Madeleine Grunde-McLaughlin
Hi! I am a final-year Ph.D. student at the intersection of Human-Computer Interaction and Artificial Intelligence at the University of Washington. I am advised by Jeff Heer and Dan Weld, and I have collaborated closely with Ranjay Krishna. I have interned with the Google Applied Sciences team under Dan Liebling and with AI Frontiers under Adam Fourney and Saleema Amershi. I received my Bachelor’s degree in Cognitive Science from the University of Pennsylvania.
Most tasks given to AI are underspecified, meaning that there are multiple possible approaches that are valid under the task conditions. As a result, current AI systems incorporate implicit assumptions into their outputs, assumptions that may contradict users’ personal context and preferences. I evaluate and design systems to improve human verification of AI’s implicit assumptions and enable steering towards preferred alternatives.
Broadly, I work to evaluate AI behavior and effectively communicate its capabilities to people in order to improve decision-making and retain human agency. I often study the data analysis domain and am particularly interested in supporting scientists in environmental applications.
My research has informed the design of Microsoft products and been included in university course listings across the United States. I publish in leading HCI and AI venues, including TOCHI, CHI, CSCW, and CVPR, and have received two Best Paper Honorable Mention awards.