Machine Learning Bias: Algorithms in the Courtroom

This case examines the complexities of using machine learning (ML) in judicial decision-making through the development and application of the Correctional Offender Management Profiling for Alternative Sanctions (COMPAS) tool for assessing recidivism risk. It contrasts two perspectives: one highlighting algorithmic efficiency and objectivity, the other raising concerns about bias, fairness, and ethical dilemmas. Students explore the ML development process, identifying nine types of biases and evaluating competing definitions of fairness, such as predictive parity and equalized odds. The case fosters understanding of ML, bias mitigation, and management responsibilities, equipping future managers to navigate AI’s societal and ethical challenges responsibly.