Exploring the impact of generative AI-facilitated personalized learning on EFL students’ speaking and writing skills

The use of Generative Artificial Intelligence (GenAI) in education has created opportunities for individualized learning. However, its effects on the speaking and writing skills of EFL students are still insufficiently examined. Current research predominantly emphasizes AI-assisted writing support, with limited empirical data on its efficacy in enhancing speaking proficiency. Seventy-eight EFL undergraduate students from a private university in North Cyprus were recruited for this study. The participants included 53 males and 25 females ( N = 78). The participants were between 18 and 25 years old. They were assigned to control and experimental groups. Pretests and posttests were used to assess fluency, accuracy, coherence, and lexical resources in both speaking and writing. The experimental group completed the same sequence of GenAI-supported tasks using standardized prompts, while the control group completed equivalent tasks using traditional instructional methods. This strategy facilitates the evaluation of the intervention’s efficacy by comparing performance variations between the two groups. Results from paired t-tests, independent t-tests, and ANCOVA showed that the experimental group significantly outperformed the control group, with large effect sizes, in both skills. Within the scope of this study, GenAI-facilitated personalized learning was associated with measurable gains in speaking and writing proficiency when implemented as a structured pedagogical co-assistant rather than a substitute for instruction; broader generalization beyond this sample and institutional context awaits further research. The study discusses implications for classroom practice, educational policy, and responsible GenAI integration in English language teaching.