LinedatLinedat
Beta

Capabilities

The Linedat capability catalogue, anchored to regulation and with their limits

Trust & evidence

Append-Only Decision Evidence (via Trigger)

The evidence of your decisions is not edited: it is superseded, chained.

Framework / standard: ISO/IEC TR 38505-2 §5 (decide phase) · EU AI Act Art. 14

Many tools promise "append-only" and then back it up with a convention. Linedat backs it up with a database trigger that rejects UPDATE: a decision is not rewritten, it is superseded.

Immutable at the database level

Decision evidence links to the assets consulted, the query hash, the AI session that assisted and the reasoned rationale. A trigger rejects any UPDATE; a correction creates a new version pointing to the previous one. The truth is in the chain, not in a mutable row.

AI assistance trace

If a decision was assisted by AI, it is linked to the corresponding AI usage entry (EU AI Act Art. 14), without re-executing the chat. It links asset identities, not results: there is no data leakage.

The limits (what we do not claim)

It records the identities of the assets consulted and the query hash, not the query results. The query text is tenant opt-in. It is database-level immutability for decision evidence, distinct from the hash-chain of the audit logs.

How Linedat helps

Linedat makes the evidence of your governance decisions genuinely immutable — by trigger, not by convention — ready to defend a decision before a regulator.

FAQ

Respuestas sobre implementación y capacidades

A database trigger that rejects UPDATE. A correction can only create a new version that supersedes the previous one; the original remains.

Govern your data and your AI with Linedat

Connect your sources and in minutes you have catalog, quality, lineage and AI governance. Free to start.