Abstract
The spring semester of 2023 in higher education was impacted by the release of Large Language Models (LLMs) to the general public. OpenAI’s ChatGPT was at the forefront, with Google's Bard, Microsoft’s Bing, and others following suit. In response to the ongoing discussions in higher education, we conducted a survey to examine the use and understanding of LLMs among students and instructors at the Faculty of Mathematics and Natural Sciences at the University of Bergen, Norway. Given that the survey was distributed in February and March 2023 - shortly after ChatGPT was launched and before similar tools from other providers were in wide use - we focused primarily on ChatGPT. Our aim was to gain a deeper understanding of existing perceptions and misconceptions about ChatGPT, enabling us to design more informed implementation strategies for its integration into our courses. Further, by considering usage patterns and perceptions shortly after the launch of ChatGPT, we established a baseline for later comparisons as these tools become increasingly embedded in our routines.