Year: 2026
Type: article
Abstract: Recent technological advancements in artificial intelligence (AI) and machine translation (MT) have fundamentally transformed EFL writing instruction and assessment.This study investigated the effectiveness of traditional non-AI methods, AI-assisted learning, and hybrid approaches in undergraduate EFL writing courses at a private Japanese university.Over one academic semester, 38 students in intensive language courses participated in this mixed-methods research, which included surveys and writing samples.Results revealed that despite institutional policies and instructor guidance, students consistently used AI and MT tools for written assignments whenever digital access was available.Only in-class, paper-based, monitored writing produced nearly 100% authentic student work.While hybrid methods reduced AI and MT use, they did not eliminate it.Notably, neither instructor intuition nor AI detection tools proved reliable enough for making academic decisions, as they produced too many false positives.Based on these findings, a balanced approach is recommended: incorporating AI as a learning tool while maintaining traditional analog writing instruction to develop foundational skills.This includes reintroducing handwritten test notebooks for in-class writing evaluation.The study raises important questions about the future of EFL writing instruction, including its continued relevance, the potential of purpose-built digital writing instruction and evaluation tools, and the evolution of student-teacher collaboration in writing instruction.
Authors: Wilkins, Michael, Peet, Thomas
Language: English
Subfield: Language and Linguistics
Field: Arts and Humanities
Domain: Social Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W7154869896