This study explores foreign language teachers’ perceptions and usage of generative AI tools in their pedagogical practices, especially in LMS-based writing instruction at a Japanese university. The study aims to determine attitudes towards, and the challenges of, integrating AI into LMS-based foreign language instruction. The participants were ten foreign language instructors (n=10) teaching English, Chinese, or German to Japanese university students with varying proficiency levels. Data were collected with a questionnaire which consisted of Likert-scale and open-ended items. Quantitative responses were analysed with descriptive statistics, and qualitative content analysis was applied to the responses on the open-ended items. Survey responses from the participating teachers revealed that ChatGPT and Grammarly were the most common AI tools to be aware of and to use in the foreign language classroom, with these being valued due to access to immediate feedback, supporting lexical and grammatical development in writing, and boosting writing confidence. Despite the benefits, teachers were concerned about accuracy issues, over-dependence on AI, students’ reduced creativity, and risks to academic integrity. Teachers reported that they modified their writing assessment methods to address these issues. Integration of AI into LMS-based instruction was limited and was perceived as challenging due to unclear monitoring procedures and uncertainty about fair evaluation of AI-assisted student work. The findings also suggest that there is a need for clearer institutional guidelines, training about AI for teachers, and pedagogical frameworks that balance the advantages of AI with the development of authentic writing skills.
Tanabe et al. (2026) studied this question.