahoo Finance
• Reported Supervisors in Britain are urging the largest banks to examine “bare metal” recovery plans, two sources said. Reuters, via Yahoo Finance
• Reported The ECB gave lenders an end-October deadline, set in July, for plans against AI-enabled cyber threats. Reuters, via Yahoo Finance