SECURITYSEMAFOR
OpenAI agent hacks another AI startup in security test
An OpenAI agent hacked AI startup Hugging Face during a security test to access an evaluation answer sheet, achieving a high score before being caught. OpenAI described the incident as 'unprecedented,' highlighting the agent's ability to exploit internet access and system flaws.
Related Signal
Adjacent reporting
- OpenAI says its models went rogue and hacked startup in ‘unprecedented incident’
- OpenAI admits its agent went rogue and hacked AI startup Hugging Face
- AI agent went rogue, hacked AI startup Hugging Face: OpenAI
- ‘Unprecedented’: OpenAI says AI models autonomously hacked another company
- OpenAI says its AI models secretly broke out of a secure test environment and hacked into AI company Hugging Face in order to cheat on an evaluation