AI Agents Could Pose New Risks to Banks: US, European Regulators Step Up Scrutiny of AI Safeguards, Chinese Models and Cybersecurity: Report
Financial regulators in the U.S. and Europe are intensifying their examination of AI, focusing on whether cyberattacks, autonomous systems, or inadequate controls could destabilize banks and markets.
In the U.S., the Federal Reserve, the Office of the Comptroller of the Currency (OCC), and the Federal Deposit Insurance Corporation (FDIC) are questioning lenders about advanced AI systems and related safeguards, Reuters reported, citing people with knowledge of the matter.
The inquiries include whether banks use Chinese-developed models, employ autonomous AI agents for cyber defense, and can intervene when systems increasingly perform tasks with limited human direction.
European supervisors, including the Bank of England and the European Central Bank, are pressing banks to improve recovery plans in case AI-related disruptions affect financial markets.
Regulators are reviewing system limits, monitoring procedures, approval requirements, auditability, human oversight and protections designed to prevent AI agents from exceeding their authority.
The inquiries involve banks of different sizes, the report added. Supervisors are still collecting information and have not established prescriptive requirements for lenders.
The regulatory push follows growing concern among financial executives that AI adoption could create new channels for cyberattacks and operational failures.
Federal Reserve, OCC, and FDIC did not immediately respond to Benzinga’s request for comments.
AI Cyber Risks Raise Bank Security Concerns
Earlier this month, CrowdStrike Holdings Inc. (NASDAQ: CRWD) said a suspected 26-year-old man in China’s Guangdong province may be linked to a cyber campaign targeting at least nine South Korean banks using AI coding tools. The cybersecurity firm assessed with moderate confidence that the individual was Chinese-speaking and financially motivated but did not attribute the attacks to a named threat actor. Shinhan Bank reported compromised data involving about 25,000 customers, while KB Kookmin Bank reported a leak affecting 119 customers.
In August, JPMorgan Chase & Co. (NYSE:JPM) CEO Jamie Dimon reportedly urged corporate leaders to join a U.S.-focused group addressing AI-related risks. The initiative, the Alliance for Critical Infrastructure, backed by JPMorgan, is aimed at improving security across critical systems as companies rapidly adopt AI.
Earlier this month, Dimon also warned that AI cyber risks were real while JPMorgan continued financing the sector. He highlighted Anthropic’s Mythos model and concerns that advanced systems could take unintended actions online.
Disclaimer: This content was partially produced with the help of AI tools and was reviewed and published by Benzinga editors.
Image via Shutterstock