Technology acceptance model analysis of perceptions and use of AI writing tools among Yemeni EFL students in academic writing

Artificial Intelligence (AI) has emerged as a transformative tool in language learning, creating opportunities to facilitate teaching efficiency and support individualized learning, particularly in the development of writing skills. This study examines Yemeni EFL students’ perceptions of using AI tools in writing, their perceived usefulness, and the challenges associated with integrating these tools. The Technology Acceptance Model (TAM) served as the theoretical framework to understand how Yemeni EFL students accept and adopt AI writing tools. The study employed a quantitative method. An online survey was completed by 51 Yemeni EFL students from three universities (Sana’a, Hajjah, and Saba). In addition, a pretest–posttest component involving 30 of these students using the QuillBot application was conducted to examine changes in academic writing performance. The study identified several perceived benefits of AI tools, including increased writing confidence and the use of AI for proofreading, summarizing, paraphrasing, grammatical correction, translation, and immediate feedback. However, the study also noted challenges such as students’ lack of motivation, over-reliance on AI tools for simple tasks, the possibility that AI may conceal students’ writing deficits (potentially leading to unfair instructor evaluation), and other technological issues. The findings revealed that students held positive perceptions of AI tool integration in academic writing. However, the pretest–posttest component showed that both groups improved over time, and the group × time interaction was not statistically significant, indicating no differential improvement attributable to QuillBot use F (1, 28) = 0.586, p = .450). These findings highlight the importance of responsible and well-planned integration of AI into the teaching of academic writing. Policymakers and teachers can use these preliminary findings to help design programs that combine AI with writing instruction, pending further research with more rigorous designs.