GENERATIVE-AI TECHNOLOGIES, WITH CHATGPT AS A PEDAGOGICAL EXEMPLAR, IN HIGHER EDUCATION: A CRITICAL EXAMINATION OF INNOVATION DIFFUSION AND THE MEDIATING ROLE OF STUDENT NORMATIVE-ETHICAL AGENCY

Abstract

AI-driven generative technologies, with ChatGPT as a leading example, are transforming higher education by redefining how knowledge is created, evaluated and applied.  Addressing issues of misuse, ethics, and academic dishonesty is an essential part of developing any technology and making it as useful as possible.  To comprehend the process of diffusion in its entirety, it is necessary to examine student use and sharing of ChatGPT through the lenses of Unified Theory of Acceptance and Use of Technology (UTAUT), Social Cognitive Theory (SCT), and students' individual normative-ethical agency.  Perceptions of academic/perceived anxiety, perceived technological capability, perceived utility, perceived user-friendliness, normative influence, and perceived learning capability are some of the factors that this study aims to examine in order to understand how and why students use ChatGPT. Secondly, to assesses the moderating influence of perceived adherence to normative Academic Ethics in defining these variables relationship.

A cross- sectional quantitative research design was employed, using web based survey distributed to university students across Pakistan. This generated an overall 1,200 valid responses, with a 70.4% response rate. The 41 item survey, adapted from a few validated scales, was piloted to achieve clarity after which it was critiqued by experts. The survey was then tested again for clarity and to obtain reliability. The dataset was put through an extensive configuration which included descriptive statistics, common method bias, and an assessment of robustness concerning outlier treatment and tests for multicollinearity which gave assurance. The Structural equation modeling was applied to test the study hypothesis relationship. The result suggests the need to nurture responsibility along with ethical boundaries of policy, training, and digital awareness. The results also indicate the factors which include Perceived Usefulness, Perceived User Friendliness, Normative Influence, Perceived Learning Capability, and Perceived Technological Capability regarding a positive assessment of ChatGPT use are greatly diminished by Perceived Academic Anxiety.

The findings show that Although Reputational Normative Academic Ethics is the only one that has constraining direct influence and still enables the reduction of the other negative influence. All students, to varying degrees, acknowledge the usefulness and social acceptability of ChatGPT while ethically engaged students employ its functionality more restrictively. This suggests that educational institutions ought to formulate policy, training, and outreach strategies that achieve integration of digital literacy and its ethical application. Educational institutions, in collaboration with developers and policymakers, have the potential to encourage the responsible, critical, and sustainable integration of technologies such as ChatGPT into academic practices. This centers on the primary integration of promoting responsible usage and critical engagement.