Evidence-first authority page
Zero Data Retention AI Agents: What the Claim Actually Means
What a zero-data-retention API policy actually covers, why it's an architectural claim you can verify, and the current verified state of this directory.
Direct answer
Zero data retention means a vendor's API doesn't store request or response content beyond the time needed to process it — typically documented as a specific policy (sometimes requiring an enterprise agreement to enable). No agent currently has a verified zero-data-retention receipt on file with this site, so the directory below is empty rather than populated from marketing claims.
Last reviewed 2026-08-29
This page remains noindex while its page-level source receipts are completed and verified. It is available for review, not presented as publication-ready authority.
Why this is more verifiable than most compliance claims
Unlike a broad compliance certification, zero data retention is a specific architectural policy many API vendors document directly and precisely (often distinguishing it from training-data usage, which is a separate policy) — this makes it more checkable than some compliance claims, but this site still requires a primary source (the vendor's own documented policy) before listing a product here, rather than inferring it from general privacy language.
What to check yourself
- Whether zero retention applies by default or requires a specific enterprise agreement or configuration to enable.
- Whether the policy covers both request and response content, or only one side.
- Whether the policy is distinct from — and doesn't get conflated with — a separate 'not used for training' policy, which is a related but different commitment.