CS6603: AI, Ethics, and Society

This course examines how AI and ML systems — already used by industry, government, and schools to make consequential decisions — can encode and amplify human bias, and what responsibility computing professionals have for those outcomes. It combines statistics fundamentals with applied fairness techniques across word embeddings, facial recognition, and predictive policing algorithms, using tools like AI Fairness 360 and Google’s What-If Tool. The course text is Weapons of Math Destruction by Cathy O’Neil.

Final Project — Bias Mitigation

A capstone project applying the semester’s fairness and bias-mitigation techniques to transform a biased real-world dataset into a more equitable one.

Written Critiques

Short analytical papers arguing a position on AI ethics case studies, including the ethics of autonomous-vehicle decision-making and a critique of bias-detection results produced using Google’s What-If fairness tool.