Insert a user request into an internal FAQ prompt.
Prevent the request from changing safety rules or permissions.
Choose two appropriate implementations.
Separate trusted instructions from untrusted input and keep authorization outside model-generated text.
Detailed explanation
Hidden prompts do not stop malicious input or output.
A trusted server-side control must enforce permissions.
Text input has no authority.
Untrusted text must not override system policy.
Check official documentation
Compare the following points with the official documentation.
AWS公式生成AIセキュリティとAIF-C01 Domain 2.3のプロンプトインジェクション対策を確認する。
Official references
What to confirm
役割分離とサーバー側認可の必要性を説明できる。
Key points
- Trust boundary
- Role separation
- Authorization
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
Trust boundary
A boundary separating controls and instructions that are trusted from data supplied by users or external sources.
Prompt injection
An attack that places instructions in input or retrieved content to alter a model's intended behavior.