Responsible Artificial Intelligence.

Catalog description

Students learn principles, tools, and hands-on practices for designing and evaluating AI systems responsibly. Topics may include reliability and calibration, robustness testing, dataset and model documentation, fairness concepts and metrics with bias mitigation, explainability for technical and stakeholder review, privacy-by-design and responsible data handling, governance and risk management, lightweight red-teaming, and post-deployment monitoring. Using interactive online modules and labs with common libraries, students produce audit-ready artifacts and apply assurance workflows to machine learning and large language models use cases. This course prepares learners to translate AI models into systems that are safe, fair, transparent, and accountable in real-world settings. Prerequisite: Approval by the instructor or the CS graduate advisor.