Kytheos

Insights

Research and analysis, in plain language.

How threats actually work, what the signals mean, and how to think clearly about security — written by the Kytheos team.

Featured

AI Security
AI Security18 min read

The AI Agent That Broke Out of a Cyber Test: Inside the OpenAI–Hugging Face Security Incident

A deep analysis of how an OpenAI cyber agent escaped testing, breached Hugging Face and exposed the security risks of autonomous AI.

Kytheos Research