PRODUCTION SCENARIO
A company runs a customer-facing chatbot on a self-managed open source model outside Amazon Bedrock. The company wants the chatbot's prompts and responses checked by the same Amazon Bedrock guardrail its other applications already use, without moving the model to Amazon Bedrock.
Which approach will meet this requirement?
Answering here is anonymous. Nothing is saved unless you sign in.
Show answer and explanation
Answer: Call the ApplyGuardrail API with the prompt and response.
A guardrail can be used directly through the ApplyGuardrail API without invoking a foundation model, so the chatbot's prompts and responses can be evaluated while the model stays where it is. Attaching the guardrail to a knowledge base only covers traffic through that Bedrock resource. Custom Model Import would bring the model into Bedrock, which the requirements rule out. An inference profile routes Bedrock invocations across Regions and wraps nothing external.