GenAI tutors in higher education: Testing support for student learning

Artificial intelligence, specifically generative AI, has developed substantially in just a few years with tools like ChatGPT capturing imaginations while raising concerns. Capitalizing on this growth, “AI” has become one of the latest features required for advanced products across a broad range of fields. Higher education is no exception, and companies are working to integrate GenAI (Generative Artificial Intelligence) into their offerings. Universities have strong incentives to engage students in their studies and graduate on time. While GenAI tools appear to be a powerful educational ally, potentially addressing both these broad aims, evidence is necessary to justify the investment – time and money – needed to make this partnership a reality. As part of a series of studies, the present research was to experimentally test the benefits of GenAI for student learning and its essential corollaries. In this experimental study a customised AI Tutor provided students with interactive tutoring in a principles of marketing course. The relationship between students’ use of a generative AI tutor and their interest, self-efficacy, and achievement (each for the course of study) across pre-and-post treatment were tested. Results confirm the challenges facing any new technology which seeks to enhance human behaviour and outcomes in educational settings. Experimental findings are reviewed in detail and directions for future efforts in this area are discussed.