Assess students' ability to evaluate and balance fairness, accuracy, and privacy trade-offs in supervised machine learning for high-stakes decision-making. Scope includes identifying sources of bias, selecting and interpreting fairness metrics (e.g., demographic parity, equalized odds, calibration), measuring model accuracy and utility, applying privacy-preserving methods (differential privacy, federated learning), designing evaluation experiments and audits, quantifying trade-offs, and proposing technical and policy-aligned mitigations based on case studies and stakeholder impact analysis.