Kassem Fawaz
Associate Professor, University of Wisconsin–Madison Department of Electrical & Computer Engineering
Talk Title
Exploring LLMs for Privacy-Aware Social Companion Robots
Abstract
Social robots are embodied agents that engage with people following human norms of communication. They listen and speak with people, interact using non-verbal cues, and share the physical environment with them. Without privacy awareness, social robots cannot meet user expectations regarding how they collect, process, and share information in their operating environment. For example, a social robot can share information from group interactions with other family members, occupants, or visitors of the home. In three parts, this talk discusses our current work at establishing design principles for privacy-aware social robots. The first part describes our analysis of family preferences when sharing access to autonomous agents, such as ChatGPT. The second part discusses our efforts to understand whether recent advances in large language models (LLMs) can enhance privacy awareness in robots. The third part discusses our ongoing work on co-designing approaches to signal and express privacy awareness to social robots.
Bio
Kassem Fawaz is the Grainger Institute of Engineering Associate Professor in the Electrical and Computer Engineering department at the University of Wisconsin–Madison, where he serves as the inaugural associate chair for research. He earned his Ph.D. in Computer Science and Engineering from the University of Michigan. His research interests include the security and privacy of user interactions with AI-powered systems. He was awarded the Caspar Bowden Award for Outstanding Research in Privacy Enhancing Technologies in 2019. He also received the National Science Foundation CAREER award in 2020, the Google Android Security and PrIvacy REsearch (ASPIRE) award in 2021, the Facebook Research Award in 2021, the Chancellor Teaching award in 2022, and the Vilas Associates Award in 2024. His research has been funded by the National Science Foundation, the Federal Highway Administration, and the Defense Advanced Research Projects. His work on privacy has been featured in several media outlets, such as the BBC, Wired, the Wall Street Journal, the New Scientist, and ComputerWorld.