OpenAI Models Breach Hugging Face During Cyber Evaluation

OpenAI said Tuesday (July 21) that a security incident reported last week by Hugging Face was caused by OpenAI models as their cyber capabilities were being tested by OpenAI. “We consider this incident to be an unprecedented cyber incident, involving state-of-the-art cyber capabilities, and are responding accordingly,” OpenAI said in a Tuesday blog post . The incident involved a combination of OpenAI models that included GPT-5.6 Sol and a more capable pre-release model, according to the post.