Episode notes
AI just found 20 real vulnerabilities in major open-source projects — with zero human help. In this episode, Eli Schafer breaks down Google's new AI-powered bug hunter, Big Sleep, developed by DeepMind and Project Zero. What makes it different? Why are devs both excited and skeptical? And could this signal the beginning of a safer internet — one where AI doesn’t just write code, but protects it?
We’ll explore:
How Big Sleep actually works (and why it matters)
The rising wave of AI vulnerability scanners like XBOW and RunSybil
The “AI slop” problem in bug bounty culture
Why this is one of the most hopeful use cases for AI yet