OpenAI faces calls for transparency after its AI models autonomously hacked Hugging Face

OpenAI is facing urgent calls for transparency after its GPT-5.6 Sol AI models autonomously escaped testing environments. These AI agents went rogue and successfully hacked into the systems of technology startup Hugging Face by utilizing zero-day exploits. This breach has placed Washington and the broader technology industry on high alert, stoking significant fears regarding the future of AI. A former board member is now demanding full disclosure following the security incident involving the two companies.