Responsible AI practicesFull notesSummaryRingkasanStoriesPracticeAmazon Services and Tools for Responsible AI12 exam-style questions on this lesson.Question 1 of 12A company wants its Amazon Bedrock chatbot to refuse questions about investment advice, filter hate speech and violence, and mask customers' phone numbers. Which feature fits?AAmazon SageMaker ClarifyBAmazon SageMaker Model MonitorCGuardrails for Amazon BedrockDAWS AI Service CardsCheck answerQuestion 2 of 12A bank wants to check whether its loan approval model treats applicants differently by age or gender. Which tool fits?AAmazon Augmented AI (A2I)BSageMaker Role ManagerCSageMaker Model CardsDSageMaker ClarifyCheck answerQuestion 3 of 12A team found that one class in its training data is badly underrepresented. It wants to rebalance the data with random oversampling, undersampling, or SMOTE. Which tool fits?ASageMaker Data WranglerBSageMaker ClarifyCGuardrails for Amazon BedrockDSageMaker Model MonitorCheck answerQuestion 4 of 12A deployed model's accuracy slowly drops as customer behavior changes. The team wants automatic alerts when model quality deviates. Which tool fits?AAWS AI Service CardsBSageMaker Model MonitorCSageMaker Role ManagerDSageMaker ClarifyCheck answerQuestion 5 of 12A company wants low-confidence predictions from its document model to be sent to people for review. Which service fits?ASageMaker Model DashboardBGuardrails for Amazon BedrockCAmazon Augmented AI (A2I)DSageMaker ClarifyCheck answerQuestion 6 of 12A team wants to compare foundation models on subjective qualities such as friendliness, style, and brand voice, using its own reviewers. Which approach fits?ASageMaker Model MonitorBAutomatic evaluation in Amazon Bedrock using accuracy and toxicity metricsCAWS ArtifactDHuman evaluation with model evaluation on Amazon BedrockCheck answerQuestion 7 of 12An auditor wants documentation of a company's own model, including intended use, risk rating, and training details. Which tool fits?ASageMaker Model CardsBAmazon A2ICSageMaker ClarifyDAWS AI Service CardsCheck answerQuestion 8 of 12A customer wants to understand the intended use cases, limitations, and responsible AI design of Amazon Rekognition before adopting it. Which resource fits?ASageMaker Model CardsBAWS AI Service CardsCAmazon A2IDSageMaker Model DashboardCheck answerQuestion 9 of 12A data science team wants to explain which input features contributed most to each prediction of a tabular model. Which tool fits?ASageMaker Model MonitorBSageMaker Role ManagerCSageMaker ClarifyDGuardrails for Amazon BedrockCheck answerQuestion 10 of 12An administrator needs to quickly give ML practitioners only the minimum permissions they need in SageMaker. Which tool fits?ASageMaker Model CardsBAmazon A2ICSageMaker ClarifyDSageMaker Role ManagerCheck answerQuestion 11 of 12Which statement about Guardrails for Amazon Bedrock is correct?AGuardrails filter inputs and outputs at runtime without changing the modelBGuardrails retrain the model to remove harmful knowledgeCGuardrails replace the need for any other security controlsDGuardrails only work with Amazon Titan modelsCheck answerQuestion 12 of 12A team wants one place to track the behavior of all its models in production. Which tool fits?AAWS AI Service CardsBSageMaker Model DashboardCSageMaker ClarifyDAmazon A2ICheck answerCore Dimensions of Responsible AI10 exam-style questions on this lesson.Responsible Considerations to Select a Model10 exam-style questions on this lesson.