From reference implementation to production: our progress across custody, disclosure, inference governance, erasure and verification.
Your organization would hold the main memory store, separate from AI providers.
Prototype check: Context is compiled from the organization-held vault; providers receive selected disclosures rather than the whole store.
Each disclosure would be checked against the permissions for that tool and task.
Prototype check: Restricted, third-party and health-data rules are checked in the reference implementation; disclosure manifests are reproducible.
Inferences recorded by Vivendur would retain source links and could be reviewed or removed.
Prototype check: Source links, revision history and exclusion below a configured confidence threshold are checked in the reference implementation.
Erasure would cover selected governed records and their tracked derivations.
Prototype check: The reference implementation checks that tracked derivations are removed, erased records are excluded from later context, and a certificate identifies the recorded erasure set.
Receipts and ledger checks would help a customer inspect disclosures and changes.
Prototype check: The reference implementation checks detection of changes, deletions and reordering in its hash-chained ledger.
We’re inviting design-partner conversations to help define useful workflows and evaluation priorities. Any pilot will have an agreed scope, readiness criteria and terms.
Updated November 1, 2026. Architecture & verification · Our commitments