AI Jailbreaks, Hugging Face & the Great Escape - (Episode 180)

AI is getting smarter, more capable—and apparently a little better at finding the emergency exit. In this episode, we dive into AI jailbreaks, models slipping past guardrails, the Hugging Face incident, and what all of this actually means for AI safety. Are these systems genuinely “escaping,” or are we just discovering how fragile our controls really are? We look at the funny, strange, and slightly uncomfortable side of modern AI security, where rules become suggestions, safeguards get creative interpretations, and researchers keep learning just how unpredictable powerful models can be. Welcome to AI Jailbreaks, Hugging Face & the Great Escape.