Fed’s Bowman Says AI Is Blessing and Threat to Bank Security

Federal Reserve Vice Chair for Supervision Michelle Bowman addressed the potential benefits and risks of artificial intelligence for banks at a recent workshop. Bowman highlighted the growing threat AI poses to banks, given the increased availability and use of AI by malicious actors. She pointed out that AI can accelerate vulnerability identification, create sophisticated social engineering campaigns, lower the barrier to entry for cybercriminals, and adapt attacks in real time.

Bowman emphasized the importance of strong cyber hygiene, including up-to-date asset inventories, phishing-resistant multifactor authentication, rigorous identity and access controls, and robust vulnerability identification and patch management programs. She noted that AI is becoming a critical part of these security measures, acting both as a defensive tool and an evolving risk.

Bowman acknowledged that preparing for cyber threats can be challenging for community banks and highlighted the Federal Reserve's approach to IT examinations, which considers each bank's risk profile and emerging threats. She encouraged community banks to provide feedback on how the Federal Reserve can offer more clarity on its expectations.

The International Monetary Fund recently found that AI does not need to create new types of cyberattacks to be a threat, as it can accelerate vulnerability discovery and exploitation across shared technologies, turning isolated incidents into correlated disruptions affecting multiple institutions simultaneously. Companies must consider the level of autonomy granted to AI systems, the resources they can access, and the containment measures in place when things go wrong.