$20 Million Gift from Thomas Francis Dunn, AB’81, MBA’86 and Susan Knapp Dunn, AB’82, Launches a New Lab at the Harris School to Advance Emerging Field of Algorithmic Public Policy

With an extraordinary gift from University of Chicago Trustee Thomas Francis Dunn, AB’81, MBA’86, and Susan Knapp Dunn, AB’82, the University of Chicago Harris School of Public Policy is establishing the Bike Shop @UChicago, a new lab dedicated to building and leading the emerging field of algorithmic public policy.

The usual vision for AI is to imagine an algorithm that can do what people can do. This new initiative instead seeks to help solve some of society’s most pressing problems by designing and scaling algorithms that can do things people cannot, algorithms that enhance human capacity rather than simply automate. Inspired by Steve Jobs’ description of computers as “bicycles for the mind,” the Bike Shop @UChicago’s goal is to build tools that expand the reach, consistency, and effectiveness of human decision-making for social impact.

The Bike Shop @UChicago is guided by two complementary ambitions. Its scientific mission is to help steer the development of new AI technologies through foundational research and by training the next generation of scholars working at the intersection of AI, public policy, and behavioral science. Its social impact mission is to translate those advances into scalable tools—interventions that governments can adopt and deploy at a scope and speed that traditional policy approaches often cannot achieve.

Led by Jens Ludwig, the Edwin A. and Betty L. Bergman Distinguished Service Professor at the Harris School and Pritzker Director of the University of Chicago Crime Lab and co-director of the Education Lab, the Bike Shop @UChicago will serve as a hub for research, education, and field-building at the intersection of public policy, data science, and behavioral science. The work will be carried out in close partnership with a sister initiative, the Bike Shop @MIT, led by Prof. Sendhil Mullainathan, Ludwig’s longtime collaborator.

What Is Algorithmic Public Policy?

Algorithmic public policy uses algorithms in two complementary ways. First, it combines human judgment with machine learning and data-driven tools to improve complex decisions—such as risk assessment, resource allocation, and service delivery—where evidence shows that systematic improvements are possible. Second, it uses algorithms to reduce the costs governments face in implementing proven policies at scale, making it easier to adopt approaches that research has shown to be effective.

“We are tremendously grateful to Tom and Susan Dunn for supporting this new field of study, for which our society has an urgent need,” said University of Chicago President Paul Alivisatos. “Their generosity strengthens the scholars, ambitious ideas, and innovative work that will define a new field and ensure that advances in AI and data science translate into trustworthy public systems, helping government deliver results and improved outcomes for millions of people.”

“UChicago and the Harris School have a long-standing strength in rigorous, evidence-based public policy and in asking hard questions about how government can work better,” said Mr. Dunn. “We wanted to help build on that foundation by supporting the scaling of cutting-edge research on algorithms into solutions public servants can actually put to work.”

Why Now?

Governments spend far more on social programs than philanthropic organizations or the private sector do, yet they often struggle to identify, adopt, and sustain the most effective approaches. Traditional strategies are often slow, expensive, and difficult to scale.

Recent advances in artificial intelligence and machine learning have created a genuine inflection point. For the first time, new technologies make it possible to pursue two goals simultaneously: extending human judgment in complex policy decisions and dramatically lowering the cost of turning evidence-based insights into operational systems governments can actually use.

Taken together, these advances make it feasible to both rethink how policy decisions are made and scale what works far beyond isolated pilots. The Bike Shop @UChicago is designed to seize this moment—building the scientific foundations of algorithmic public policy while translating that work into deployable tools with real-world impact.

“The Bike Shop @UChicago will define algorithmic public policy by solving a central challenge in evidence-based governance: getting what works to extend beyond the lab,” said Ethan Bueno de Mesquita, dean and Sydney Stein Professor of the Harris School. “Our goal is to turn breakthrough research into scalable exemplars—tools governments can adopt, learn from, and replicate.”

How the Bike Shop Works

The Bike Shop @UChicago will pursue two core lines of work: advancing research that makes new types of algorithmic tools possible (“bicycles of the mind” for public policy), and deploying machine-learning interventions that deliver proven impact at scale.

Examples of initial “bicycles for the mind” include:

• Reimagining Public Safety: Led by Oeindrila Dube, the Philip K. Pearson Professor at the University of Chicago, this project asks: What if it were possible to substantially improve the outcomes of American policing by combining AI with insights from behavioral science to help officers navigate the countless high-stakes decisions they have to make every shift?
• Personalizing Education: Led by MIT’s Mullainathan, this project asks: Could we transform education by using AI to build teachers a map of student confusion that allows for a more personalized and effective pedagogy?

Examples of machine learning interventions designed for scale include:

• Reforming Municipal Finance: This project, led by Christopher Berry, the William J. and Alicia Townsend Friedman Professor and Director of the Mansueto Institute for Urban Innovation at UChicago, asks: What if we could use machine learning algorithms to scale dramatic improvements to the efficiency and fairness of how cities across the U.S. collect billions of dollars each year from their citizens?
• Transforming Climate Change Adaptation Efforts: Led by UChicago Assistant Professor Amir Jina, University Professor Michael Kremer, and Associate Professor Pedram Hassanzadeh, this project asks: What if we could scale AI weather-forecasting technologies to provide decision-relevant information to vulnerable populations, like rural farmers, that could enhance resilience to climate change and impact income levels and volatility in developing countries around the globe?

Educating the Field and Looking Ahead

Beyond research, the Bike Shop @UChicago will train the next generation of scholars and practitioners shaping this field. It will support student research assistantships and internships, develop new courses in algorithmic policy methods, host summer institutes and conferences, and convene researchers and government partners to accelerate learning and adoption.

Designed to both build and grow alongside the field itself, the Bike Shop @UChicago will expand through additional partnerships and support—strengthening technical infrastructure, deepening government integration, and advancing promising ideas from prototypes to tools embedded in public systems.