Ethical AI for Decisions in Today’s World
Learn how to design, evaluate, and govern AI systems that make fair, transparent, and accountable decisions in high-stakes real-world settings.
Ethical AI for Decisions in Today’s World
Learn how to design, evaluate, and govern AI systems that make fair, transparent, and accountable decisions in high-stakes real-world settings.
Discover how ethical principles translate into concrete design choices when building and deploying artificial intelligence (AI) systems. In this course, you'll learn to map any AI application as a data–model–decision pipeline, identifying where bias, harm, and unintended consequences can enter at each stage. You'll explore why unfair outcomes are often predictable—rooted in incomplete data, historical proxies, or feedback loops that amplify inequities over time.
What you'll learn
- Map any AI system as a data–model–decision pipeline and pinpoint where ethical risks, bias, and harms can emerge at each stage
- Diagnose common data and model issues and assess their impact on different groups and stakeholders
- Apply practical mitigation strategies, including guardrails, domain-informed data repairs, and uncertainty-aware decision support like calibration and conformal intervals
- Formulate value-aligned objectives and constraints, navigate competing goals through trade-off reasoning, and communicate ethical limitations transparently
Meet your instructors
-
Associate Professor
Swati Gupta
Swati Gupta
Associate Professor
Swati Gupta is the Class of 1947 Career Development Associate Professor and an Associate Professor at the MIT Sloan School of Management in the Operations Research and Statistics Group.
Her work focuses on deep theoretical questions in optimization and AI—many of which are grounded in real-world challenges—striving to bridge the gap between rigorous theory and meaningful practice. She broadly works on (i) static and dynamic algorithms for incomplete and erroneous data, (ii) multi-criteria decision-making using optimization and learning, and (iii) bridging discrete, continuous, and quantum optimization. Her work spans various domains such as hiring, admissions, districting, e-commerce, platforms, supply chains, healthcare, quantum optimization, and energy. She frequently interacts and collaborates with industry, as well as with doctors, physicists, law and policy scholars. She has led various high-impact projects, including serving as the lead of Ethical AI at the NSF AI Institute on Advances in Optimization from 2021-2023.
Swati currently serves as an associate editor of the Open Journal of Mathematical Optimization. She has served on the technical program committees for top conferences in optimization, machine learning, and algorithmic fairness, e.g., Integer Programming and Combinatorial Optimization (IPCO 2026, 2024), Neural Information Processing Systems (NeurIPS 2023, area chair), and ACM Conference on Fairness, Accountability and Transparency (FAccT 2022, area chair).
Swati received a PhD in operations research from MIT in 2017 and a joint Master’s and Bachelor in Technology in computer science from IIT Delhi in 2011.