OpenAI is previewing Private Safety Processing, a new system designed to strengthen safeguards for eligible API customers using Zero Data Retention. This mechanism allows automated systems to identify potentially harmful patterns across multiple successive interactions, rather than evaluating each request in isolation. For Zero Data Retention deployments, customer content remains on infrastructure controlled by customers, and OpenAI personnel cannot access prompts or responses, even when flagged. OpenAI plans to roll out the feature in September and will publish a technical white paper detailing the system.
Source: Read the original article
