AWS Certified AI Practitioner AIF-C01 Practice Question

Amazon SageMaker Model Monitor captures a sample of the live requests and predictions at a hosted endpoint, compares their statistical distribution with a baseline calculated from the training data, and publishes CloudWatch metrics so alerts can be triggered when drift or other data-quality issues appear. SageMaker Clarify provides bias detection and explainability tools-including bias-drift and feature-attribution-drift monitoring when integrated with Model Monitor-but it is not the primary service for general data or model-quality drift detection. SageMaker Autopilot automates model building and tuning but does not watch deployed endpoints. SageMaker Ground Truth Plus delivers managed data labeling rather than model-quality surveillance.