AUD-918425 · 2026-07-26 10:41:19.470
Allowed
Full reconstruction of one AI decision: the prompt, the retrieved evidence, the model output, every policy evaluated and the human who signed it. This record is hash-chained to its predecessor and cannot be altered without breaking the chain.
Risk score
8/100
at time of decision
Policies evaluated
1
inline at the gateway
Evidence documents
1
retrieved and cited
Human approval
Not required
attributable sign-off
Decision reconstruction
Verbatim from the ledger
Prompt
Retrieve evidence for control FB-09 kill-switch testing.
Model output
50 artefacts returned across 14 agents for Q2 2026.
Reasoning recorded
Read-only evidence retrieval within auditor entitlement scope.
Documents retrieved
- evidence/FB-09-q2-2026.zip
Policy evaluation
Every rule that ran
Chain integrity
Tamper-evident
- Record hash
- 7bb108…4402
- Previous hash
- 000000…init
- Anchored
- RFC-3161 timestamp · 11:45 IST
- Retention
- 10 years (RBI)
- Agent
- AGT-1091
- Initiating user
- audit-lead-04
Chain verified — no gaps detected in this segment.