Assess students' ability to evaluate trade-offs between model interpretability and predictive performance in AI-driven clinical decision support systems, covering interpretability methods (feature importance, rule-based models, LIME/SHAP), performance metrics (AUC, calibration, sensitivity/specificity), clinical utility and safety, fairness and regulatory/ethical considerations, and practical strategies (model selection, hybrid approaches, validation, monitoring). Quiz tasks include comparing model options in clinical scenarios, justifying trade-off decisions based on patient outcomes and stakeholder needs, and proposing validation/communication plans to balance transparency and accuracy.