An ecommerce team deployed a forecasting model to a SageMaker real-time endpoint for near real-time inventory management, but model accuracy degrades over time. They need a long-term automated approach to detect issues that should trigger retraining. Which solution meets this requirement?

The correct answer is to use Amazon SageMaker Clarify to monitor model and feature attribution bias to inform model retraining. SageMaker Clarify is designed for monitoring model quality, drift, and bias in production. It can detect data drift, concept drift, and feature attribution drift, which are common causes of model degradation over time. By monitoring these metrics, Clarify can trigger alerts or automated retraining workflows when performance issues arise due to data or concept shifts. SageMaker Debugger focuses on debugging training jobs, not monitoring deployed endpoints. AWS X-Ray is for application performance monitoring and distributed tracing, not model-specific quality or bias detection. SageMaker Ground Truth is for data labeling and dataset creation, not for ongoing model monitoring or retraining triggers.