<![CDATA[In higher education, the integration of artificial intelligence (AI) has experienced rapid development. However, empirical evidence regarding the effectiveness of AI in improving students' academic writing skills remains limited, particularly in Indonesia. This quasi-experimental study investigated the effect of AI-assisted writing instruction on the academic writing proficiency of undergraduate English as a Foreign Language (EFL) students at Universitas Muhammadiyah Palopo. This study involved 60 students divided into two groups: an experimental group and a control group. Students' writing was assessed using a validated rubric that focused on four aspects: content and organization, grammar and mechanics, vocabulary and style, and citations and references. The instrument demonstrated strong inter-rater reliability (Cohen's κ = 0.87). The results showed that the experimental group achieved significantly higher scores than the control group. These results were analysed using an analysis of covariance (ANCOVA). This can be seen from the higher scores (Cohen's d = 2.00) and (N-Gain = 0.42) in the experimental group compared to the control group, which achieved (N-Gain = 0.20). Furthermore, data from the perception questionnaire (Cronbach's α = 0.84) indicated that most students responded positively to the use of AI tools. Reported benefits included increased writing efficiency, improved grammatical accuracy, and better organization of ideas. However, some participants also expressed concerns about the risk of over-reliance on the technology. Overall, the findings suggest that AI-assisted writing instruction, when implemented within a structured and targeted pedagogical approach, can significantly improve academic writing skills in EFL higher education contexts. These results offer practical implications for curriculum design and teaching strategies.]]>
Copyrights © 2026
