OpenAI says it can scan for abuse without retaining customer prompts. The proof is still to come
A preview called Private Safety Processing is intended to preserve zero-data-retention access for eligible API customers while detecting patterns across interactions.