Behavioral Determinants of AI-Driven Arabic Language Learning: Insights from the Extended UTAUT2 Model

Background/purpose. Artificial Intelligence (AI) is transforming education through adaptive learning systems, intelligent tutoring, and personalized instruction. Although AI has become increasingly prominent in language education, its specific impact on enhancing Arabic proficiency has not been adequately explored, especially in terms of student engagement, linguistic development, and sustained adoption. This study examines the behavioral factors influencing the use of Generative AI (GenAI) in Arabic language learning by applying the Extended Unified Theory of Acceptance and Use of Technology (UTAUT2) framework. Materials/methods. To achieve this, quantitative data were collected from students at Sultan Qaboos University in Oman, including both Arabic majors and students from other disciplines. This study utilizes ordinal logistic regression to examine the key factors influencing students' adoption of AI in Arabic language learning. It explores how performance expectancy, effort expectancy, facilitating conditions, social influence, hedonic motivation, and habit shape students’ engagement with AI-driven tools. Results. Findings indicate that performance expectancy and social influence are the strongest drivers of adoption, while effort expectancy and facilitating conditions play a key role in sustained engagement. Despite AI’s potential to enhance pronunciation, grammar, and vocabulary acquisition, adoption remains inconsistent due to usability issues, content accuracy concerns, and varying levels of AI literacy. Conclusion. The study highlights the need for structured AI integration, improved digital literacy, and interactive AI-driven learning environments to enhance Arabic language acquisition in higher education.