A retrieval chatbot executed an instruction contained in an uploaded document and exposed sensitive information.
Users can upload documents.
Which description is most accurate?
Prompt injection occurs when untrusted input contains instructions that try to override trusted instructions or cross a data boundary.
Detailed explanation
The attack mixes data with instructions and attempts to change the model's intended behavior.
The attack mixes data with instructions and attempts to change the model's intended behavior.
The key issue is treating untrusted content as an instruction.
The key issue is treating untrusted content as an instruction.
This is an input and instruction-boundary risk, not storage availability.
This is an input and instruction-boundary risk, not storage availability.
The behavior is a security attack, not a benign prompting technique.
The behavior is a security attack, not a benign prompting technique.
Try it yourself
An example you can run in a temporary verification environment.
AWS公式AIF-C01ガイドのDomain 2.3とAmazon BedrockのPrompt management・Guardrails関連説明を確認する。Expected result
信頼できない入力とシステム指示を分離し、権限・出力検査・ガードレールを設計する必要を説明できる。Key points
- Untrusted input
- Instruction priority
- Permission separation
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
- Command output formatting can vary slightly by distribution or tool version.
- Run the example in a temporary directory or process when possible.
Foundation review
Read the scope first
Check whether the command acts on the current shell, a new process, an existing process, or a file.
Verify the observable result
Use the supplied command and compare the output with the expected result.