Brown CS News

“Artificial intelligence (AI) systems,” writes Brown CS faculty member Serena Booth, “increasingly influence both high stakes and everyday decisions across many sectors of the economy. However, providing clear and reliable instructions for intelligent systems is difficult even for relatively narrow applications. Failures occur because instructions are created by people, and human reasoning is shaped by limited information, context, and common cognitive mistakes. As AI becomes more widespread, improving how systems interpret human intent will be essential for safety and reliability.”

To address that challenge, she’s just received a National Science Foundation (NSF) CAREER Award to study how people communicate goals to machines and design AI systems that can interpret imperfect instructions by reasoning about the intent behind them. The expected outcomes include safer decision-making technologies and new tools that help organizations deploy AI more effectively. CAREER Awards are given in support of outstanding junior faculty teacher-scholars who excel at research, education, and integration of the two within the context of an organizational mission.

“This project,” Serena explains, “develops computational foundations for learning AI specifications from imperfect human input. The research integrates reinforcement learning, Bayesian inference, and computational cognitive modeling with empirical studies of human decision making to better characterize how people communicate goals and where specification errors arise.”

The work is organized around three research thrusts:

1. Modeling and Inferring AI Specifications develops probabilistic models of human reasoning that capture systematic specification errors and uses these models to enable AI systems to infer more accurate goals from flawed instructions.

2. Richer Inputs and Representations expands how AI systems learn from people by incorporating different forms of input such as preferences, demonstrations, explanations, gestures, and structured debate. New algorithms and elicitation interfaces will integrate these signals and resolve inconsistencies across modalities.

3. Personalization and Governance develops methods for learning multiple reward models that reflect differences in human preferences, enabling scalable personalization and avoiding one-size-fits-all objectives. In parallel, the project will develop educational programs that prepare students to design and govern AI systems. These activities include revising an undergraduate AI course to emphasize human decision-making in the design of AI systems and expanding the AI Policy Summer School to help build a national workforce that is fluent in both AI technology and public policy.

A member of the Brown CS faculty since 2025, Serena’s research focuses on how people communicate goals, preferences, and constraints to AI systems. Her work develops methods to help people design specifications for AI systems and robots while avoiding common problems such as misspecification, misuse, or unintended behavior. She’s also worked on AI policy issues addressing the use of AI systems in high-stakes applications such as housing and banking. 

Recently, she co-chaired ACM’s US Technology Policy Committee’s Subcommittee on AI and Algorithms, was named a Canadian Institute For Advanced Research (CIFAR) Azrieli Global Scholar, led a first-of-its-kind AI Policy Summer School to teach the nation’s top students to conduct CS policy work, and received a Reinforcement Learning Conference Outstanding Paper Award.

Serena joins numerous previous Brown CS winners of the award, including (most recently) Chen Sun, Ritambhara Singh, Peihan Miao, Vasileios Kemerlis, and Srinath Sridhar.

For more information, click the link that follows to contact Brown CS Communications Manager Jesse C. Polhemus.