Beyond Automation: AI as a Pedagogical Mediator in Collaborative Learning

Abstract

Research on Artificial Intelligence in Education (AIED) has predominantly conceptualized AI as a tool for automation, personalization, or instructional efficiency, often privileging individual learning outcomes over social and relational processes. This instrumental orientation has resulted in a theoretical misalignment between contemporary AI systems, particularly generative and interactive models, and established pedagogical understandings of collaborative learning as a socially mediated, co-regulated process. As a consequence, the role of AI in shaping interaction, epistemic agency, and shared regulation in collaborative learning remains under-theorized. This paper advances a conceptual framework that reconceptualizes AI as a pedagogical mediator in collaborative learning environments. Drawing on socio-cultural theory, distributed cognition, connectivist perspectives, and recent research in AI-supported collaborative learning, the framework positions AI as an active participant in the orchestration of interaction, epistemic sense-making, and regulatory processes, without displacing learner or teacher agency. Rather than treating AI as a tutor, peer, or automated orchestrator, the framework foregrounds mediation as a pedagogical function through which agency, authority, and responsibility are dynamically redistributed across human and non-human actors. The paper makes three contributions. First, it offers a theoretically grounded reconceptualization of AI’s role in collaborative learning that moves beyond tool-centric and individualistic paradigms. Second, it articulates key mediating functions of AI, interactional, epistemic, and regulatory, and examines their implications for collaborative learning design. Third, it critically addresses ethical, professional, and governance challenges associated with AI-mediated collaboration, with particular attention to equity, transparency, and teacher professional judgment.