OpenAI is adding a new safety capability that allows enterprises to detect misuse of its
OpenAI is adding a new safety capability that allows enterprises to detect misuse of its AI systems across multiple interactions without retaining prompts or responses, enabling risk monitoring while preserving its Zero