Navigating the gAI Landscape: Insights from a Physics Education Survey

Abstract

A survey of physics students at Durham University sheds some light on the students' current use and understanding of gAI and their concerns for the future and the education sector. Physics students highlight that the main use of gAI is for computational assistance, followed by its use as a valuable tool for simplifying complex physics concepts and providing alternative explanations. Overall, students do not trust gAI and cross-check its output. They agree with the department’s policy concerning gAI and express mixed views regarding its use for providing feedback and assessing students’ work. Differences across levels have been found, with the students’ confidence in effectively using gAI increasing with level, together with resistance to the possibility of using gAI to assess their work. Though online assessments remain popular, their popularity decreases with the level.