Sonja Schmer-Galunder » Faculty Research Areas

Sonja Schmer-Galunder is the Glenn and Deborah Renwick Leadership Professor in AI Ethics and a Professor of Practice in the Herbert Wertheim College of Engineering at the University of Florida. She directs the Renwick Program for Safe, Ethical and Beneficial AI, housed in the Engineering Leadership Institute, and co-directs the NLP&Culture Lab.

Trained in the social sciences, she brings philosophical and anthropological frameworks to AI safety and alignment. Her research spans four areas:

• Alignment: cultural alignment, thick alignment and value pluralism
• AI Safety: monoculture and resilience, and correlated errors as a source of systemic failure
• AI Ethics: global AI ethics, non-Western philosophical frameworks, advising industry partners on interdisciplinary approaches
• AI Governance: tech diplomacy, global partnerships, the digital sovereignty of low-resource countries, and oral-tradition languages
• Multicultural & Multilingual NLP: cultural values and bias across languages, and language technologies for low-resource and oral-tradition languages

Before joining UF, she spent 14 years as a Principal Investigator and Principal Research Scientist at Smart Information Flow Technologies (SIFT), where she led government-funded research on data bias, community annotation studies, misinformation, hate-speech moderation, prosocial discourse and human-AI teaming. She also led several long-duration space-analog studies at the HI-SEAS habitat in Hawaii.

At UF she teaches AI Ethics for Technology Leaders and is launching the university’s first Global AI Ethics course, supported by a Provost’s initiative award. She is developing an interdisciplinary MS in Safe and Ethical AI and leads the Renwick Faculty Fellows program. Beyond UF, she works with international bodies on AI governance and serves on the AI Council of the Global Tech Diplomacy Network. The Renwick Program is an affiliated member organization of the International Association for Safe and Ethical AI (IASEAI).

She writes about anthropology, culture and AI in her monthly newsletter, Field Notes.
Selected Funded Research (as PI unless noted)

• U.S. Army VANESSA (2018–ongoing)
• DARPA Civil Sanctuary (2021–2024): LLM-based moderation of online hate speech and cultivating prosocial discourse
• DARPA HABITUS (Co-PI, 2019–2024): anthropological methods for culturally grounded AI, studying values in maternal-health decisions in Nigeria
• DARPA/AFRL TYBALT (2019–2022): AI approaches to detecting and countering harmful intent toward democratic societies, using Bandura’s theory of moral disengagement
• DARPA Understanding Group Bias (2018–2020): measuring cultural and gender bias in language models, with Stanford University
• DARPA MEDULLA (2017–2021) and OBLONGATA (2022–2024): collective allostatic load in small teams; four month-long space-analog missions at HI-SEAS
• NASA (Co-PI, 2016–2018): team composition strategies for long-duration space exploration, sub-award with the University of Hawaiʻi
• Defense Health Program NAPP (2015–2017) and USAMRAA CALM (2012): mobile health apps for sleep hygiene and anger management in military populations