What happened
Federal Reserve Chairman Jerome Powell and Treasury Secretary Scott Bessent convened major U.S. bank CEOs on April 7, 2026, at the Treasury Department to address cyber risks from Anthropic's Claude Mythos model. The meeting included leaders from Bank of America, Citigroup, Goldman Sachs, Morgan Stanley, and Wells Fargo. Anthropic released Claude Mythos Preview with limited access due to its capability to identify and exploit vulnerabilities across major operating systems and web browsers. This follows Anthropic's launch of Project Glasswing, a defensive collaboration with partners including JPMorgan Chase, to use Mythos for cybersecurity.
Why it matters
The high-level meeting signals AI's advanced capabilities pose a systemic cyber threat to the U.S. financial system, elevating risk for security architects and compliance officers. This direct engagement from senior officials underscores the immediate need for financial institutions to integrate AI-specific threat models into their defence strategies. The limited release of Anthropic's Mythos, despite its Project Glasswing partners, highlights recognised dual-use potential, impacting procurement teams evaluating frontier AI models. This follows a U.S. appeals court upholding the Pentagon's blacklisting of Anthropic as a supply-chain risk.
Subscribe for Weekly Updates
Stay ahead with our weekly AI and tech briefings, delivered every Tuesday.
