High school students exploring machine learning and its societal implications: Opportunities and challenges
Magnus Høholt Kaspersen, Karl-Emil Kjær Bilstrup, Maarten Van Mechelen, Arthur Hjort, Niels Olof Bouvin, Marianne Graves Petersen
Abstract
The increased use of AI and machine learning (ML) calls for a general AI literacy, in particular regarding understanding how ML works, the process behind creating ML models, and reflecting on its implications. Where existing learning tools focus on the first two, we explore opportunities and challenges for meaningfully engaging students in understanding and reflecting on ML in their everyday life. We designed VotestratesML, following a Constructive Design Research approach, as an ethics-first learning tool that allow students to explore implications of ML for democratic elections. Based on deployments of VotestratesML in two high school social studies classrooms, we found that safely exploring ML from a concrete starting point helped students reflect and form opinions about its use, that promoting iterative exploration through collaboration and competition motivated them to explore, and that foregrounding ethics in the design and grounding ML in a well-known subject area allowed them to engage with ML on a personal level.