Exploring Factors Influencing Continuance Intention of Pre-Service Teachers in Using Generative Artificial Intelligence

Abstract

Generative Artificial Intelligence (GAI) holds significant potential to enhance pre-service teacher professional development. However, research has primarily focused on initial acceptance, neglecting post-acceptance behaviours, particularly the factors influencing continued GAI use among pre-service teachers. To address this research gap, this study extends an Expectation-Confirmation Model (ECM) to include information quality and AI self-efficacy as additional determinants. Using the partial least squares structural equation modelling (PLS-SEM) approach, we analysed data from 506 Chinese pre-service teachers. Findings reveal that information quality positively impacts perceived usefulness and expectation confirmation, both of which enhance use satisfaction. Together with AI self-efficacy, these elements emerged as key predictors of intention to continue using GAI, with perceived usefulness as the most direct factor. Contrary to the hypothesis, personal major did not moderate these relationships. This study contributes to a deeper understanding of the behaviours and motivations of pre-service teachers post-GAI adoption, offering new insights into the sustained development and integration of GAI in teacher education.