As researchers across the University of Utah build, study and use generative artificial intelligence (AI), they’re uncovering high-stakes ethical questions that can’t easily be solved by technologists or humanists alone.

Physician Ryan A. Metcalf is exploring how AI might help doctors decide who truly needs a blood transfusion—a common, lifesaving treatment that is also costly and often overused—without sidelining clinical judgment at the bedside.

Economist Ellis Scharfenaker is asking who will control AI’s growing economic power as it reshapes work, with the potential to reduce drudgery and improve safety but also to intensify surveillance, deskilling and inequality.

Political scientist Yuree Noh is using AI to analyze a massive global dataset on censorship and surveillance and wonders how to ensure a large language model’s judgments hold up across countries—including authoritarian ones—without reinforcing biases that could shape policy. “I’m thinking about aid allocation, for example,” Noh said. “What if these systematic biases are affecting those who have the least power to push back?”

A man lectures in front of a screen that says "Can you teach AI the Ethics of AI?"
Event co-lead Jeff Phillips asked what it means to teach a chatbot not just ethical theories and case studies, but the harder human work of making sound moral judgements. Credit: Kelly Hermans

Researchers grappled with these questions and others at a first-of-its-kind AI and Ethics Workshop held Friday, April 3, at the University Guest House. About 75 people attended the daylong interdisciplinary event, led by One-U Responsible AI Initiative faculty fellow and philosophy professor C. Thi Nguyen and his collaborator Jeff Phillips, a computer science professor and member of the initiative’s Faculty Engagement Committee.

“AI is invading or driving everything, depending on your perspective,” Phillips said. “We need to pause and think, ‘Is it ok to do it that way?’”

As part of Nguyen’s fellowship, he and Phillips are building the U’s first AI and ethics course cross-listed in philosophy and computing. They used the event, in part, to start to build an interdisciplinary cohort around the subject.

Researchers across the U are examining problems at the intersection of AI and ethics, but many remain siloed in their own departments. “The workshop was centered around facilitating these conversations with people who would normally not get a chance to talk,” Phillips said.

Continue Reading Kelly Hermans’ “U researchers confront urgent AI ethics questions” on @theU.