US Regulators and Wall Street Leaders Convene Over Systemic AI Cyber Risks

Top U.S. financial regulators and the leadership of the nation's most systemically important banks held an urgent summit on Tuesday to evaluate the cybersecurity implications of a new artificial intelligence model. Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell convened with the chief executives of Citigroup, Morgan Stanley, Bank of America, Wells Fargo, and Goldman Sachs. The meeting centered on "Mythos," an advanced AI developed by Anthropic. Unlike general-purpose AI, Mythos is specifically engineered for cybersecurity tasks, possessing the capability to discover software bugs and construct sophisticated exploits. Regulators are now treating AI not just as a technological challenge, but as a potential catalyst for systemic financial instability. Anthropic has limited the release of Mythos to a select group of technology and financial institutions under a program called "Project Glasswing." The company has previously coordinated with U.S. officials regarding the model's offensive and defensive capabilities. Simultaneously, Anthropic is challenging a Pentagon designation that labels the firm a supply-chain risk in a separate legal dispute. The concerns extend beyond traditional banking into the decentralized finance (DeFi) sector. Experts warn that Mythos' ability to identify zero-day vulnerabilities in real-time and at low cost could jeopardize the underlying infrastructure of blockchain-based finance. This shift in regulatory framing suggests a new era of oversight where AI-driven cyber threats are viewed as macro-prudential risks.