Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell convened an urgent meeting with Wall Street’s top executives on Tuesday at Treasury headquarters in Washington, addressing heightened cybersecurity risks posed by Anthropic PBC’s latest AI model, Mythos. The previously unreported gathering, arranged on short notice, underscores regulators’ growing concern over a potential wave of sophisticated cyber attacks targeting the financial sector.
The meeting brought together CEOs from systemically important banks, including Citigroup Inc.’s Jane Fraser, Morgan Stanley’s Ted Pick, Bank of America Corp.’s Brian Moynihan, Wells Fargo & Co.’s Charlie Scharf, and Goldman Sachs Group Inc.’s David Solomon. JPMorgan Chase & Co.’s Jamie Dimon was unable to attend. Regulators emphasized the need for banks to fortify their defenses against vulnerabilities that could be exploited by AI systems like Mythos, which Anthropic has described as capable of identifying and targeting flaws in major operating systems and web browsers when directed by a user.
Anthropic has taken a cautious approach to Mythos’ rollout, limiting initial access to a select group of major technology and finance firms under “Project Glasswing.” Participants include Amazon.com Inc., Apple Inc., and JPMorgan Chase & Co., with the initiative focused on securing critical systems before broader release or the emergence of similar AI models. The company has also engaged with US officials prior to the launch, discussing the model’s offensive and defensive cyber capabilities.
Beyond cybersecurity, Anthropic faces additional pressure from the US government. The Pentagon has classified the company as a supply-chain risk, a designation Anthropic has contested in court. Earlier this week, a federal appeals court declined to temporarily halt the Pentagon’s label, leaving the legal battle unresolved.
Regulators view AI-driven cyber threats as among the most pressing risks to the financial industry, particularly for institutions deemed critical to global stability. The urgency of Tuesday’s meeting reflects a broader push to preemptively address vulnerabilities as AI technology advances rapidly.
Anthropic, the Treasury, and the Federal Reserve offered no immediate comments on the discussions, while spokespeople for the summoned banks also declined to speak on the matter. The focus now shifts to how these institutions will adapt their cybersecurity frameworks in response to Mythos’ capabilities, with Project Glasswing serving as an early testing ground for defenses against next-generation threats.
Information for this story was found via the sources and companies mentioned. The author has no securities or affiliations related to the organizations discussed. Not a recommendation to buy or sell. Always do additional research and consult a professional before purchasing a security. The author holds no licenses.
