Athoillah, Muhammad. (2026). A Descriptive Study of EFL Students’ Use of AI Tools in Academic Writing on English Department Students Batch 2023. Sarjana’s Thesis. English Language Education Study Program, Faculty of Teacher Training and Education, Universitas Lambung Mangkurat. The first advisor: Dr. Moh. Yamin, S.Pd., M.Pd., the second advisor Elsa Rosalina S.Pd., M.Pd.

Keywords: Academic Writing, AI Tools, EFL Students, Learner Agency, Scaffolding.

The rapid integration of Artificial Intelligence (AI) in higher education has transformed conventional academic writing practices, yet the specific processes and operational behaviors of EFL learners remain under-explored. This study aimed to describe how and for what purposes EFL students integrate AI tools into their academic writing tasks.

Utilizing a descriptive qualitative design, this research investigated the experiences of English Department students from batch 2023 at Universitas Lambung Mangkurat. Data were collected through a screening questionnaire, semi-structured in-depth interviews with selected participants, and qualitative document analysis of the students' writing drafts. The data were analyzed using a hybrid thematic analysis approach.

The findings revealed that students utilize AI tools as a dynamic scaffolding mechanism across specific writing stages: as supportive scaffolding for idea generation during prewriting, minimized intervention during drafting to protect their original text, and as reflective scaffolding for linguistic refinement during revising. Furthermore, the study identified a diverse spectrum of AI integration among learners, categorized into three distinct patterns: Minimalist-Mechanical, Scaffolding-Relational, and Holistic-Cognitive Support.

In conclusion, EFL students do not passively accept AI outputs; instead, they demonstrate active learner agency and self-regulation through continuous critical negotiation with the technology to maintain their unique authorial voice and academic integrity. Based on these findings, it is suggested that academic institutions should shift from prohibitive AI policies toward pedagogical frameworks that foster critical curation and ethical AI companionship in EFL writing instruction.