Dimensions of artificial intelligence acceptance among pre-service EFL teachers: exploring usability, usefulness, social norms, ethics and intentions using ChatGPT

Purpose This study aims to examine the factors influencing acceptance and intention to use ChatGPT among Yemeni pre-service English as a Foreign Language (EFL) teachers. Using a quantitative cross-sectional research design, 597 participants were selected through stratified random sampling from Yemeni universities. Design/methodology/approach A questionnaire was used to collect data, divided into two parts comprising 31 items. Data were analysed using three statistical techniques. Exploratory factor analysis was initially used to uncover the essential structures within the questionnaire data, identifying emergent patterns and confirming scale constructs from multiple perspectives. Confirmatory factor analysis was subsequently conducted to validate these structures, assessing the reliability and validity of the constructs by evaluating the fit of the hypothesized models. Finally, structural equation modelling was applied to further investigate the relationships between the constructs. Findings The results revealed that perceived ease of use, usefulness and social norms had significant positive effects on the intention to use ChatGPT, while perceived ethics had a significant negative influence. Perceived ease of use and perceived usefulness also positively influenced social norms and perceived ethics. Mediation analysis showed that social norms positively mediated the relationships between perceived ease of use and intention to use, as well as perceived usefulness and intention to use, while perceived ethics negatively mediated these relationships. The findings affirm the importance of user-friendly and beneficial artificial intelligence (AI) tools and the need to address ethical concerns to promote acceptance. Originality/value This study offers timely insights into how Yemeni pre-service EFL teachers perceive and accept ChatGPT in their teaching practice. It goes beyond common adoption models by exploring not only usability and usefulness but also ethical concerns and social influences, which are often overlooked in similar research. By focusing on a resource-limited educational context, the study brings a fresh perspective to AI integration in language education and provides practical guidance for teacher preparation programs aiming to introduce emerging technologies responsibly.