The Problem of AI
Emerging Governance Flashpoints in Artificial Intelligence
Artificial intelligence and new digital technologies are transforming society in profound ways. The same technologies advancing medicine, finance, and every industrial sector are creating generational disruptions to everything from work, the environment, and the creative economy to family life, the wellbeing of children, and what it means to be human in the age of AI. Difficult governance questions follow. Who owns creative works and culture? How do we keep humans in the loop of machine decisions with consequences for national security, privacy, speech, and human safety? Who is responsible when autonomous technologies cause harm and how can the most vulnerable members of society be protected? How can discriminatory bias be identified and mitigated in opaque technologies hidden from view? How can systems authenticate whether someone is a live human or whether a video or image is real or a deepfake? How should cyber governance evolve when hacks are no longer carried out by humans? What regulatory structures are necessary to promote innovation and productivity while addressing these societal concerns? How these many questions unfold will shape humanity for decades. Yet, regulatory frameworks around agentic AI, companion robots, chatbots, and nearly every other manifestation of artificial intelligence are lagging behind technological change. Students in this course will confront the major AI governance flashpoints of our time and analyze emerging frameworks of technology governance rising to address these concerns.
Introduction to Technology, Ethics, and Society
Emerging technologies provide new capabilities for improving our lives. But they can also pose novel and significant risks to our societies. For instance, artificial intelligence systems trained using machine learning techniques and enormous data sets are being deployed for a rapidly increasing number of important tasks. These range from the mundane, such as recommending movies and targeting advertising, to the crucial, as in hiring or even pre-trial detention decisions. These AI systems raise pressing new ethical issues. For instance, many of these systems display discriminatory bias. How can this bias be mitigated? Who should be accountable for ensuring that it is? Moreover, the effectiveness of machine learning depends on collecting huge amounts of information about people, i.e., “big data.” Who should be allowed to collect such data, and what data should be collected? What rights do individuals have to privacy and anonymity? The complexity of these technologies also poses novel problems of governance and accountability. What kinds of explanations are people owed for how an opaque AI system treats them? How can we hold developers and deployers of opaque systems accountable for their behavior? This course explores ethical questions like these that are raised by contemporary artificial intelligence and data-driven technologies. Readings will be drawn from philosophy, computer science, and other related fields. The goal is to prepare students to engage in critical ethical reasoning as developers, users, and stakeholders of new technology.
The Science of AI
The World of Data
Many aspects of our lives have been converted to data. Once our lives and the world around us are data-fied, the data can be converted into new forms of information and insight. In this class, we will learn about the basics of data science from data collection and cleaning to descriptive and predictive data analysis to visualization and storytelling. We will contemplate the insights we can learn about data and discuss principles for identifying and creating useful data that can be assessed to give insight into the world we live in. We will also discuss the limitations and biases associated with data and the ethical challenges that arise when using certain types of data. Through a data-centric lens, we will also learn the basics of Python programming.
Computational Language Processing
This course will introduce students to the basics of Natural Language Processing (NLP), a field that combines linguistics and computer science to produce applications, such as generative AI, that are profoundly impacting our society. We will cover a range of topics that form the basis of these exciting technological advances and will provide students with a platform for future study and research in this area. We will learn to implement simple representations such as finite-state techniques, n-gram models, and topic models in the Python programming language. Previous knowledge of Python is not required, but students should be prepared to invest the necessary time and effort to become proficient over the semester. Students who take this course will gain a thorough understanding of the fundamental methods used in natural language understanding, along with an ability to assess the strengths and weaknesses of natural language technologies based on these methods.
Math for Machine Learning
This course is designed to lay a strong mathematical foundation for students who wish to pursue AI or Data Science-related topics. The course will cover matrix algebra, differentiation of functions of many variables, integration on high-dimensional domains, basic problems of machine learning, such as linear regression, and the use of continuous optimization to solve them. Students will gain insight into why modern tools like Large Language Models (LLMs) work or fail in different circumstances. This course is suitable for math majors, math/statistics minors, as well as students from other departments.
The Applications of AI
Wicked Problems
This course examines how teams learn to work on wicked problems in an AI-augmented world. Wicked problems are complex challenges defined by uncertainty, competing values, and incomplete information, where progress depends as much on problem framing as on solution choice. Students work in peer and intergenerational teams on live, unresolved challenge problems that require careful problem articulation, structured planning, and defensible decision-making. The course follows a three-phase structure: foundational preparation, challenge lab work, and synthesis and reflection. Students begin with structured training in using AI systems and managing AI agents as part of human-AI teams. In the challenge lab phase, student teams work collaboratively—both with peers and in structured, time-bounded engagement with experienced practitioners—to frame problems, develop and iterate solution plans, and stress-test decisions under real-world constraints. AI is used throughout the course to support problem framing, option exploration, and stress-testing of plans in simulated environments. A distinctive feature of the course is short, intensive collaboration blocks with experienced executive practitioners, exposing students to real decision contexts and constraints and reinforcing learning through practice. Executive participants will engage in time-bounded collaboration with students to introduce real-world constraints and support intergenerational learning and workforce-relevant collaboration in challenge-based problem solving.
Ethics of AI and Health (also has The Problem of AI tag)
Artificial intelligence is re-shaping health and healthcare at a blistering pace. Doctors are using machine-learning algorithms to diagnose illnesses faster and more accurately than a human can. Smart devices are tracking and analyzing the personal health metrics of millions. Surgical teams are using augmented reality underpinned by AI algorithms to guide scalpels and increase precision. These advances hold both ethical opportunities and ethical challenges–from the promise of cheaper, more effective, more personalized care, and advancements in research and therapeutics to the perils of algorithmic bias, the loss or devaluing of human care, the cementing of health inequities, and the surveillance and exploitation of minoritized and vulnerable groups. This course will provide an introduction to critical issues in AI and Health Ethics that aims to equip students with key ethical concepts, theories, and frameworks to help navigate this complex emergent terrain. Particular foci will include AI developments in areas of physical, mental, sexual, and social health, and the course will canvass key readings in bioethics, political philosophy, and feminist and critical race theory.