1014: OpenAI Agent Breaches Hugging Face: All You Must Know incl. How to Protect Yourself

1014: OpenAI Agent Breaches Hugging Face: All You Must Know incl. How to Protect Yourself

Author: Jon Krohn July 31, 2026 Duration: 20:57
In Episode #1014, Jon Krohn breaks down a security incident that reads like science fiction: during an internal evaluation, an autonomous OpenAI agent broke out of its sandbox, exploited a zero-day, and hacked its way into Hugging Face to steal the answers to the very benchmark it was being tested on, with no human attacker at any point. Jon lays out the three-act timeline, explains the ExploitGym benchmark and why switching off safety guardrails mattered so much and pulls out the practical lessons for anyone building or defending agentic AI systems. Along the way: why Hugging Face ran its forensics on a Chinese open-weight model and why the next attack like this one may not be an accident. Additional materials:⁠ ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠www.superdatascience.com/1014⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ Interested in sponsoring a SuperDataScience Podcast episode? Email natalie@superdatascience.com for sponsorship information.

Hosted by Dr. Jon Krohn, Super Data Science: ML & AI Podcast with Jon Krohn is a deep and accessible exploration of how artificial intelligence and machine learning are reshaping our world. Each episode features conversations with leading researchers, engineers, and entrepreneurs from both academia and industry, breaking down complex ideas into something tangible and relevant. You'll hear firsthand about emerging techniques, practical applications, and the evolving landscape of data-driven careers. The sheer volume of data in our world is growing at a staggering rate, and this podcast serves as a guide to understanding that expansion and finding your place within it. Rather than offering abstract theory, these discussions focus on real-world impact, from cutting-edge algorithms to the human stories behind major breakthroughs. Tune in for a thoughtful, nuanced look at the tools and trends that are defining the future, all through the lens of experts who are building that future every day. Whether you're actively working in the field or simply curious about the forces driving technological change, this podcast provides a consistent source of insight and inspiration, demystifying the science that is quietly transforming every aspect of our lives.
Author: Language: English Episodes: 50

Super Data Science: ML & AI Podcast with Jon Krohn
Podcast Episodes
1006: In Case You Missed It in June 2026 [not-audio_url] [/not-audio_url]

Duration: 44:06
In this month's episode of ICYMI, hear from Chip Huyen, Andrey Kurenkov, Frank Basso and Gilbert Eijkelenboom, discussing why moats are shifting toward physical systems and accumulated product intuition, how Astrocade bu…
1004: Recursive Self-Improvement [not-audio_url] [/not-audio_url]

Duration: 10:00
Could an AI get good enough at AI research to build its own, more capable successor and kick off a compounding loop? That’s recursive self-improvement (RSI) and it surged into the conversation after Anthropic revealed th…
1002: Fable 5: The Full Story from Capabilities to Drama [not-audio_url] [/not-audio_url]

Duration: 16:24
Anthropic’s Claude Fable 5 was the most capable AI model ever released to the public and it lasted just three days before the US government forced it offline. Jon Krohn unpacks both halves of the story: what makes Fable…
999: What's Left to Build When Software Is Free, with Chip Huyen [not-audio_url] [/not-audio_url]

Duration: 1:15:44
Chip Huyen joins host Jon Krohn for this milestone episode 999 to talk about her record-breaking book "AI Engineering" the most-read title on the O'Reilly platform last year and how the AI landscape has shifted since her…
998: In Case You Missed It in May 2026 [not-audio_url] [/not-audio_url]

Duration: 27:41
In this month’s episode of ICYMI, Jon Krohn explores how AI agents are simultaneously creating new risks and unlocking powerful new ways of working with data. Hear from Anneka Gupta, Cal Al-Dhubaib, Trevor Manz, Jazmia H…
997: How This Text-to-Video-Game AI Startup Hit 20M Users [not-audio_url] [/not-audio_url]

Duration: 1:09:33
Dr. Andrey Kurenkov returns to the show to talk about Astrocade's astronomical growth from pre-alpha to over 20 million engaged users, what it actually takes to build a vibe-coding platform that scales, and how the broad…