Artificial Intelligence In Language Education: Exploring Prompting Strategies To Foster Argumentative Writing Skills

Résumé: The rapid advancement of Artificial Intelligence (AI), particularly Large Language Models (LLMs), calls for a thorough examination of not only of the opportunities for innovation but also the conditions necessary to foster a productive and informed human-machine relationship in education. This article explores the integration of prompt engineering as a critical transversal skill for effectively implementing AI-based technologies in language education while promoting the development of digital competencies among educators and learners. The study observes variations in interactions between learners and ChatGPT during educational activities designed to enhance argumentative writing skills. Specifically, it examines the reliability and feasibility of ChatGPT in providing meaningful and relevant feedback on argumentative writing through the analysis of task-specific interactions between secondary school students and the language model. Additionally, it explores how the iterative process of prompt construction and refinement adopted by participants shapes ChatGPT’s responses when evaluating learners’ argumentative texts. By analysing the impact of different prompt strategies on the chatbot’s outputs, the study offers practical guidelines for leveraging AI to foster language acquisition, AI literacy, and critical thinking through the evaluation, validation, and optimization of learners’ interactions with ChatGPT.

Publié dans la revue: Journal of Studies in Language, Culture and Society (JSLCS)