QuestionQ288

Security, Compliance, and Governance for AI Solutions

A company deploys a foundation model (FM). The company observes that the FM produces answers to user-submitted questions about politics. The company needs to ensure that the model does not send responses to political questions to users.

Which AWS solution meets this requirement?

  • A Amazon Bedrock Guardrails
  • B Amazon Bedrock Agents
  • C Amazon SageMaker Clarify
  • D Amazon SageMaker Model Monitor
Explanation

Amazon Bedrock Guardrails can define politics as a denied topic and block detected content in both user queries and model responses, returning blocked messaging instead of the model’s political answer.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!