Search for a command to run...
Abstract The use of Artificial Intelligence (AI) at higher education institutions (HEI) in South Africa has emerged as a transformative academic tool that is aimed at advancing teaching, learning and research. This research study was aimed at investigating the use of AI in academic writing for master’s and doctoral students at a University of Technology (UoT) in Southern Gauteng South Africa, with specific focus on ethical application and AI impact on postgraduate studies. This qualitative study explored how postgraduate students at a South African University of Technology use generative AI particularly ChatGPT tool for academic writing and research projects. The researcher analysed open ended questionnaires and interviews with 59 students across two faculties using thematic analysis. Many students reported efficiency gains in idea generation, outlining, and language editing, including concerns about information accuracy, authorship, and dependence on the tool. The study found that unregulated application of AI may impact negatively on postgraduate student’s academic writing skill, critical thinking and increases integrity risks where assessments rely on the use of AI tools alone. The researcher proposes a balanced approach combining clear AI use policies, early ethics education, and assessment redesign emphasizing process evidence (e.g., drafts, protocols, and oral defences). The findings inform practical governance for the institution seeking to harness AI while safeguarding quality and integrity in postgraduate studies. The study also acknowledges that, although AI tools facilitate idea generation and technical tasks, it may however undermine the value of academic writing, critical thinking, overreliance and quality on research outputs. To uphold ethical standards and preserve intellectual rigour, the study advocates for a more balanced application in the use of AI for master’s and doctoral studies for the development of critical scholarly advancement.