Sonja Schmer-Galunder

Glenn and Deborah Renwick Leadership Professor in AI and Ethics
Professor of Practice

Bio

Sonja Schmer-Galunder the Glenn and Deborah Renwick Leadership Professor in AI and Ethics at the University of Florida’s Department of Computer & Information Science & Engineering (CISE). She is also a member of the Florida Institute for National Security (FINS) and the Working Group in AI Ethics and Policy.

Prior to joining UF, she was a Principal Investigator and Research Scientist at Smart Information Flow Technologies, where she was working on socio-technical AI research projects for the U.S. government. Her research is concerned with values and beliefs embedded in language and AI systems, moral reasoning and justifications of harmful intentions towards social groups, as well as the use of LLMs to mitigate harm and polarization and foster prosocial behaviors and empathy. She led several projects on gender bias, misinformation, moderation of hate speech, prosocial discourse and use of politeness in human and human-AI interactions (e.g. PI on DARPA Civil Sanctuary, DARPA Understanding Group Bias and co-PI on DARPA Habitus). Leveraging her background in social anthropology, she informs socio-technological innovations by focusing on their cultural, social, and ethical impacts. She aims to prevent harm from AI by fostering a deep understanding of people, culture and societies, their values and norms, and their ideology and demographic diversity.

She is interested in collective behavior and collective intelligence, ranging from small teams in extreme environments to larger groups tied to ideology. She has led several space simulation studies at the HI-SEAS research station in Hawaii, where she collected behavioral and physiological data from gender-balanced teams of six individuals who were living and working in confinement ranging from 1 to 12 months to better understand collective allostatic load expressed in cognition, physiology, and social behavior, and developed ideas for human-AI teaming to optimize performance, health and well-being (e.g. PI for Phase 1, 2 and 3 SBIRs for research on Collective Allostatic Load, Effective Human Teaming Supported by Social Sensing and Force Protection in the Online Information Environment).

She is currently developing an AI Ethics program for CISE focusing on 1) Understanding Social and Ethical values and how they are reflected in AI systems, 2) Privacy, Consent and Autonomy and 3) AI-enabled Democratic Collective Decision Making.

Research Areas

Human-Centered Computing

Education

Masters, Social Anthropology, 2004, Lunds University, Sweden

Publications

View a full list Here