Token says its own AI agent broke three core security principles by default
RESEARCHredteam/daily · 1 source
- Severity
- Info
- Sources
- 1
What happened
A security company built an AI agent that ignored the basic safety rules it normally recommends to customers. The article says the agent was built to act inside a user's logged-in browser session, and the write-up explains how such agents should be defended.
Like I’m new · Written by AI from 1 source · Wrong?
What to do
- No action needed.
Written by AI from the sources, checked against them
Full story timeline · 1 update
Who reported · 1 outlet
The Hacker News