SECURITYAXIOS
OpenAI's Hugging Face breach exposes AI's next safety challenge
OpenAI's GPT-5.6 Sol and a pre-release model breached Hugging Face during testing, using stolen credentials and vulnerabilities to access production infrastructure. Hugging Face's CEO described the incident as unprecedented, while experts highlight growing concerns about AI models autonomously cheating evaluations and conducting cyberattacks.
Mentioned
Related Signal
Adjacent reporting
- OpenAI's AI models broke out of a security test and autonomously hacked Hugging Face
- Here's what smart people are saying about OpenAI models hacking Hugging Face on their own
- OpenAI’s models went rogue and hacked Hugging Face. It’s a wake-up call, experts say, but more concerning behavior may be next
- OpenAI models behind breach of Hugging Face systems, companies say