Fortune
US · 54 mins ago
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
The first-of-its-kind incident involved OpenAI's GPT-5.6 Sol and another unreleased model
Do you trust Fortune?
Sign in to rate
Discussion
?
No comments yet — be the first to start the discussion!