Fed's Bowman Sets Out 'Sound Practices' for Banks Using Artificial Intelligence

The Federal Reserve's Vice Chair for Supervision, Michelle Bowman, used opening remarks on 7 July to lay out what she described as sound practices for banks adopting artificial intelligence. Her comments reflect a supervisory balancing act: encouraging banks to harness AI to improve efficiency, fraud detection and customer service, while ensuring the technology is deployed with appropriate governance, risk management and human oversight. Bowman has generally favored a measured, principles-based approach to supervision that avoids stifling innovation, and her remarks emphasized that existing risk-management expectations apply to AI just as they do to other tools, rather than proposing an entirely new rulebook. The guidance touches on issues such as model governance, data quality, third-party and vendor risk, explainability, and the need for banks to understand and control the systems they rely on. The intervention comes as regulators worldwide grapple with how rapidly advancing AI models are reshaping financial services, and as supervisors weigh both the opportunities and the operational, cyber and consumer-protection risks the technology introduces across the banking sector.