Using AI in the Bilingual Classroom to Enhance Writing Skills in Second Language Learning: A Case Study

Abstract

This is a study of how AI was used in a bilingual classroom for second language learners in English to help master the skills needed at their grade level for written expression skills during the 2024-2025 school year. The study's objective is to see if AI is useful, how it is useful, and how it could be improved. We will focus our paper on the results from one grade level as a case study of 16 to 17-year-old language learners in Paris, France, in which an online AI teaching tool was used for several months by two teachers to track students’ progress. The AI program was used to help teachers with time management in a two-year program where class time had been reduced, and students showed a need for the reinforcement of basic grammar and vocabulary skills to succeed in the bilingual program. It was noted that students did improve overall on the accuracy of their basic grammar level in English, and on the precision of their analytical skills. The study found that the feedback given instantaneously to students was extremely helpful for them and for the teachers, enabling teachers to focus specifically on particular areas of need rather than broad schemes of work that would address global issues in language learning for that age group. One challenge to the students which they often protested, was that in their speed to respond before time ran out, they made spelling mistakes due to mistyping words. This would lower their grade and not be a realistic representation of their abilities. As this group of students will be passing an exam in their final year of study in 2026, it was found that honing the language skills in the first part of their two-year program; helped them better understand the methodology which in turn left them more time to focus on the in-depth analysis of the program content. For students struggling with basic language skills, the use of AI helped increase their grade point average. The main limitation noticed by teachers and students was that most students produced typed work, which did not help them practice their handwriting which would be needed the day of the exam as this was an option that would require payment. Since this project was experimental, we thought it better to have more information before committing to a subscription.