AI security tool exposes thousands of security flaws, costing banks millions

What happened

Bank of America, for instance, hired 1,000 contractors and spent hundreds of millions of dollars to address flaws identified by Anthropic's AI model, Mythos. This powerful tool, capable of autonomously discovering and exploiting unknown vulnerabilities, has been granted access to a select group of organizations, including major banks.

AI/ML AI security tool exposes thousands of vulnerabilities, costing banks millions October 18, 2026 Share By SC Staff (Adobe Stock) Per Tech Radar. JPMorgan Chase and Goldman Sachs also report dedicating significant resources and personnel to managing the risks exposed by such AI.

While these tools offer enhanced security, they also highlight the urgent need for financial institutions to bolster their cybersecurity budgets and remediation capabilities to keep pace with rapidly evolving AI-driven threat detection. The implications for economies, public safety, and national security could be severe if these vulnerabilities are not addressed promptly.