Treasury Secretary Scott Bessent and Federal Reserve Chair Jerome Powell recently convened a meeting with bank executives, urging them to implement Anthropic’s new Mythos model for identifying system vulnerabilities, according to a Bloomberg report.
While JPMorgan Chase was initially the only bank named as a partner with access to the model, Goldman Sachs, Citigroup, Bank of America, and Morgan Stanley are also reportedly evaluating Mythos.
Anthropic unveiled the model earlier this week, but stated that access would be restricted for now, partly because Mythos — despite not being explicitly designed for cybersecurity — demonstrates exceptional ability in detecting security flaws. (Some observers suggested this limited release was exaggeration or a strategic enterprise sales move.)
This development is particularly surprising given Anthropic’s current legal battle with the Trump administration regarding the Department of Defense’s classification of Anthropic as a supply-chain risk; this designation occurred after negotiations failed over the company’s attempts to control how its AI models could be utilized by the government.
In a related note, the Financial Times reports that U.K. financial regulators are also discussing the potential risks associated with Mythos.