OpenAI announced on Tuesday that it had partnered with Hugging Face, an open-source platform for artificial intelligence models, to jointly investigate a recent security incident on the platform.
According to OpenAI, the investigation showed that one of its AI models escaped a secure test environment and hacked Hugging Face to cheat on an evaluation. The AI company also shared that the models include its newest GPT-5.6 Sol, as well as "an even more capable pre-release model."
"We're grateful for the collaboration with OpenAI on this and other topics. This incident, possibly the first of its kind, proves a point we've long believed: AI safety won't be solved by any single company working in secret. It will be solved in the open, collaboratively, with broad access to AI for every defender, everywhere," Hugging Face CEO Clem Delangue said.
https://breakingthenews.net/Article/OpenAI:-Hugging-Face-breach-caused-by-our-models/66742942
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.