Abstract This paper presents a research project, funded by Durham University, investigating the potential of Generative AI (GenAI) for Modern Foreign Languages (MFL) learning, teaching and assessment within the School of Modern Languages and Cultures (MLaC) at Durham University. The study explores how Generative AI can support writing skills and feedback in the MFL context. Through comparative analysis of tasks completed with and without ChatGPT, and feedback by both teachers and ChatGPT, the study investigates the pedagogical implications of emerging technologies in language teaching, with a focus on writing. The project involves 8 language areas, namely: Arabic, Chinese, French, German, Japanese, Italian, Russian, and Spanish, and recruited two participants for each language with a CEFR-level of B1-B2. Participants completed diagnostic surveys, written tasks with reflections, sought feedback from ChatGPT and their teachers and reflected on them, and engaged in semi-structured interviews and focus groups. The project highlights the potential of GenAI for supporting students’ learning and concerns about its implications for academic integrity and the development of linguistic skills, especially for beginner learners. It also emphasizes the need for clear institutional guidelines on the ways in which students can and cannot use GenAI to complete their academic work.
Versini et al. (Wed,) studied this question.