Sustainable Education: STEM Students’ Use of Generative AI in Education: Perceived Usefulness and Intentions to Persist to Degree Completion

This article highlights the sustainability of STEM education and the role of emerging technology, specifically Generative Artificial Intelligence (GenAI), in supporting students’ continued educational progress. GenAI is rapidly becoming part of undergraduate education, yet little is known about how students’ use of GenAI functions within the social-cognitive and contextual processes associated with academic persistence. This study examined patterns of GenAI use and tested GenAI use as a proximal affordance within the Social Cognitive Career Theory Performance/Persistence Model among a national sample of 367 undergraduate STEM students across multiple STEM disciplines. GenAI use was widespread: 83% of students reported using GenAI consistently or at least sometimes in their academic work, and most used more than one GenAI product. Students reported using GenAI across roles ranging from assistance with routine academic tasks and communications to critical thinking and decision-making. The derived SCCT Performance/Persistence Model provided an adequate fit to the data. Emotional and instrumental supports contributed to students’ social-cognitive processes, while academic barriers negatively predicted both academic self-efficacy and intentions to persist. Importantly, GenAI use positively predicted intentions to persist to graduation when examined alongside these traditional contextual and social-cognitive variables. These findings position the use of GenAI as an emerging proximal affordance within STEM students’ academic environments and suggest that its self-directed use may represent a developing form of student agency in service of educational progress and persistence.