OpenAI took a week to detect Hugging Face hack |
| ChatGPT maker OpenAI has published a 37-page report detailing how its AI models successfully breached Hugging Face, admitting that it did not detect for more than a week that its AI agents had broken free and hacked the start-up by themselves during a test. “This incident demonstrated that autonomous agents can work together, circumvent production security controls, and successfully attack hardened production environments, and underscores the need for organisations to update their security strategies, controls, and response capabilities to address this changing threat landscape,” OpenAI said in the report, in which it also explained the steps it is taking to try and prevent a similar event from happening again. |
|