Answers are grounded in this paper and cite their sources.
This study investigated the potential of AI-assisted teachers using an AI Copilot to improve IELTS writing task 2 instruction in online learning environments. Following a mixed-methods design and using an AI Copilot, the study examined students’ writing performance and the quality of teacher instruction informed by AI and personalized feedback in an IELTS writing course focusing on Task 2. The study quantified score improvements and revealed how students perceived AI-driven insights as crucial in improving their writing through semi-structured interviews. Fourteen IELTS candidates participated in a 24-session online course, with each session lasting 90 min and conducted twice per week over six weeks. They completed pre- and post-tests to assess writing score improvement. Additionally, semi-structured interviews provided qualitative insights. A paired-sample analysis compared the participants' pretest and post-test writing scores. The results indicated that the integration of AI copilots within a teacher-led framework significantly improved EFL learners' writing performance in one-on-one online IELTS writing courses in four components of Task Achievement (TA), Coherence and Cohesion (CC), Lexical Resource (LR), and Grammatical Range and Accuracy (GRA). Qualitative analysis indicated participants found the AI-assisted teacher's instruction, personalized feedback, and sample writings highly valuable for understanding writing structure, organization, and meeting target band scores. However, the limited sample size necessitates careful interpretation of these findings. This study proves that AI-assisted online instruction can enhance the effectiveness of IELTS writing task 2 education by providing consistent, personalized support at scale. These results further emphasize the potential of AI integration into teacher professional development programs, particularly for online and distance learning modalities.
No takes yet. Share an insight, caveat, or question.
Asadi et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: