AI Tutoring in Primary ICT Education: The CARE Conceptual Framework

[conference]

Full reference

Ioannis Berdousis, Evi Togia, Vassilis Poulopoulos, and Manolis Wallace, AI Tutoring in Primary ICT Education: The CARE Conceptual Framework, International Conference on Human-AI Collaboration & Augmented Intelligence (HAICAI 2026), Athens, Greece, April 23–24, 2026

Abstract

Artificial Intelligence (AI) tutoring systems are increasingly integrated into educational settings, creating the need for pedagogically grounded frameworks that clarify the underlying principles guiding their design and implementation. This paper proposes the CARE conceptual framework, an acronym for Curriculum alignment, Agency, Reflective epistemic scaffolding, Equity, as a model for AI tutoring in primary computing education. CARE integrates four interrelated dimensions: curriculum alignment, teacher agency and oversight, reflective epistemic scaffolding, and equity, including gender and inclusion considerations. Together, these dimensions conceptualize AI tutoring as an educational practice embedded within structured learning goals, guided cognitive support processes, inclusive participation, and sustained teacher supervision. Drawing on curriculum theory, sociocultural perspectives on learning, inclusive education research, and responsible AI principles, CARE provides a coherent lens for aligning AI tutoring with core educational values. The framework establishes a conceptual foundation for the responsible integration of AIsupported tutoring in primary computing education.