A public chatbot receives harmful requests and may expose sensitive information.
Apply controls to both model inputs and outputs.
Which Amazon Bedrock feature best matches this requirement?
Amazon Bedrock Guardrails can apply content, denied-topic, and sensitive-information controls to generative-AI inputs and outputs.
Detailed explanation
Guardrails can evaluate and control model inputs and outputs using configured policies.
Guardrails can evaluate and control model inputs and outputs using configured policies.
Route 53 provides DNS rather than model input/output safety controls.
Route 53 provides DNS rather than model input/output safety controls.
Retention is a storage setting, not real-time model I/O control.
Retention is a storage setting, not real-time model I/O control.
A dashboard visualizes metrics but does not reject or transform model input/output.
A dashboard visualizes metrics but does not reject or transform model input/output.
Try it yourself
An example you can run in a temporary verification environment.
Amazon Bedrock Guardrailsの公式ドキュメントで、入力・出力評価とポリシー設定を確認する。Expected result
監視と、モデル入出力を実際に制御する機能を区別できる。Key points
- Guardrails
- Input and output controls
- Sensitive-information protection
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.