Abstract

Artificial intelligence (AI) is transforming higher education by enabling personalised learning, automated assessment, and intelligent instructional support. This study aims to (1) investigate the factors influencing the success of artificial intelligence in education (AIEd) and (2) develop a predictive equation for AIEd success. This quantitative study employed covariance-based structural equation modelling (CB-SEM) combined with the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), extended by incorporating Trust (TR) and Self-Efficacy (SE) as additional constructs. Data were collected from 381 university lecturers in Thailand who had prior experience with AIEd via an online questionnaire. Confirmatory factor analysis (CFA) was used to assess measurement model validity, while CB-SEM was employed for hypothesis testing. The model demonstrated good fit (CMIN/df = 1.95; GFI = 0.985; CFI = 0.918; RMSEA = 0.050; SRMR = 0.052). Results indicated that Intent to Adopt AI for Teaching (IA) was the strongest predictor of AIEd success (SAU), with a standardised path coefficient of β = 0.423 (p < 0.001), explaining 60.84% of the variance in SAU. Facilitating Conditions (FC), Trust (TR), and Price Value (PV) also demonstrated significant positive effects. Notably, Performance Expectancy (PE) and Effort Expectancy (EE) influenced SAU indirectly through Trust rather than directly through IA, suggesting a trust-mediated pathway to AIEd adoption. Limitations regarding the convergent validity of the IA construct are acknowledged, and implications for educational technology policy and practice are discussed.

This work is licensed under a Creative Commons Attribution 4.0 License.