Artificial Intelligence in Education Bridging or Widening the Digital Divide in Developing Countries

Abstract

Artificial intelligence (AI) is rapidly transforming education through personalized learning, intelligent tutoring, automated assessment, generative content creation, learning analytics, academic research support, and educational administration. Although these technologies offer significant opportunities for improving educational quality and accessibility, their implications for developing countries remain complex. Educational systems in low- and middle-income countries frequently operate under constraints that include inadequate digital infrastructure, unreliable connectivity, limited financial resources, insufficient teacher preparation, linguistic diversity, and unequal access to technology. Consequently, AI has the potential both to reduce longstanding educational inequalities and to create a new layer of digital exclusion. This article provides an integrative review of the influence of AI on education in developing countries, with particular attention to teaching, learning, assessment, higher education, administration, inclusion, and educational equity. It synthesizes recent academic literature and international policy reports, with particular emphasis on developments associated with generative AI since 2022. The review identifies personalized learning, expanded access to educational expertise, teacher support, multilingual assistance, administrative efficiency, and improved accessibility as important opportunities. However, connectivity inequalities, affordability, algorithmic bias, inadequate AI literacy, academic integrity concerns, privacy risks, cultural and linguistic underrepresentation, and excessive dependence on automated systems present substantial challenges. To explain these competing effects, the article proposes an Inclusive AI in Education Framework in which infrastructure, affordability, human capacity, localization, governance, and pedagogical integration determine whether AI contributes to greater educational inclusion or reinforces existing inequalities. The paper argues that the educational value of AI in developing countries will ultimately depend less on the availability of sophisticated technologies than on the institutional and societal capacity to deploy them equitably, responsibly, and in accordance with local educational priorities.