Fostering Transversal Skills Through Open Schooling Supported by the CARE-KNOW-DO Pedagogical Model and the UNESCO AI Competencies Framework

Abstract

Abstract This study breaks new ground in educational theory by empirically validating how artificial intelligence (AI) competencies emerge through social learning ecosystems in underserved communities. Through an innovative cross-national mixed-methods design, we examined 330 secondary school students across the UK, Greece, and Brazil who developed AI competencies while addressing Sustainable Development Goals (SDGs) through open schooling initiatives. Our theoretical innovation synthesises UNESCO's AI Competencies Framework with the CARE-KNOW-DO pedagogical model, generating transformative insights into AI literacy. We conceptualise AI literacy as a dynamic interplay of understanding, applying, and creating AI through social practice, where learners develop ethical awareness (CARE), epistemic-technical competence (KNOW), and responsible agency (DO) within authentic learning ecosystems. Employing a concurrent triangulation methodology of teaching and learning, we collected comprehensive longitudinal data through teacher ethnographic accounts, student self-assessments, STEM researchers’ observations and granular analysis of AI-enabled project work over a year. This robust methodological approach revealed seven dimensions of AI-supported competency linked to transversal skills development across three distinct but interrelated domains. CARE (Understanding AI) highlights how proactive exploration and affective engagement emerge socially. KNOW (Applying AI) demonstrates how problem-solving, STEM for sustainability, and future prospects become embodied through application. DO (Creating AI) reveals how scientific citizenship and authentic learning with AI catalyse new forms of participation. This research advances sociocultural learning theory in three significant ways: (1) it empirically explicates how AI competencies develop through social practice in underserved communities, challenging existing theoretical paradigms; (2) it provides a comprehensive, evidence-based framework for equitable AI education integration across disciplines; and (3) it demonstrates how open schooling methodologies can foster transformative learning experiences that prepare students for an AI-driven future while promoting social justice and sustainable development. The findings have profound implications for educational policy and practice, suggesting a fundamental reconceptualization of AI literacy development in secondary education. This work establishes a new theoretical foundation for understanding how schools can systematically develop AI competencies while addressing critical issues of equity, inclusion, and sustainable development in an increasingly AI-mediated world.

Topics & Concepts