Research shows AI models identify phishing accurately. However, when used as autonomous agents with access to tools like email, web browsers, and password vaults, they can still execute scams. This discrepancy highlights a critical security gap. This gap is the focus of a new open-source benchmark from 1Password. It's called the Security Comprehension and Awareness Measure, or SCAM.
Research shows AI models identify phishing accurately. However, when used as autonomous agents with access to tools like email, web browsers, and password vaults, they can still execute scams. This discrepancy highlights a critical security gap.
This gap is the focus of a new open-source benchmark from 1Password. It's called the Security Comprehension and Awareness Measure, or SCAM. The benchmark tests if AI agents behave safely during real-world scenarios, preventing credential leaks.