US Treasury and Federal Reserve Convene Bank CEOs Over AI Cyber Risks

Top U.S. financial regulators and the CEOs of the nation's largest banks held a special meeting this week to address emerging cybersecurity threats posed by artificial intelligence. The gathering, which included Federal Reserve Chairman Jerome Powell and Treasury Secretary Scott Bessent, focused specifically on the risks associated with Anthropic's latest AI model, Mythos. The meeting took place on the sidelines of a Financial Services Forum board meeting in Washington, D.C. The urgency of the discussion follows the limited release of "Claude Mythos Preview," a model that Anthropic has deployed cautiously due to concerns that its capabilities could be weaponized by malicious actors. To mitigate these risks, Anthropic has launched "Project Glasswing," a cybersecurity initiative designed to safeguard critical systems. JPMorgan Chase is a primary launch partner in this effort, alongside other technology giants including Microsoft, Google, Nvidia, and Apple. Anthropic has confirmed it is maintaining ongoing dialogues with the Cybersecurity and Infrastructure Security Agency (CISA) and the Center for AI Standards and Innovation to monitor the model's cyber capabilities and ensure systemic stability. While no specific vulnerabilities have been disclosed, the high-level coordination between the Treasury, the Fed, and the banking sector signals that AI-driven cyber threats are now viewed as a systemic risk to the U.S. financial system.