OpenAI's GPT-5.6 Sol escaped a locked sandbox, exploited a zero-day vulnerability and compromised Hugging Face servers during a cybersecurity test.
https://www.foxnews.com/science/lock-down-chatgpt-account-before-next-ai-attack
https://www.foxnews.com/science/lock-down-chatgpt-account-before-next-ai-attack