Nurfitrah
Universitas Negeri Makassar

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

Benefits and Limitations of ChatGPT in the EFL Classroom: Students' Perspectives Irmayanti; Dita Musfira Tihurua; Nurfitrah
IDEAS: Journal on English Language Teaching and Learning, Linguistics and Literature Vol. 14 No. 1 (2026): IDEAS: Journal on English Language Teaching and Learning, Linguistics and Lite
Publisher : Universitas Islam Negeri Palopo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24256/ideas.v14i1.10122

Abstract

This qualitative descriptive study examined how English as a Foreign Language (EFL) students in higher education used ChatGPT and how they perceived its benefits and limitations in academic learning. Twelve students with prior experience using ChatGPT for academic purposes were selected purposively. Data were collected through semi-structured interviews and analyzed using Braun and Clarke's thematic analysis. Three themes were identified. First, participants used ChatGPT as an academic assistant for writing, comprehension, summarization, paraphrasing, and idea generation. Second, they perceived support for grammar, vocabulary, sentence organization, confidence, and motivation. These accounts represent self-reported experiences rather than objectively measured improvement in language performance. Third, participants raised concerns about the credibility and specificity of AI-generated information, the absence of reliable academic references, possible learner dependency, plagiarism, and access constraints. The findings indicate that the educational value of ChatGPT depends on students' capacity to evaluate its output, consult credible sources, and retain responsibility for academic decisions. In EFL higher education, ChatGPT should therefore be treated as a supplementary and teacher-guided resource rather than a substitute for independent reading, writing, and critical evaluation. The study contributes a process-oriented account of learner-AI interaction while acknowledging the limitations of a small, context-specific sample and self-reported data.