OpenClaw AI agent tricked into phishing attacks, with user data compromised

Varonis has a suggestion on how to make AI agents more careful.

from Latest from TechRadar https://ift.tt/7lxZUjt