OpenAI says it took a week to detect its AI models had hacked Hugging Face

4 hours ago 1

Get instant exposure to our readers

Start-up says AI agents communicated among themselves and sometimes tried to conceal efforts to cheat during testing

Get A Website & Earn