Fed plans new rules to safely support AI use in banking amid rapid tech advances

The Federal Reserve is developing new supervisory guidelines to support the safe adoption of artificial intelligence (AI) in banking, emphasizing the need to balance innovation with security. Vice Chair Michelle Bowman highlighted the rapid advancement of AI capabilities, such as Anthropic's Mythos model, which can detect cyber vulnerabilities but also pose risks if misused. Regulators aim to update their frameworks, coordinate across agencies, and engage with the industry to manage emerging AI risks effectively. A report on sound AI practices in finance is expected in the third quarter, reflecting the broader impact of AI beyond banking.