Explaining Students’ Generative AI Acceptance in Higher Education: An Extended Technology Acceptance Model with Digital Trust, Self-Efficacy, and Perceived Academic Value

Abstract

Background of study: Generative artificial intelligence (GenAI) is becoming a routine academic resource, yet students differ substantially in whether they regard it as trustworthy, manageable, and educationally valuable.  Aims and scope of paper: This study extends the Technology Acceptance Model (TAM) by integrating digital trust, AI self-efficacy, and perceived academic value to explain students’ behavioral intention to accept GenAI in higher education.  Methods: A cross-sectional multi-institutional survey and covariance-based structural equation modeling (CB-SEM) are proposed. For manuscript-development purposes, the reporting format is demonstrated with a clearly identified synthetic dataset (N = 438); these values are illustrative and must be replaced by verified empirical data before submission.  Result: The illustrative model shows acceptable fit (CFI = .957, TLI = .949, RMSEA = .049, SRMR = .043). AI self-efficacy predicts perceived ease of use; digital trust and perceived academic value predict usefulness and attitude; and attitude, usefulness, academic value, and self-efficacy jointly explain 68% of the variance in behavioral intention.  Conclusion: GenAI acceptance is better explained when classical TAM beliefs are complemented by students’ confidence, calibrated trust, and judgments of academic value. Institutions should therefore combine technical access with AI literacy, trust-building governance, and pedagogically meaningful use cases.