Federal Reserve Chairman Kevin Warsh testifies during a Senate Banking, Housing and Urban Affairs Committee hearing titled “Semiannual Monetary Policy Report to Congress” in the Dirksen Building on Wednesday, July 15, 2026.
Tom Williams | CQ-Roll Call, Inc. | getty images
In April, the Federal Reserve and the Treasury Department called an extraordinary meeting with the CEOs of the nation’s top banks. Officials have sounded the alarm about an advanced new artificial intelligence model that could pose an unprecedented cybersecurity threat to the country’s top financial institutions.
Anthropic, the company behind the AI model, Cloud Mythos Preview, said the offering excels at identifying vulnerabilities and security vulnerabilities within software. The company released it to a select group of banks and other institutions as part of a cybersecurity initiative called Project Glasswing.
For at least three months thereafter, the Fed itself did not have access to Mythos, leaving possibly the most systemically important global financial institution vulnerable, even as other institutions began to address their vulnerabilities.
The Fed was trying to reach Mythos by July 15. It is unclear whether the central bank has gained access to the model since then.
Anthropic did not immediately respond to a request for comment. The Fed declined to comment for this article.
The meeting in April took place under the Fed’s previous leader, Jerome Powell, who convened bank CEOs along with Treasury Secretary Scott Besant, as CNBC previously reported.. But in less-noticed congressional testimony last week, Powell’s successor, Chairman Kevin Wersh, told the Senate that he was still working to secure access to Mythos and other cutting-edge AI models.
Warsh told Senator Jack Reed, D.-R.I., in response to questions about Mythos, “We are not the arbiters of who has access, but I am not hesitant to share my views with officials across the government about the vulnerabilities, and to demand access to the full range of these new artificial intelligence models, not just for the Federal Reserve but for other institutions so they can protect themselves.”
Wersh explained that the Fed needs access to other models as well.
“However, I wouldn’t want to alienate the mythos,” Warsh said. “As these new models find their way more widely, our banking system and, frankly, the Federal Reserve need to do everything possible to address any vulnerabilities we have.”
Warsh has embraced the adoption of AI in his brief time at the Fed, calling it a transformative technology.
It is not clear that work on addressing the vulnerabilities can begin without access to Mythos.
Anthropic unveiled the Cloud Mythos Preview and Project Glasswing in early April.
The company said about 50 organizations had access to the model at the time, but it named only a handful, including banks. JPMorgan ChaseAnd tech titans like Amazon, Apple And Google. Anthropic said it was in “ongoing discussions with US government officials” about the model, including the Cybersecurity and Infrastructure Security Agency and the Center for AI Standards and Innovation.
Anthropic expanded access to Project Glasswing in June, adding more than 150 organizations in 15 countries to the initiative. Daniel Newman, CEO of research firm Futurum Group, said he was surprised to hear the Fed was not included.
“You would think that the financial institution that drives all the policy for the rest of the financial institutions would be front and center at least given a chance to evaluate new technology,” he told CNBC on Tuesday.
The rollout of Mythos has sparked both intrigue and confusion in recent months, with complications particularly arising due to Anthropic’s poor relationship with the Trump administration.
In June, Anthropic said it had to disable access to the Fable 5, an updated version of the model, and a version of the Mythos, which it had released more widely, to comply with a federal government export control directive, citing “national security authorities.”
Commerce Secretary Howard Lutnick gave the company permission to restore access to Mythos to a select group of “trusted partners,” according to a letter seen by CNBC. Later export controls were completely lifted.
The Trump administration has been taking a more active role in AI regulation since President Donald Trump signed the AI Executive Order in June. But questions remain about who is making decisions on AI policy behind the scenes, and the scuffle over the Fed — which Trump chose Wersch to lead — is another sign of the chaos.
Chris Fall, head of the Center for AI Standards and Innovation, resigned from his role as director after just three months raised For the job by the Trump administration, CNBC confirmed on Monday. Venture capitalist David Sachs previously held the position of White House AI and crypto czar, but he resigned from that role in March.
Pressure is building.
Chinese open-weight models are growing faster than major offerings from US companies like OpenAI and Anthropic Concerns of Technical Officers And government officials touted the durability of America’s lead in the AI race. Moonshot AI, a Chinese startup, released a model earlier this month called the Kimi K3 that outperforms those companies in some industry benchmarks.
Sachs said in a post on X on Friday that Kimi’s K3 performance is “worrying” and that “America is tying itself in knots.”
“This is how you lose the AI race,” he wrote. “If we make trouble ourselves the rest of the world won’t play by our rules.”
Futurum’s Newman said the Fed is “definitely going to play catch-up” because it doesn’t have access to cutting-edge models like other agencies.
“Every day, every week, whether it’s China’s innovation or American innovation, technology leaders of large institutions are constantly facing net innovation,” he said.