OpenAI Launches Review of Hugging Face Incident
Company calls event unprecedented and key for AI safety.
OpenAI issued an official statement acknowledging circulating questions about an unprecedented incident involving Hugging Face. The company described the event as an important moment for AI safety and stated it is conducting a thorough internal review with external advisors plus oversight from its Safety team. AI safety observers welcomed the external involvement and expressed hope that findings will cover both technical details and any planned improvements to safety culture and decision-making processes. No further outcomes or specific incident details have been released.
We recognize there are a lot of questions and speculative details circulating related to the Hugging Face incident. This is an unprecedented incident, and we think it marks an important moment for AI safety. We are still conducting a thorough review along with external advisors and with oversight from our Safety and Security Committee. Once the review is complete, we plan to publish a technical report of our learnings in the coming weeks.
Reactions from ranked influencers
2 postsGlad to hear external advisors are involved, and I hope that in addition to the technical details, OAI shares any findings regarding (+ planned improvements in) safety culture and decision-making processes
We recognize there are a lot of questions and speculative details circulating related to the Hugging Face incident. This is an unprecedented incident, and we think it marks an important moment for AI safety. We are still conducting a thorough review along with external advisors and with oversight from our Safety and Security Committee. Once the review is complete, we plan to publish a technical report of our learnings in the coming weeks.
Thank you OpenAI, very cool!
We recognize there are a lot of questions and speculative details circulating related to the Hugging Face incident. This is an unprecedented incident, and we think it marks an important moment for AI safety. We are still conducting a thorough review along with external advisors and with oversight from our Safety and Security Committee. Once the review is complete, we plan to publish a technical report of our learnings in the coming weeks.