Moral-Technical Relationship: A Structural Analysis of Trust and Academic Integrity as Predictors of Generative AI Adoption in Higher Education

Abstract

Generative AI has become a powerful tool in higher education, helping students and researchers work more efficiently. However, this rapid adoption raises important questions about how users view AI ethics and trust. This study investigates what actually drives AI adoption among students, researchers, and lecturers in Indonesia, using an extended version of the UTAUT2 model. Based on 250 survey responses analyzed using PLS-SEM, the model successfully explains 27.8% of the variance in users' intention to adopt AI tools. The results show that Performance Expectancy, Effort Expectancy, Social Influence, and Moral Obligation significantly influence AI adoption. In contrast, Trust and Perceived Ethics were found to be non-significant factors. These findings indicate that the academic community in Indonesia currently has a 'utility-first' mindset: users prioritize how useful and easy the AI is for their immediate tasks, rather than focusing on the ethical risks or the trustworthiness of the system. This highlights a critical need for universities to provide stronger ethical training and guidelines as AI usage continues to grow