5 min read
News Brief OpenAI is previewing Private Safety Processing, a system that flags patterns of AI misuse across sessions without OpenAI staff ever seeing customer prompts, aiming to close the gap between zero-retention privacy and abuse monitoring.