“We’re in this together”: responding to student concerns about large language models in higher education

Abstract

We, the self-proclaimed “LLMs in Education” group, comprising university educators (and one partner researcher) from the University of Bergen, Norway, conducted two workshops in August and December 2023 to explore the use of tools powered by large language models (LLMs) in education. We analyzed student responses from two surveys, one national (n=660) and one local (n=26), to understand their perceptions and concerns related to tools such as ChatGPT. Specifically, we focused on responses to the question “How do you think these tools should be used in education, in a way that is fair and that supports your learning?”

We categorized the concerns that emerged in our reading of this student feedback, and we were able to group these categories into two broad themes: questions and concerns about the tools themselves, and questions and concerns related to acceptable use, fairness, and ethics. We provide our collaborative responses to the identified questions and concerns in both categories.

In sum, we advocate for honesty, transparency, and working with students as we navigate these new tools in teaching and learning. Students are definitely looking to their instructors to identify boundaries and best practices for LLMs in our courses, but as with many new technological advances, we may assume that our students are far beyond us in terms of savvy use of these tools. This is where a continued dialogue will be helpful in the months ahead. We will continue to ask our students about their use and perceptions of LLMs, etc., and to be fair, we should be honest with them about our own practices, confusion, and concerns.