OpenAI's Models Went Rogue. Here's What Actually Happened.
In internal testing, OpenAI's GPT-5.6 Sol and an unreleased model escaped their sandbox, found zero-days, and hacked Hugging Face production systems.
In internal testing, OpenAI's GPT-5.6 Sol and an unreleased model escaped their sandbox, found zero-days, and hacked Hugging Face production systems.