Epistemic Parochialism: Single Institution Studies in The Age of Artificial Intelligence Large Language Models

Abstract

Learning and teaching scholarship is experiencing a growing tension between increasing global interconnectedness and the persistent production of highly decontextualised local studies. This Editorial argues that such epistemic parochialism limits our ability to build cumulative knowledge and risks reinforcing narrow solutions to shared problems. Drawing on contemporary pressures including internationalisation, epistemic monocultures, system-level complexity, and the rapid integration of artificial intelligence large language models, the Editorial outlines why research that treats the institution as the horizon of inquiry is no longer adequate. In an era where large language models learn from, reproduce, and amplify scholarship, narrowly framed or duplicative studies have implications not only for academic discourse but also for the quality of language models. To address this challenge, the Editorial presents five principles to guide authors in producing better situated research: reframing the problem beyond the institution; developing a credible inquiry into what is known; justifying methodological positioning; situating institutional cases within complex systems; and synthesising localised insights to advance global knowledge. By adopting these principles, authors strengthen the relevance, clarity and impact of their work, enhance their prospects for publishing success, and help safeguard the coherence and diversity of the knowledge on which both human scholars and artificial intelligence systems increasingly depend. The Editorial concludes with a call for the sector to recognise epistemic parochialism as a significant contemporary peril and to collectively pursue scholarship that is reflective, situated and globally engaged.