From Health to Climate to Classrooms: Meet 13 New Faculty Fellows Putting Responsible AI to Work for Utah
October 8, 2026
The University of Utah One-U Responsible AI Initiative at the Scientific Computing and Imaging Institute this week announced its third cohort of faculty fellows, totaling 13 researchers across 11 departments. The initiative’s largest and most interdisciplinary cohort to date, the fellows will advance practical, people-centered uses of AI—from personalized health care and urban planning to climate research and AI-enabled learning.
Read more about faculty fellowships and other opportunities from the One-U Responsible AI Initiative.
Amir Arzani
Associate Professor; Department of Mechanical Engineering, John and Marcia Price College of Engineering; Scientific Computing and Imaging Institute
Thematic Areas: Health Care and Wellness; Teaching and Learning
Amir Arzani’s work explores how AI can responsibly help doctors plan personalized treatments and help students learn. He is developing agentic AI tools that use digital twin models of a patient’s body to explore treatment options (for example, how to deliver effective cancer treatment while protecting healthy tissue). His goal is to make these technologies easier to use and accessible in clinics and Utah industry. In the classroom, Arzani’s goal is to create AI-enabled virtual labs where students use agentic AI to explore engineering problems, test new ideas, and learn to question and verify AI’s results.
Daniel Brown
Assistant Professor; Kahlert School of Computing, John and Marcia Price College of Engineering
Thematic Areas: Health Care and Wellness; Teaching and Learning
Daniel Brown is a member of the Utah Robotics Center and director of the Aligned, Robust, and Interactive Autonomy (ARIA) Lab. His research lies at the intersection of AI, human-computer interaction, and robotics. His work explores how robots and other AI systems can learn what people want, make safe and interpretable decisions, collaborate with humans and other AI systems, and improve through experience. As a faculty fellow, he will develop and study assistive robots for hospitals and homes, focusing on improving transparency and trust, learning individual user’s preferences, and understanding how people and robots adapt to one another over time to improve independence, health, and well-being while reducing the burden on caregivers and health care workers.
Edward DiBella
Professor; Department of Radiology, Spencer Fox Eccles School of Medicine
Thematic Area: Health Care and Wellness
Ed DiBella develops AI techniques to improve medical imaging. He combines AI approaches with raw MRI data to reduce scan time for patients and generate higher-quality images. DiBella is also working with colleagues and predoctoral students to build a better database of MRI images of the left atrium in people with atrial fibrillation. Such databases allow for widespread comparison of AI methods to improve image quality and clinical understanding. Beyond his MRI efforts, DiBella is collaborating with faculty in computing, biomedical engineering, and the Scientific Computing and Imaging Institute to establish a predoctoral training program focused on clinical problem-solving using responsible AI.
Dusti R. Jones
Assistant Professor; Department of Family Medicine and Public Health, Spencer Fox Eccles School of Medicine; Huntsman Cancer Institute
Thematic Area: Health Care and Wellness
As a faculty fellow, Dusti Jones will study how AI can create personalized health videos that provide support when people are facing challenges such as quitting smoking or reducing alcohol use. Rather than delivering the same message to everyone, these videos would adapt to a person's current situation and needs. Jones will evaluate whether AI-generated content improves engagement and health outcomes while also examining important questions about trust, transparency, bias, and unintended consequences. The project will also help develop tools and infrastructure that allow researchers to create and evaluate AI-generated health interventions responsibly. The overall goal is to establish evidence-based guidelines for using AI safely and effectively to support health behavior change.
Makoto Kelp
Assistant Professor; Department of Atmospheric Sciences, College of Science
Thematic Area: Environment
Makoto Kelp’s research focuses on AI safety in Earth science, including weather, climate, and air pollution. Systems such as AI foundation models learn from vast amounts of data and can outperform conventional forecasting models, but their reliability under changing environmental conditions remains uncertain. For example, a model may forecast everyday air quality well but struggle during severe wildfire smoke or dust events. Kelp investigates what these AI models learn about the atmosphere and how that knowledge drives their predictions. His goal is to build scientific interpretability and control into these systems so we can understand how they work and guide their behavior. He is developing evaluation standards and governance frameworks to help scientists and public agencies use them responsibly in decisions affecting public health and the environment.
Thomas Kraft
Assistant Professor; Department of Anthropology, College of Social and Behavioral Science
Thematic Areas: Health Care and Wellness; Environment
Thomas Kraft’s research investigates how anthropological photographs and videos can be processed using AI to solve complex problems related to health and aging in understudied populations. This work is cross-species (including humans and closely-related primates), spans current and historical time periods, and is highly interdisciplinary. Examples of ongoing projects in Kraft’s lab include: 1) the development of obesity prediction models that generalize across diverse human populations; 2) the generation of photo-based, non-invasive biomarkers of health and aging in wild primates to inform veterinary interventions; and 3) cross-cultural investigation of the biomechanics and kinematics of human resting postures over the past century. The goal of Kraft’s work is to accelerate ethical AI research in the field of anthropology, and contribute to applied topics ranging from conservation biology to telehealth.
Himanshu Mishra
Professor; Department of Marketing, David Eccles School of Business
Thematic Area: Teaching and Learning
As AI integration in curricula expands, higher education is being asked to make curricular decisions faster than the evidence to guide them is developing. We still know little about whether different forms of integration produce different learning experiences, whether AI supports or displaces disciplinary learning, or which approaches instructors find sustainable. A review of syllabi shows that instructors converge on one principle—that AI output is provisional and must be verified—but differ in how they approach it: some ask students to critique or audit AI-generated work, others create with AI and document their process, some treat prompting as an assessed skill, and some use AI as a learning assistant. Mishra’s goal is to examine which practices faculty found most useful, how practices and outcomes varied across disciplines, and which practices most improved student learning.
Yuree Noh
Assistant Professor; Department of Political Science, College of Social and Behavioral Science
Thematic Area: Teaching and Learning
Noh studies the gap between what people say and what they really think. In authoritarian countries, people may give the answer they think is expected. One of her projects pairs AI with human reviewers to measure censorship and surveillance worldwide, and tests whether the AI consistently misjudges particular countries or groups. She examines a similar gap in classrooms, where AI lets students turn in polished work without building the reasoning behind it. Another project will compare students who design surveys without AI, with AI that hands them answers, and with AI that flags a problem (e.g., privacy risks) and asks them to explain it. The goal is to learn which approach builds independent thinking.
Bei Wang Phillips
Associate Professor; Kahlert School of Computing, John and Marcia Price College of Engineering; Scientific Computing and Imaging Institute
Thematic Area: Teaching and Learning
Bei Wang Phillips’s research makes AI safety easier to see, understand, and improve. Instead of relying solely on measures of accuracy or fairness, she uses topology, geometry, and visualization to show where, why, and for whom AI systems may fail: an approach she calls the "shape of AI safety." For example, a hiring system may appear accurate overall while disadvantaging a particular demographic group. A chatbot may learn to give answers that please users or meet a performance target rather than provide truthful information. A medical diagnostic system trained at one hospital may become unreliable when used with images from a different scanner or patient population. By making these weaknesses visible, this work helps students design and evaluate AI systems more carefully. The goal is safer and more socially responsible AI.
Naomi Riches
Research Assistant Professor; Department of Obstetrics and Gynecology, Spencer Fox Eccles School of Medicine
Thematic Areas: Environment; Health Care and Wellness
Naomi Riches examines how environmental exposures such as wildfire smoke, air pollution, radiation from historical nuclear testing, and heat may contribute to reproductive and adverse pregnancy outcomes such as fetal growth restriction and stillbirth. By linking large health and environmental datasets, she aims to identify exposure patterns and mixtures that may increase risk. A central goal of her research is responsible AI: developing transparent, interpretable, and reproducible methods rather than relying on black box algorithms. As a fellow, she will compare established mixture methods with interpretable machine-learning approaches, evaluate whether findings and model performance differ across geographic and socioeconomic groups, and carefully document how data are linked and analyzed. By making these methods more trustworthy and accessible, this work could support evidence-based decisions to improve maternal and infant health.
Guang Tian
Assistant Professor; City & Metropolitan Planning, College of Architecture and Planning; Scientific Computing and Imaging Institute
Thematic Areas: Environment; Teaching and Learning
Utah faces serious environmental challenges. Winter inversions trap pollution along the Wasatch Front, while rapid growth and urban expansion lead to longer drives, heavier traffic, higher emissions, and the loss of natural land. Rising housing costs affect where people live and how far they travel. Guang Tian uses AI to help communities respond. He combines data from multiple sources to examine how transportation and land-use choices affect air quality, access to jobs and services, affordability, and quality of life. He is building urban digital twins (virtual models of cities) that allow planners and residents to test options such as new transit routes or different growth patterns before investing. His responsible AI approach protects privacy, checks for unfair outcomes, explains results clearly, and involves communities in decisions.
Jon Wang
Assistant Professor; School of Biological Sciences, College of Science
Thematic Areas: Environment; Teaching and Learning
Jon Wang leads the Dynamic Carbon and Ecosystems (DYCE) lab, which uses large datasets from drones, aircraft, and satellites to examine the ways climate change is transforming ecosystems and the carbon cycle. For example, his lab recently used high-performance computing to process hundreds of terabytes of airborne laser scanning, satellite imagery, and extensive field data with statistical learning and map 40 years of forest biomass change due to severe wildfire across Canada and Alaska. Current work is leveraging AI approaches to delineate, label, and characterize hundreds of millions of tree crowns across the western U.S. in order to track the forest-killing effects of drought and insects, the extent and rate of woody plant encroachment, and consequences of severe wildfire on semi-arid forests.
Eliane S. Wiese
Assistant Professor; Kahlert School of Computing, John and Marcia Price College of Engineering
Thematic Area: Teaching and Learning
Eliane Wiese studies how to help students build the judgment they will need in an AI-enabled world. An expert in education and human-centered computing, she explores ways to teach computing students to foreground human concerns, from incorporating ethics into AI lessons to examining code readability as students learn programming. As a fellow, she will explore how AI can support higher-level reasoning in early computing classes. Instead of having students use AI to write code, AI can analyze their code to show instructors their approach and can generate different solutions for students to critique and improve. By showing that technical solutions can produce the same output but differ in how well they fit a context, students can learn to consider alternatives, weigh tradeoffs, and use their own judgment.