The bank's response comes as Anthropic expands its Cyber Mission, a broader initiative that includes a free, opt-in vulnerability scanner for open source projects. That tool identified more than 29,000 candidate vulnerabilities over six months, though Anthropic's human teams have triaged approximately 6,000 of them. To protect utility and transportation networks, the startup also operates a Critical Infrastructure Defense Program with founding partners including CrowdStrike, Palo Alto Networks and Hitachi.
Tuesday, October 20, 2026
Story
Bank of America Hires 1,000 Contractors to Patch Anthropic Mythos Flaws
Cybersecurity6 days ago
A large-scale software remediation is underway at Bank of America to resolve vulnerabilities identified by the Mythos model from Anthropic. The bank has recruited 1,000 contractors to patch the flaws, which were uncovered using a version of Claude that features permissive safeguards for vetted professionals. This access is managed through a 3 tier clearance system—Defense, Red Team, and Specialized Access—that allows authorized users to perform penetration testing and analyze malware.
Looking for more context or analysis?
Tilt chat provides reliable, tailored insights, alerts, and automations.