Student perceptions of ChatGPT and artificial intelligence tools in higher education: Evidence from early experiences

Due to the growth of generative artificial intelligence and the advancement of ChatGPT in particular, there are unprecedented debates about its use and use in the landscape of higher education. The extensive use of these technologies has not been accompanied with empirical research studies on student perceptions, attitudes, and instructional design significance, which forms a significant knowledge gap that hampers cognizant policy formulation and designing of instruction. This research holds its place as it deals with this gap. In this study, undergraduate and postgraduate students from various institutions in four continents were involved in the study. The outcomes showed that positive attitudes to ChatGPT adoption were significantly predicted by the perceived usefulness (β = 0.436, p < 0.001) and perceived ease of use (β = 0.328, p < 0.001), and the influence of ethical concerns on this relationship was negative (β = -0.187, p < 0.01). Surprisingly, students also showed advanced knowledge regarding the correct application of AI when they were asked about the legitimate application of AI in brainstorming and organizing research with 73.2% of the students acknowledging the valid application in these fields but in a context of assessment, they showed concerns about wanting AI to directly write the answers. The results add to the technology acceptance theory by generalizing Technology Acceptance Model (TAM) to the generative AI setting, as well as, offer practical implications to educators and policy makers engaged in the process of implementing the use of artificial intelligence in educational institutions.