U.S. Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell summoned bank executives for a meeting this week, where they encouraged the use of Anthropic’s latest model to detect vulnerabilities, according to Bloomberg.
While JPMorgan Chase was the only bank listed as one of the initial partner organizations with access to the model, Goldman Sachs, Citigroup, Bank of America, and Morgan Stanley may also be testing Mythos.
While Anthropic announced the model this week, the firm said it would be limiting its access, partly because Mythos is “too good at finding security vulnerabilities,” despite not being specifically trained for cybersecurity. Some have questioned this claim, saying it is hype, or part of the enterprise sales strategy.
READ: Pentagon clashes with Anthropic over military AI use safeguards (January 30, 2026)
In its official blog post, Anthropic introduced Mythos as a breakthrough in autonomous cybersecurity, warning that the model could be dangerous if it fell into the wrong hands. Rather than allowing broad access, the company is reportedly testing it through a closed initiative called “Project Glasswing,” giving access to Mythos only to select researchers and organizations.
Anthropic says that releasing it to the public would be like handing advanced hacking capabilities to anyone with a laptop.
The hype around Mythos has led to mixed reactions. While some have called it a big leap in security, others have questioned Anthropic’s claims, saying this might be a strategic plan designed to grab attention. The Wall Street Journal, however, had reported that the U.S. government officials had met Wall Street banks, telling them to get ready for Mythos and do a thorough review of their digital security.
READ: US judge gives Anthropic reprieve from Pentagon blacklisting (March 27, 2026)
While not much information about Mythos is currently available to the public, Anthropic has said it is designed to explore software independently, identify vulnerabilities, and chain them together into full-blown exploits. Rather than simply flagging a potential problem, the model can dig into the software, find hidden flaws, and figure out exactly how to break into a system, all by itself.
The endorsement of Mythos by Trump officials is especially notable since the administration had designated Anthropic a “supply chain risk.” The AI company is currently challenging the designation in court. The Pentagon had reportedly blacklisted Anthropic following the company’s refusal to allow the military to use AI chatbot Claude for U.S. surveillance or autonomous weapons.
Meanwhile, the Financial Times reported that UK financial regulators are also discussing the risk posed by Mythos.