University of Utah Awards Nearly $400K for First-Ever Responsible AI Seed Grants
March 5, 2026
The University of Utah One-U Responsible Artificial Intelligence Initiative (One-U RAI) will award its first round of seed grants, totaling $386,855, to nine research teams advancing AI tools and uncovering new insights on AI’s societal impacts. Projects will empower patients and their families, shed light on AI’s role in the classroom, and help people better understand environmental exposures that harm their health.
From a highly competitive pool of 40 proposals, One-U RAI committee members selected nine projects that reflect a balance of disciplines and initiative thematic areas—environment, health care and wellness, and teaching and learning. Awardees span 17 departments from across campus.
“The novelty and overall quality of applications and their potential impacts are impressive and a testament to the breadth and depth of applied AI research at the U,” said Manish Parashar, executive director of the Scientific Computing and Imaging Institute, which leads One-U RAI. “We are excited to serve as a springboard for these nine projects and look forward to results that directly benefit patients, students, Mountain West residents, and beyond.”
The seed grant program is geared toward projects with budgets up to $50,000 and strong potential for real-world impact. Teams must include at least one of 49 people on campus affiliated with One-U RAI as a current committee member, faculty fellow, or postdoctoral fellow mentor. Evaluation criteria included scholarly merit, innovation, interdisciplinary collaboration, and likelihood of attracting future external funding. Funds will be awarded on April 1, and projects are expected to be completed in one to two years.
In addition to seed grants, the $100 million One-U RAI funds cluster hires and faculty, postdoctoral, and visiting fellowships. Beyond growing interdisciplinary AI research at the U, the initiative strengthens cyberinfrastructure and promotes ethical AI use and development across the state and region, including through the community consortium.
AI Literacy in Undergraduate Biology: Student Domain Knowledge and Responsible Practice
Principal Investigator (PI): Frederick Adler, Mathematics, Biology
Co-PI: Sara Yeo, Communication
Thematic Areas: Environment; Health Care and Wellness; Teaching and Learning
This project examines how biology students engage with AI-powered tools and investigates the relationship between domain-specific knowledge and AI literacy. Using survey and focus group data, it evaluates students’ ability to critically assess, verify, and integrate AI-generated content within established scientific practices. The findings will inform development of an integrative pedagogical model that embeds responsible and ethical AI use within biological sciences education.


The Walkables Project
PI: Brian Baucom, Psychology
Co-PIs: Daniel Mendoza, Atmospheric Sciences; Simon Brewer, Environment, Society, and Sustainability; Nina de Lacy, Psychiatry
Thematic Areas: Environment; Health Care and Wellness
The team will investigate how environmental conditions affect sleep, mood, and stress by using wearable sensors to gather detailed, real‑time data from individuals. The researchers will build explainable AI models to predict when someone may be at higher risk for poor neuropsychiatric health and identify which types of sensor information matter most. The pilot will lay the groundwork for a larger research program aimed at improving early detection of mental‑health risks.




Extending NBSchat With Responsible AI: A Hybrid Scripted–Generative Chatbot to Support Parents Receiving Sickle Cell Trait Newborn Screening Results
PI: Karen Eilbeck, Biomedical Informatics
Co-PIs: Guilherme Del Fiol, Biomedical Informatics; Kimberly Kaphingst, Communication; Katherine Sward, Nursing, Biomedical Informatics
Thematic Areas: Health Care and Wellness; Teaching and Learning
This project will expand NBSchat, a scripted chatbot Eilbeck’s team developed to support parents whose newborns screen positive for sickle cell trait. The updated system will allow parents to ask their own questions—rather than limiting them to predefined questions—while ensuring information remains safe and accurate. The project will explore how generative AI can improve personalization and support during a stressful time. Findings will inform best practices for using AI in sensitive public health communication and position the team for future funding.




Safe and Effective Integration of AI-Based Seizure Detection into Neonatal Intensive Care
PI: Jennifer Keene, Pediatrics
Co-PIs: Amanda Sandoval Karamian, Pediatrics; Sabrina Malone Jenkins, Pediatrics; Vineet Pandey, Computing
Thematic Area: Health Care and Wellness
Dr. Keene and her team will examine whether an AI tool can help electroencephalography (EEG) technologists identify newborn seizures—rarely visible to the human eye—faster and more accurately without increasing harmful false alarms. Building on preliminary analysis of 15,000-plus hours of EEG data, the study examines how AI and technologists could function together in real-time clinical care. Findings will guide the safe and responsible deployment of AI in neonatal neurocritical care and inform future clinical trials.




Co-Designing a Responsible AI Framework for Equitable Diabetes Self-Care in Low-Income Older Adults
PI: Jisook Ko, Nursing
Co-PI: Vineet Pandey, Computing
Thematic Area: Health Care and Wellness
This project focuses on creating a responsible AI framework to support type 2 diabetes self-management among low-income older adults through a 12-month community co-design process. By integrating health data and embedding principles of transparency, fairness, and accessibility into system design, the team will generate validated protocols and prototype specifications. The work establishes a replicable model for equitable AI deployment and will set the stage for competitive federal funding proposals.


Conversational AI Agents as Personalized Teaching Assistants
PI: Bei Wang Phillips, Computing
Co-PI: Rob MacLeod, Biomedical Engineering
Thematic Area: Teaching and Learning
Phillips and MacLeod will develop EduPilot, a conversational agentic AI teaching assistant, to integrate real-world computational workflows into biomedical engineering education. By translating natural language prompts into multi-step data analysis and visualization tasks, EduPilot lowers technical barriers while reinforcing conceptual understanding and scientific reasoning. Deployment in two courses will advance responsible AI integration in STEM education, strengthen workforce preparation, and promote reproducible computational practice.


Responsible AI-Driven Model Development for Proactive Household Environmental Exposure Assessment Using Low-Cost PM2.5 Sensors
PI: Darrah Sleeth, Occupational and Environmental Health
Co-PIs: Yue Jiang, Computing; Scott Collingwood, Pediatrics; Nancy Daher, Occupational and Environmental Health; Katherine Sward, Nursing, Biomedical Informatics
Thematic Area: Environment
Sleeth’s team will develop a responsible AI monitoring framework that transforms sensor data on indoor air quality into actionable, real-time exposure alerts without using audio, video, or personal identifiers. Through laboratory simulations and in-home testing, the team will train and validate machine learning models to detect household pollution from everyday activities, such as spraying a cleaning solution or lighting a candle. The resulting privacy-preserving system aims to empower individuals to proactively manage indoor air quality and advance environmental public health.




Responsible AI for Rare Clival Tumors: Exact Patient-Level Unlearning Without Retraining
PI: Tolga Tasdizen, Electrical and Computer Engineering
Co-PI: Tyler Richards, Neuroradiology
Thematic Area: Health Care and Wellness
Tasdizen and Richards will develop an unlearning-capable MRI model to improve diagnosis of rare skull-base tumors while making patient consent withdrawal practical and auditable. By separating a shared backbone from removable patient-specific “plug-in” updates, the model can exclude withdrawn data without full retraining. The approach will be evaluated for tumor segmentation, classification accuracy, and measurable privacy protection, establishing a framework for adaptable and accountable medical AI.


The Effects of Customized AI Course Support on Cooperative Learning in College Classrooms
PI: Peilu Zhang, Population Health Sciences
Co-PIs: Pavitra Govindan, Economics; Bernardo Modenesi, Population Health Sciences
Thematic Area: Teaching and Learning
As AI becomes integrated into classrooms, there is little evidence on how these tools affect peer-to-peer cooperative learning—a common educational practice with established benefits. This project will evaluate how a custom AI assistant impacts students working in groups in a large introductory economics course. Through a randomized controlled trial that compares group discussions with and without AI support, the study will assess effects on learning outcomes, stress, engagement, and attitudes toward collaboration. Findings will guide the pedagogically grounded design of AI tools in higher education.


