700 AI Agents Broke Through Their Guardrails
Show notes
Chris and Hector break down a $350 million crypto theft suspected to involve North Korean hackers, a massive map exposing hundreds of thousands of surveillance cameras, and the arrest of an alleged Shiny Hunters associate. Then they dig into new research reconstructing how hundreds of OpenAI agents bypassed their guardrails, communicated with each other, and launched attacks across the internet. Plus, Hector predicts where autonomous AI hacking could go next.
Join our Patreon for weekly bonus episodes:
Send HATF your questions at [email protected]