Federal Reserve Chairman Jerome Powell and Treasury Secretary Scott Bessent met with major U.S. bank CEOs this week to discuss the possible cyber risks raised by Anthropic's Mythos model, CNBC confirmed Friday.
The bank heads were already in Washington, D.C., for a Financial Services Forum board meeting when a special gathering was called on Tuesday to discuss Mythos, according to people familiar with the matter, who asked not to be named in order to share information about a confidential matter.
The CEOs had a dinner early in the week when they got the call to meet at the Treasury Department, one of the people said. JPMorgan's Jamie Dimon was the only major banking CEO who could not attend the meeting, they added.
The Federal Reserve declined to comment. Treasury and Anthropic did not respond to requests for comment.
Bloomberg and the Financial Times were first to report the special meeting.
Earlier this week, Anthropic rolled out the new artificial intelligence model, Claude Mythos Preview, in a limited capacity due to concerns that hackers could exploit its capabilities.
Banking giant JPMorgan Chase was among the initial launch partners for the cybersecurity initiative, known as Project Glasswing. Other partners include Apple, Google, Microsoft and Nvidia.
An Anthropic official told CNBC that it's been in "ongoing discussions" with the U.S. government, including the Cybersecurity and Infrastructure Security Agency and the Center for AI Standards and Innovation, about Claude Mythos Preview's cyber capabilities.
Late last month, cyber stocks slumped after Fortune discovered a draft blog post from Anthropic, which revealed the model's advanced cyber capabilities and possible risks.
"The dangers of getting this wrong are obvious, but if we get it right, there is a real opportunity to create a fundamentally more secure internet and world than we had before the advent of AI-powered cyber capabilities," CEO Dario Amodei wrote in a post to X, accompanying Glasswing's rollout.
Hackers have used Anthropic's models in the past to orchestrate AI-fueled attacks.
In November, the company disclosed that a Chinese group used Claude to automate a hack on government and corporate targets.
Anthropic is currently fighting the Department of Defense over its recent labeling of the AI lab as a supply chain risk to national security. This week, a federal appeals court denied the company's request to temporarily block the blacklisting.
