When GenAI becomes the default: a grounded theory study of GenAI-first problem solving among master’s nursing students

This is the authors' abstract. We don't add key points for this paper.

Generative artificial intelligence is increasingly used in health professions education, but less is known about how repeated use becomes integrated into students’ learning and research routines and begins to shape problem-solving behaviour. Guided by the grounded theory approach of Strauss and Corbin, we used purposive and theoretical sampling to conduct semi-structured, in-depth interviews with 23 master’s nursing students from multiple provinces in China between August and October 2025. Data were analysed using constant comparison with sequential open, axial, and selective coding. The first 21 interviews contributed to theory development; two additional interviews were used to assess theoretical saturation. No new categories or changes in category relationships were identified, supporting saturation. We developed a process model describing how generative artificial intelligence becomes embedded in postgraduate learning and research, characterised by a forward shift in the problem-solving pathway in which generative artificial intelligence increasingly functions as the initial point of cognitive engagement. This shift is driven by contextual conditions (impeded task progress, time pressure, insufficient support, and ambiguous norms), consolidated through pathway reconstruction (artificial intelligence-first use and cognitive task outsourcing), and stabilised via dual reinforcement mechanisms (Functional Benefits and Emotional Benefits). Although students develop awareness of potential risks, this pattern is typically maintained through delayed regulation and stage-based rationalisation, resulting in a form of negotiated coexistence rather than immediate discontinuation. Generative artificial intelligence use among master’s nursing students reflects a dynamic shift in learning processes, shaped by both structural and individual factors, in which generative artificial intelligence becomes routinely positioned at the start of problem solving. Educational responses should move upstream to shape when and how generative artificial intelligence is used, strengthen research competence and human support systems, and integrate generative artificial intelligence literacy with critical verification training to promote reflective and bounded use.