This study investigated the impact of DeepL AI on students’ English writing performance. It aimed to examine how DeepL is implemented in classroom settings and the extent to which it enhances students’ writing abilities.

The research used quantitative experimental method, with 27 subjects divided into experimental and control group. The study was grounded in the concept of AI-assisted writing, where tools like DeepL utilize neural networks and natural language processing to support students in producing better writing (Kruse, Chitez, & Rapp, 2023). The theoretical framework combined Douglas Brown’s (2018) writing assessment theory and the TPACK (Technological Pedagogical Content Knowledge) framework by Mishra and Koehler (2006) model to support the analysis.

Results showed that the experimental group’s mean score improved by 27.5 points—from 53.75 to 81.25—while the control group only improved by 1.75 points. The t-test value (10.36) exceeded the critical value (2.09), indicating a statistically significant difference. These findings suggest that DeepL AI provides effective real-time feedback, improves grammatical accuracy, and enriches vocabulary use.

Although DeepL AI should not replace traditional methods, it can serve as a powerful complementary tool in the writing classroom. The study recommends integrating AI-based writing tools to enhance learning outcomes without promoting overdependence.