Analysis of Generative AI Acceptance Using the Technology Acceptance Model (TAM) at Universitas Merdeka Madiun
TL;DR
The findings indicate that Perceived Usefulness is reflected in enhanced efficiency for reference searching and academic task completion, and Perceived Ease of Use is manifested through intuitive user interfaces, although effective prompt engineering skills remain unevenly distributed.
Abstract
Generative artificial intelligence has become one of the most influential technologies in higher education, with various platforms actively utilized by the academic community to support diverse scholarly activities. Despite the rapid surge in adoption within this sector, empirical evidence regarding how academic communities accept this technology within higher education institutions remains limited, leaving institutional acceptance patterns poorly understood. This study aims to analyze the acceptance of generative artificial intelligence by the academic community at Universitas Merdeka Madiun using the Technology Acceptance Model (TAM) proposed by Davis (1989) as an analytical lens, focusing on four dimensions: Perceived Usefulness, Perceived Ease of Use, Attitude Toward Using, and Behavioral Intention to Use. Employing a descriptive qualitative approach with a case study design, data were collected through direct observation, in-depth interviews, and documentation involving seven informants selected via purposive sampling. The informants consisted of one Rector, one Head of the Center for Data and Information (Pusdatin), two lecturers, and three students, all of whom possess direct experience utilizing generative AI in academic settings. The findings indicate that Perceived Usefulness is reflected in enhanced efficiency for reference searching and academic task completion. Perceived Ease of Use is manifested through intuitive user interfaces, although effective prompt engineering skills remain unevenly distributed. Furthermore, Attitude Toward Using is generally positive but accompanied by critical concerns regarding technology dependency and academic integrity. Lastly, Behavioral Intention to Use is robust yet highly dependent on further institutional support in the form of standardized guidelines and digital literacy training.
