AI conversation at UND spans across campus

GRAND FORKS — Higher education is a central point in the conversation around artificial intelligence, according to Anna Kinney. The University of North Dakota is thinking about AI in terms of workforce, mental health, community partnerships and concerns, education and opportunities. Kinney, the AI instructional manager for the Teaching Transformation and Development Academy, said there is also an untenable pace of progression for the technology that needs to be kept up with. ADVERTISEMENT “There is a ton going on,” she said. “It’s really huge.” At the State of the University address last year, President Andrew Armacost said a goal — or “moonshot” — for UND was to be the AI university for North Dakota, being a national example of how AI can be adopted. However, the university had been looking at AI for a few years before that. There is change management involved as AI continues to change over time, which, in Kinney’s perspective, is about taking care of people. Kinney’s position as AI instructional manager includes building an “AI across the curriculum program.” She focuses on helping faculty think about their pedagogy and how to think about AI in disciplinary curriculum and what goals should there be for students as they navigate their degrees to ensure they leave with AI literacy. Though there is excitement from some about the opportunities of AI, there are also those who are skeptical about it. There are concerns about AI’s impact on critical thinking, mental health and how to use AI ethically, humanely and appropriately. UND’s AI and Human Innovation Initiative, which looks at the pros and cons of AI from an arts and humanities perspective, recently held its second showcase, in which colleagues discussed how to continue nurturing writing skills in students. “It is essential that we preserve those literacies, and we don’t throw them away or set them aside in service of teaching directly to a technology,” said Kinney, who used to direct the UND Writing Center. “This expertise is the literacy that students need to be successful, no matter how that technology changes.” Writing programs, writing centers and writing scholars have previously discussed the use of other technologies in the writing process, such as auto-fill and programs used for paraphrasing. She recalled that when ChatGPT emerged, an article was published in The Atlantic called “The College Essay Is Dead” by Stephen Marche. “I think that was a real rallying moment, a panic moment,” she said. “‘What happens from this?’” ADVERTISEMENT There is no part of UND that isn’t touched by writing, Kinney said. It is an assessment product to look at student learning, it is used to access funding through grants, it’s how research findings are communicated, it’s how UND recruits and advertises to students — there is “no part of our institutions that isn’t touched by this idea of an essay,” she said. UND went into motion. The TTaDA set up educational opportunities to the general campus community, which then extended out to the North Dakota University System, offering panels with experts about AI as a research area and as a tool being used in different ways. People from different areas of the university were pulled together to start conversations about how to dig into AI, followed by an intensive faculty development workshop in May 2023. The university started talking about how to tackle faculty development around AI and ensure the learning outcomes of graduates remain central to what faculty is doing, mainly critical thinking and information literacy. “We started with the skills, and then we layered AI into that as part of the context that students navigate those skills in,” Kinney said. “And it was really great to hear faculty kind of think about that from their own discipline.” Faculty in the workshops were asked to write a new assignment or revise an assignment to address AI as a complicating context. After reviewing the assignments, UND created an open educational resource to house them so colleagues locally, across the state and globally, could have access to them to adopt or adapt, as well as capture how people were thinking at this early moment of AI, Kinney said. The practice was repeated, this time requiring the integration of AI to start experimenting with the technology, which was also added to the open resource. The resource has 73 assignments with more than 15,000 downloads globally. UND has also created Co.AI — the AI Collective — which is a group of people representing diverse positions on campus. There is representation from all the colleges, from staff, faculty and students to get an inventory of AI work on the campus, including projects, ideas and use cases. Kinney, one of the collective's co-chairs, said some members are thinking about AI as an operational consideration, a teaching and learning consideration, a tool for research and as an area of research. There are also AI skeptics in the group. The Co.AI group has some work to do over the summer as it gains momentum so that, in the fall, there will be a renewed energy to pull together connections across campus and coordinate, Kinney said.