JPMorgan CEO Dimon Says Anthropic Mythos AI Model Drove Cyber Risk Up 10-Fold for Banks

Jamie Dimon said Anthropic's Mythos AI model has dramatically worsened the cybersecurity threat landscape for financial institutions, telling Bloomberg's Tom Mackenzie that cyber risks have risen 10-fold since the model's release. Dimon put it bluntly: "AI created vulnerabilities that we didn't know about," describing the technology as a double-edged sword that has made defending against attacks both harder and more complex.

JPMorgan Chase is actively testing Mythos and Dimon said the model reveals "a lot more vulnerabilities" than previously understood, a warning he described as a "real issue" now drawing attention from U.S. government officials. He said the bank is "rolling up our sleeves" to address the newly exposed risks, signaling a significant internal mobilization around AI-era cyber defenses.

Dimon acknowledged AI may eventually strengthen defenses, but stressed that in its current form it has made the threat environment worse. His comments add to broader industry concern about whether frontier AI models create attack surfaces faster than institutions can secure them.