ChatGPT Use and Academic Integrity in EFL Academic Writing: Exploring the Association Among Master’s Students

Abstract

The rapid integration of generative AI in higher education has sparked intense debate regarding its potential to undermine student ethics. While concerns about academic dishonesty abound, empirical evidence linking AI dependency to actual integrity levels remains scarce, particularly in EFL contexts. Grounded in Academic Integrity Theory, this study investigated the correlational relationship between ChatGPT use and academic integrity among first-year EFL Master’s students. A quantitative design was employed using validated questionnaires to measure usage patterns and ethical conduct. Data from 25 participants were analyzed through normality testing, descriptive statistics, and Pearson product-moment correlation. Results indicated that students reported high levels of both ChatGPT use (M = 79.00) and academic integrity (M = 81.97). Crucially, the correlation between the two variables was weak and statistically non-significant (r = .326, p = .112), indicating that no statistically significant association was detected between ChatGPT use and academic integrity in this sample. These findings suggest that frequent AI engagement does not necessarily correlate with lower ethical standards among advanced EFL learners. The study highlights the potential for students to integrate AI responsibly and underscores the importance of institutions focusing on AI-literacy training.