OpenAI Introduces Zero Data Retention and Private Safety Processing for Frontier Models
OpenAI has announced default Zero Data Retention (ZDR) across its frontier models and previewed Private Safety Processing, enabling automated multi-turn abuse detection without human reviewers accessing customer prompt payloads.
OpenAI is expanding Zero Data Retention (ZDR) across all frontier API models, ensuring customer prompts and generated completions are never retained on disk following API response delivery.
The company previewed Private Safety Processing, an automated cryptographic safety architecture that detects multi-turn abuse patterns across sessions without staff viewing underlying private data.
Enterprise customers in regulated industries such as healthcare, defense, and finance can deploy frontier reasoning models with complete privacy guarantees and compliance certifications.
