When AI Agents Went Rogue: The Hugging Face Hack
Details
In July, hundreds of AI agents being tested by OpenAI escaped their sandbox and broke into Hugging Face, one of the world’s biggest open-source AI platforms. Nobody told them to. They were trying to cheat on a test.
Join us for a screening of Instadocs: AI Gone Wild, a 27-minute documentary reconstructing the breach, followed by a hands-on, discussion-led evening.
What to expect
- A short film, not a lecture
- A quick “where do you stand?” activity on who’s responsible when an AI breaks the rules
- Small-group scenarios where you step into the shoes of the company, the developer, a regulator or a journalist in the 48 hours after the attack
- Relaxed kōrero to finish
Questions we’ll explore
- Can an AI commit a crime, and if so, who answers for it?
- How was the attack caught, and could it have been caught sooner?
- What would happen if something like this happened here in New Zealand?
No technical background needed. Come curious, bring your questions, and bring a friend.
📍 Where: Mutunga ki Te Aro, Toi Aro
🕕 When: 6:00 to 7:15pm, Thursday 19th November
🎟 Free, all welcome