Assess graduate-level understanding of principles and trade-offs in designing explainable AI for clinical decision support, including explainability methods (intrinsic vs post-hoc, global vs local), impacts on model accuracy, calibration and uncertainty, clinician trust and usability, evaluation metrics and validation, regulatory and ethical considerations, patient safety, and workflow integration; tasks include analyzing case scenarios, evaluating trade-offs, and proposing design choices that balance interpretability, performance, and safety.