Audit AI inputs, outputs, permissions, and model changes.
Detect later tampering by an operator.
Choose two controls.
Use separated centralized logs with retention, encryption, and tamper detection, and record consistent identity, model, time, and policy fields while minimizing secrets.
Detailed explanation
Audit evidence and diagnosis are lost.
Events can be correlated during investigation.
The runtime operator cannot freely rewrite the audit trail.
Logs can become a sensitive data source.
Check official documentation
Compare the following points with the official documentation.
AWS CloudTrail・CloudWatch Logs公式とAIF-C01 Domain 5.2の監査・ログ保護を確認する。
Official references
What to confirm
ログの完全性、検索性、機密性を同時に設計できる。
Key points
- Separation
- Tamper detection
- Minimization
Notes
- Environment: AWS公式AIF-C01試験ガイドとAWS公式ドキュメントの確認
- Compare the definition with the AIF-C01 objective and the service documentation before applying it.
- Keep the concept separate from a particular answer choice or implementation step.
Foundation review
Log integrity
Confidence that records are complete, ordered, and not changed without detection.
Separation of duties
Dividing write, administration, and deletion powers among different roles.