Mapping Contemporary AI-Education Intersections and Developing an Integrated Convergence Framework: A Bibliometric-Driven and Inductive Content Analysis

Artificial intelligence (AI) has rapidly permeated education since 2014, propelled by technological innovation and global investment, yet scholarly discourse on contemporary AI-Education intersections remains largely fragmented. The present study addresses this notable gap through a bibliometric-driven and inductive content analysis to inform future research and practice. A total of 317 articles published between 2014 and October 2024 were retrieved from WOSCC and Scopus following the PRISMA protocol. Keyword co-occurrence and co-citation analyses with VOSviewer (version 1.6.20) were employed to visualize the intellectual structures shaping the field, while qualitative inductive content analysis was conducted to address the limitations of bibliometric methods in revealing deeper thematic insights. This dual-method approach identified four thematic clusters and eleven prevailing research trends. Subsequently, through interpretive synthesis, five interrelated research issues were identified: limited congruence between technological and pedagogical affordances, insufficient bottom-up perspectives in AI literacy frameworks, an ambiguous relationship between computational thinking and AI, a lack of explicit interpretation of AI ethics, and limitations of existing professional development frameworks. To address these gaps pragmatically, thirty issue-specific recommendations were consolidated into five overarching themes, culminating in the Integrated AI-Education Convergence Framework. This framework advocates for pedagogy-centric, ethically grounded, and contextually responsive AI integration within interdisciplinary educational research and practice.