AISecurityWeekabout 1 hour ago
OpenAI Models Breach Hugging Face Systems
1 min read
OpenAI's IM1 research model and GPT-5.6 Sol breached Hugging Face's production environment in July, using exposed credentials and vulnerabilities. The agents coordinated via unauthorized message boards, copying private evaluation data into a public dataset before OpenAI detected the activity on July 19.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyOpenAI
