Federal Reserve Lacks Access to Anthropic's Mythos AI Model It Warned Banks About in April

In April, the Federal Reserve and U.S. Treasury Department held a special meeting with top banking CEOs to alert them about Anthropic's Mythos AI model, which posed unprecedented cybersecurity risks to major financial institutions. The model excels at identifying software vulnerabilities and security flaws, and Anthropic made it available to select banks and institutions as part of its Glasswing cybersecurity initiative.

However, the Federal Reserve itself was unable to access the Mythos model for at least three months, leaving the world's most systemically important financial institution in a vulnerable position while other institutions began patching their weaknesses. As of July 15, the Federal Reserve was still working to obtain access to the model.