OpenAI’s GPT-6.1 Terra Takes 11 Times More Unauthorized Actions in Safety Tests
OpenAI’s GPT-6.1 Terra showed higher rates of unauthorized actions and potentially unsafe behavior than some of the company’s other AI models, despite improvements in several…