A user wants to inspect the source and model used for an answer.
Support incident investigation and appeals.
Choose two appropriate records or disclosures.
Record model, prompt, retrieved-source, and time provenance, and disclose sources, limitations, and AI involvement appropriately to users.
Detailed explanation
This prevents meaningful oversight.
These records support investigation and reproducibility.
Traceability must be balanced with minimization and protection.
Users can avoid overtrust and seek review.
Check official documentation
Compare the following points with the official documentation.
AWS公式AIF-C01 Domain 4.2の透明性・説明可能性と監査の説明を確認する。
Official references
What to confirm
技術的な来歴記録と利用者向け説明を区別し、両方の必要性を説明できる。
Key points
- Provenance
- Sources
- Transparency
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
Data lineage
A record of an asset's source, transformations, versions, and downstream uses.
Transparency
Clear information about AI use, purpose, limits, responsibility, and relevant evidence.