Modeling the Impact of AI Features on Writing Performance: A TAM–SRL Integrated Framework

AI writing tools are becoming more crucial in helping university students with their English writing as generative AI in education continues to grow. This study explores how essential system features affect students’ writing performance through technology acceptance routes. And it also researches the moderating function of self-regulated learning. This study used a questionnaire survey to gather data from 531 English major students using a quantitative design. We used partial least squares structural equation modeling to examine the links between system features, perceived usefulness, perceived ease of use, attitudes, behavioral intentions, and writing performance. The findings demonstrate that perceived utility and perceived ease of use, which in turn affect attitudes and behavioral intentions, are highly predicted by AI system attributes. Behavioral intention positively affects writing performance, and self-regulated learning strengthens this relationship. This study contributes to understanding the mechanisms through which AI tools enhance writing development and provide practical implications for improving AI tool design and integrating them into writing instruction. Future research should further examine long-term effects using longitudinal or process-based data.