Toward an Ethical and Pedagogical Framework for Generative AI in Foreign Language Education: Implications for Engineering and Technology Programs in Higher Education
TL;DR
An ethical and pedagogical framework for integrating GenAI into foreign language education and a six-stage AI mediated communicative task cycle that guides learners from task orientation, guided prompting, and drafting or rehearsal to critical revision, transparent submission or performance, and reflective transfer are proposed.
Abstract
Background: Generative artificial intelligence (GenAI) is changing higher education by expanding access to feedback, simulated interaction, and personalized support. In foreign language education, however, its pedagogical value remains contested because the same tools that may scaffold oral and written production can also encourage dependence, superficial correctness, authorship ambiguity, and uncritical automation. Objective: This conceptual article proposes an ethical and pedagogical framework for integrating GenAI into foreign language education, with specific implications for engineering and technology programs in higher education. Methods: The study develops an integrative conceptual synthesis based on transparent literature mapping conducted with Elicit and cross-checked through scholarly and institutional sources published mainly between 2022 and 2026, alongside foundational works on communicative competence, autonomy, mediation, and assessment. Results: The analysis yields the MAICA Framework, a Mediated AI Communicative Agency model organized around five interdependent dimensions: pedagogical mediation, communicative production, disciplinary transfer, learner agency, and ethical and critical AI literacy. The article also formulates a six-stage AI mediated communicative task cycle that guides learners from task orientation, guided prompting, and drafting or rehearsal to critical revision, transparent submission or performance, and reflective transfer. Conclusions: GenAI should not be treated as a shortcut for language performance or a replacement for teacher expertise. Its educational contribution depends on thoughtful task design, transparent assessment, teacher mediation, and students’ capacity to question, adapt, verify, and ethically declare AI assistance. The framework offers a practical conceptual contribution for language educators working with students who need communicative competence and responsible technological agency