Jamie Dimon says broad access to Mythos would be like giving 'ballistic missiles to individuals'

Jamie Dimon warned about the risks of Anthropic's Mythos model. Alex Wong/Getty Images Jamie Dimon has sounded a warning about the vast abilities of Anthropic's Mythos model. He said broad access to Mythos would be like giving "ballistic missiles to individuals." Mythos, Anthropic's strongest model, is known for its knack at finding vulnerabilities in software. JPMorgan Chase CEO Jamie Dimon sounded the alarm on what broad access to Anthropic's Mythos-class models would look like. Speaking at the Pennsylvania Defense and Innovation Summit on Thursday, Dimon said that Mythos was a "real issue," one that the US government has been "on top of at this point." He said the government scrutiny was to ensure the model could be controlled when it is rolled out, "because you're giving ballistic missiles to individuals with Mythos, basically." Dimon was speaking on an "Investing in America" panel alongside Pennsylvania Sen. Dave McCormick. The summit also brought together other prominent figures in politics and defense, including President Donald Trump, Defense Secretary Pete Hegseth, and Palantir's defense head, Mike Gallagher. Mythos 5 is Anthropic's most capable frontier AI model, capable of finding vulnerabilities in operating systems. Critics, and Anthropic itself, have warned that the model poses strong cybersecurity risks if made available to the public. Anthropic said in April that it halted the release of the model because it was too good at finding "high-severity vulnerabilities." Instead of granting broad access, Anthropic said it would grant access only to a select group of US organizations. In June, it released Fable 5 to the general public, a Mythos-class model with guardrails against topics such as cybersecurity, distillation attacks, and biology and chemistry. But the US government blocked access to both Mythos 5 and Fable 5 about a week later, determining that Fable 5's guardrails could be bypassed. It slapped export controls on the model to restrict foreign nationals from using it. In response, Anthropic blocked access for all users. On June 30, the AI lab announced that access had been restored . "We've received notice that the Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5," Anthropic said in an X post on June 30. "We'll begin restoring access tomorrow, and will share an update soon. We're grateful to our users for their patience, and to everyone who worked with us on redeploying the models." Anthropic and JPMorgan Chase did not respond to requests for comment from Business Insider. Read the original article on Business Insider