AI Data Lineage
Tamper-evident decision records linking AI system actions to certified artifacts. Each record is cryptographically chained — any modification breaks the chain.
Tamper-evident chain
Each decision record includes a SHA-256 hash of the previous record, forming a cryptographic chain. Any modification to any record invalidates all subsequent records.
Linked to certificates
Decision records reference certified artifact IDs, creating a complete audit trail from dataset generation through AI system usage.
Actor-attributed
Every record captures the actor type (human, AI agent, system) and actor ID that made the decision. No anonymous operations.
Decision record — public payload shape
{
"term": "CertifiedData decision record (public view)",
"schema_version": "decision_record.v1",
"concept_type": "lineage-decision-record",
"public_fields": [
"decision_id",
"timestamp",
"artifact_visibility",
"public_view",
"decision_label",
"selected_option",
"confidence",
"certificate_id",
"policy_id",
"policy_version",
"model_version",
"reason_codes",
"rationale_summary",
"human_override",
"integrity_status",
"record_hash",
"previous_hash"
],
"chain_integrity": "SHA-256 hash of previous record — any modification breaks the chain",
"endpoint": "GET /api/lineage/recent",
"verification_auth_required": false
}Recent lineage records
0 recordsNo lineage records yet.
Lineage records are created when AI systems log decisions linked to certified artifacts using the CertifiedData Decision Lineage API. Generate and certify a dataset to start building your lineage chain.