Federal Reserve Chair Jerome Powell and Treasury Secretary Scott Bessent Convene Urgent Meeting with Major U.S. Bank CEOs Over Anthropic’s AI Cyber Threat
April 10, 2026 – Washington, D.C. — In an unprecedented move underscoring the cybersecurity concerns surrounding artificial intelligence, Federal Reserve Chairman Jerome Powell and Treasury Secretary Scott Bessent held a high-level meeting this week with the chief executives of major U.S. banks to discuss the cyber risks posed by Anthropic’s latest AI model, Mythos.
The meeting, which took place on Tuesday at the Treasury Department, was convened amid growing fears within the financial sector about vulnerabilities that could be exploited by hackers leveraging the advanced capabilities of Anthropic’s Mythos model. The CEOs were in Washington, D.C., attending the Financial Services Forum board meeting when the special gathering was called, signaling the urgency and significance of the issue.
Attendees and Industry Impact
Key banking leaders were in attendance, including Brian Moynihan of Bank of America, Jane Fraser of Citigroup, David Solomon of Goldman Sachs, Ted Pick of Morgan Stanley, and Charlie Scharf of Wells Fargo. Notably, JPMorgan Chase CEO Jamie Dimon was the only major banking executive unable to attend.
The confidential meeting brought together the financial industry’s top executives and the two most powerful U.S. monetary regulatory figures, a clear indication that the Trump administration is prioritizing AI-related cyber threats as critical to the stability of the nation’s financial infrastructure.
Anthropic’s Mythos Model and Cybersecurity Concerns
Anthropic, a leading AI lab, rolled out the Mythos Preview model in a limited capacity earlier this week due to apprehensions that its capabilities could be weaponized by cyber attackers. The company has been working closely with government agencies, including the Cybersecurity and Infrastructure Security Agency (CISA) and the Center for AI Standards and Innovation, to brief senior officials on both the offensive and defensive cyber applications of Mythos.
“The dangers of getting this wrong are obvious,” explained Anthropic CEO Dario Amodei in a recent statement on X, formerly Twitter, “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.”
The development of Mythos is also tied to Project Glasswing, a cybersecurity initiative with major industry players such as Apple, Google, Microsoft, Nvidia, and JPMorgan Chase as initial partners.
Government Scrutiny and Legal Battles
The federal government’s heightened vigilance comes amid ongoing tensions between Anthropic and the Department of Defense (DoD). Recently, the DoD designated Anthropic as a supply chain risk to national security, a move that prompted public criticism from President Donald Trump and Defense Secretary Pete Hegseth. Both have expressed concerns over Anthropic’s insistence on ethical limits to its AI’s use in military applications, leading to federal departments being ordered to halt deployments of Anthropic’s platforms.
Despite the ban, the DoD reportedly continued using Anthropic’s Claude AI during the Iran conflict, as previously reported by CNBC. Anthropic is currently entangled in legal challenges over the blacklisting; a federal appeals court denied the company’s bid to temporarily block the DoD’s designation, while a separate injunction allowed it to maintain contracts with other government agencies.
Industry Reaction and Future Outlook
The financial and cybersecurity sectors have been closely monitoring the implications of advanced AI models after Anthropic disclosed that hackers—including a Chinese state-linked group—have previously exploited its technology to automate cyberattacks. This revelation, made public last November, heightened concerns about the potential for AI to facilitate sophisticated cybercrime.
Following the disclosure of Mythos’s cyber capabilities, shares of cybersecurity firms experienced a slump, reflecting market unease over emerging AI threats.
The joint efforts by the Federal Reserve and Treasury provide an early example of federal regulators engaging directly with private sector leaders to navigate the risks and benefits of rapidly advancing AI technologies. It also highlights the complex intersection of innovation, national security, and financial stability in the age of AI.
About Anthropic:
Anthropic is an AI research lab focused on developing reliable, interpretable, and steerable artificial intelligence. Its latest model, Claude Mythos Preview, is designed with cybersecurity applications in mind, representing a new frontier in AI-driven defensive tools.
For ongoing coverage of technology, cybersecurity, and financial markets, follow CNBC Tech and Markets news.