Dissonance in generative AI use among student writers: How should curriculum managers respond?

Abstract

This qualitative study investigated the case of ChatGPT use by five students in an English for Academic Purposes (EAP) class comprised of 28 students in total. The results of the analysis revealed dissonance in the way the subjects of the study had intended to use ChatGPT in their written assessments and how they used it, six weeks into the writing program. The results of the study revealed that their intention to use ChatGPT may have been within the remit of fair use, but their written output as evaluated by an AI detection tool revealed that the students might have violated academic integrity. This study has clear implications for the role of educators to guide students on how to use AI in their written assessments in such a way that they uphold academic integrity. Educational institutions, therefore, should take a pedagogical stance on AI text generators, which means that proactive measures should be taken to embrace AI as an assistive tool for writing, but fair and guided use of the technology must be in place.