The Impact of Generative AI (e.g., ChatGPT) on EFL Learners' L۲ Writing Fluency and Academic Integrit

The present study was conducted to rigorously and comprehensively investigate the impact of generative artificial intelligence, particularly large language models like ChatGPT, on two fundamental components of language education: L۲ writing fluency and the academic integrity of English as a Foreign Language (EFL) learners. In recent years, the emergence of novel natural language processing technologies has caused profound transformations in learning paradigms and text generation. This research employs a descriptive-analytical approach based on a mixed-methods (quantitative and qualitative) design. The statistical population comprised ۴۵۰ EFL learners at the postgraduate level, selected through stratified random sampling. Data collection instruments included pre- and post-intervention writing tasks, attitude questionnaires, plagiarism detection software, and semi-structured interviews. Quantitative findings demonstrate that the utilization of generative AI significantly increases writing speed, lexical diversity, and syntactic complexity in the learners' final written products. However, qualitative analyses indicate that this apparent enhancement in quality does not necessarily equate to the improvement of individual cognitive skills and internal writing fluency; a severe decline in performance is observed when access to the tool is restricted. Furthermore, results concerning academic integrity reveal deep ethical challenges; the boundaries between educational scaffolding and academic dishonesty have become severely blurred, and traditional plagiarism detection systems prove ineffective in identifying machine-generated texts. Ultimately, this research proposes macro-level strategies for revising academic assessment methods and developing modern ethical codes of conduct to preserve the integrity and credibility of the educational system while harnessing technological capabilities.

کلیدواژه ها:

Generative AI ، Writing Fluency ، Second Language ، Academic Integrity ، EFL Learners ، Natural Language Processing ، Academic Assessment