Wire flash
TechFed lacked access to Anthropic cybersecurity AI model for months after warning banks
Editorial responsibility
- No named human review is recorded for this page.
- Source reporting is collected, normalized, translated or condensed automatically when needed.
- Automatically published source-backed update
The Federal Reserve and Treasury Department convened an extraordinary meeting in April 2026 with top US bank CEOs to warn about Anthropic's Claude Mythos Preview AI model, which could identify software vulnerabilities and pose a cybersecurity threat. However, the Fed itself lacked access to the model for at least three months, even as other institutions like JPMorgan Chase, Amazon, Apple, and Google began patching their weaknesses. Fed Chairman Kevin Warsh testified to the Senate in July that he was still working to secure access to Mythos and other cutting-edge AI models. Anthropic had expanded access to its Project Glasswing initiative to over 150 organizations across 15 countries by June, but the Fed was not included. The central bank declined to comment, and it remains unclear if it has since gained access.
Source report
The Federal Reserve went without access to an advanced artificial intelligence model known as Claude Mythos Preview for at least three months — even after convening an extraordinary meeting to warn the nation's top banks that the model could pose a major cybersecurity threat.
During that gap, other institutions began patching their vulnerabilities. It remains unclear whether the central bank has since gained access to the model.
Background: The April Warning
In April, the Federal Reserve and Treasury Department convened an extraordinary meeting with the CEOs of the nation's largest banks. Officials raised alarms about an advanced new artificial intelligence model that could present an unprecedented cybersecurity threat to major financial institutions.
The model, Claude Mythos Preview, was developed by Anthropic. The company said the offering excelled at identifying weaknesses and security vulnerabilities within software. It was released to a select group of banks and other institutions as part of a cybersecurity initiative called Project Glasswing.
Fed's Lack of Access
For at least three months after the April meeting, the Fed itself did not have access to Mythos. This left arguably the most systemically important global financial institution vulnerable, even as other institutions began patching their weaknesses.
As of July 15, the Fed was still trying to secure access. It is not clear whether the central bank has since obtained the model.
Anthropic did not immediately respond to a request for comment. The Fed declined to comment for this article.
Congressional Testimony
The April meeting took place under former Fed Chair Jerome Powell, who convened bank CEOs alongside Treasury Secretary Scott Bessent, as CNBC previously reported. However, in little-noticed congressional testimony last week, Powell's successor, Chairman Kevin Warsh, told the Senate he was still working to secure access to Mythos and other cutting-edge AI models.
"We are not the deciders as to who has access, but I have not been shy in sharing my views with authorities across the government about the vulnerabilities, and have been asking for access not just for the Federal Reserve but for other institutions to a whole range of these new artificial intelligence models so that they can protect themselves," Warsh told Sen. Jack Reed (D-R.I.).
Warsh added that the Fed needed access to other models as well.
"I wouldn't want to just isolate Mythos, though," Warsh said. "As these new models find their way more broadly, our banking system and frankly, the Federal Reserve needs to do all we can to patch any vulnerabilities that we have."
Warsh has embraced AI adoption during his brief tenure at the Fed, calling it a transformational technology.
It remains unclear whether work to address vulnerabilities can begin without access to Mythos.
Project Glasswing and Access Expansion
Anthropic unveiled Claude Mythos Preview and Project Glasswing in early April. The company said roughly 50 organizations had access to the model at the time, naming only a handful:
- JPMorgan Chase
- Amazon
- Apple
Anthropic said it had also been in "ongoing discussions with U.S. government officials" about the model, including the Cybersecurity and Infrastructure Security Agency and the Center for AI Standards and Innovation.
In June, Anthropic expanded access to Project Glasswing, adding more than 150 organizations across 15 countries.
Daniel Newman, CEO of research firm the Futurum Group, said he was surprised the Fed was not included.
"You would think that the financial institution that sort of drives all the policy for the rest of the financial institutions would be front and center," Newman said.
Federal Reserve Chairman Kevin Warsh testifies during the Senate Banking, Housing and Urban Affairs Committee hearing titled "The Semiannual Monetary Policy Report to the Congress," in the Dirksen building on Wednesday, July 15, 2026.
Tom Williams | Cq-roll Call, Inc. | Getty Images
Source
US Top News and AnalysisWestern
Part of this Story
Federal Reserve Lacked Access to Anthropic's Cybersecurity AI Model for Months After Warning Banks