Reframing pre-service teacher competencies in the age of AI-mediated CLIL instruction: a theoretical review

Abstract

Background – Artificial intelligence (AI) is reshaping the landscape of education, particularly in bilingual and subject-integrated contexts such as Content and Language Integrated Learning (CLIL). Yet, pre-service teacher education often lacks explicit guidance on how to embed AI into CLIL pedagogy in ways that are pedagogically meaningful, ethically responsible, and culturally inclusive. In particular, limited attention has been given to how AI mediates pedagogical processes and learner–teacher interactions in CLIL contexts. Objectives – This study conducts a theoretical review to address three guiding questions: (1) What competencies for AI-mediated CLIL instruction are identified in existing literature? (2) How do established frameworks CLIL 4Cs, TPACK, AI Literacy, and Digital Ethics interact in this emerging field? (3) How can these insights be synthesized into a new competency framework for pre-service teacher education? Methods – This study reviewed interdisciplinary literature (2010–2025) from Web of Science, Scopus, and ERIC. A three-stage qualitative coding process (open, axial, selective) was applied by two independent coders (Cohen's κ = 0.82) to identify key competency domains across frameworks. Findings were synthesized to develop the CAIIL-TC framework, incorporating comparative insights from multiple regions. Results – The CAIIL-TC framework specifies five domains: AI Content Evaluation, Generative Input Design, Contextual Re-Authoring, Ethical and Reflective Use, and Collaborative Mediation. These domains extend beyond existing models by directly addressing AI-mediated teaching tasks. Conclusion – The CAIIL-TC framework provides a structured foundation for integrating AI into CLIL teacher preparation. It offers guidance for curriculum design, accreditation standards, and professional development, while underscoring the need for future empirical validation through classroom studies, longitudinal research, and cross-cultural comparisons.