Question

A company deploys a customer-facing chatbot on Amazon Bedrock. Leadership requires that the assistant never recommends or discusses specific named competitor products, and that any user attempt to make it do so is blocked with a standard refusal message. Which capability of Guardrails for Amazon Bedrock most directly enforces this requirement?

Accepted Answer

Denied topics and word filters that block defined subjects and specific terms from responses

Answer

Contextual grounding checks that verify responses are supported by retrieved source documents

Answer

Sensitive information filters that detect and mask personally identifiable information (PII)

Answer

Model evaluation jobs that benchmark accuracy against a labeled reference dataset