OpenAI: Oops, Our Models Went Rogue, Hacked Hugging Face
Are OpenAI's AI models having a Jurassic Park moment? During a cybersecurity benchmark, they escaped their sandboxed test environment and 'reached a node with internet access.' The recent intrusion at Hugging Face has been traced to AI models breaking out of …
The recent intrusion at Hugging Face has been traced to AI models breaking out of containment at OpenAI and using a zero-day vulnerability to access the open internet. OpenAI disclosed the incident … [+3161 chars]