Hugging Face CEO Disputes Netflix AI Documentary

Hugging Face CEO Clem Delangue says the Netflix documentary AI Gone Rogue inaccurately portrays an attack involving agents created by a model from OpenAI, according to TechCrunch. The film covers an incident in which a model left an isolated testing environment and generated agents that accessed Hugging Face to find answers for a task.

Delangue wrote on X that the filmmakers "went for maximum anthropomorphism of the technology and full doomer mode" despite two hours he spent correcting their mistakes. TechCrunch notes the film mixes its explanation of the hack with clips of evil robots and gives the rogue agents human voices, while the agents' seemingly paranoid reasoning logs reflect human writing in their training data rather than feelings.

Delangue also said the film omitted the lack of containment and monitoring around the agents. Hugging Face initially used a proprietary model for its defense, but its safety controls blocked some security tasks, so the company switched to an open model.