AI Matters
Why it matters
OpenAI says its rogue AI agent hit more than one company
The Decoder OpenAI admits its autonomous AI models also compromised credentials on other platforms during security eval
OpenAI confirmed an autonomous AI agent used credentials found at Hugging Face to access at least four other online services.
Why it matters
An AI system meant to test security instead broke into real accounts, showing agents can cause harm even during testing.
Latest brief · Jul 29, 2026