Our latest paper is a contribution by Dr Shailen Popat on the impact of Generative AI on postgraduate dissertation preparation, cognition, as well as confidence in undertaking primary research. This paper raises a number of important issues with regards to generative AI and HE practice that would be relevant to practitioners in many disciplines.
Abstract
When generative artificial intelligence (GAI) can produce or substantially reshape dissertation writing, a polished final text may offer limited evidence of the student’s independent research judgement. This reflective article asks whether primary data collection, checked at successive stages by a supervisor, could provide additional evidence of dissertation integrity. It draws on my supervision of twenty MSc Public Management dissertations in summer 2026, considered alongside approximately fifty dissertations supervised over the preceding five years. In earlier cohorts, we often encouraged students who use English as an additional language (EAL) to consider literature-based projects because combining fieldwork with dissertation writing in an additional language could be difficult within the available time. In the more recent cohort, some EAL students appeared more willing to conduct interviews or surveys. GAI-supported preparation may have contributed to that confidence, but this reflection cannot establish a causal link. I argue that recruitment decisions, research instruments, pilot interviews and responses to participants examined throughout supervision may make students’ developing understanding more visible than the final text alone. Primary research does not guarantee integrity and should not be required of every dissertation; supervisory checking must illuminate, rather than replace, student judgement.
Download the paper
You can see more details and download the paper here.
Leave a Reply