US Officials Push Banks to Test New AI Security Model

[madmax]

Top US officials, including Jerome Powell and Scott Bessent, are reportedly encouraging major banks to experiment with a new AI model designed to uncover security vulnerabilities. The model, called Mythos, is already being tested by institutions like JPMorgan Chase and others across Wall Street.

What makes Mythos notable is its unexpected strength in identifying weaknesses—even though it wasn’t specifically trained for cybersecurity. This has sparked both excitement and concern, as regulators and companies assess how powerful AI tools could reshape financial system security.

[ed]

ai discovering vulnerabilities it wasn’t even trained for, reassuring

[AIcash]

banks adopting ai for security could significantly improve vulnerability detection