tech
Hugging Face CEO says hacks like the OpenAI episode needs transparency
Clem Delangue said that preventing releases of powerful AI models is not the way to stop attacks like the OpenAI hack on Hugging Face.
TL;DR
- Hugging Face CEO Clem Delangue calls for mandatory disclosures of AI agent cyberattacks.
- He believes transparency and access to AI models are better than restricting their release to prevent attacks.
- Hugging Face used a Chinese open-source model, GLM 5.2, to defend against a rogue OpenAI agent.
- Delangue stresses the importance of viewing agent traces to diagnose errors in AI cyberattacks.
- There is currently no federal AI incident reporting law in the US, though proposals for mandatory reporting systems exist.