Integrating AI in STEM Education in Africa: A Systematic Review of Best Practices and Perspectives

The integration of Artificial Intelligence (AI) in higher education is gaining global momentum, particularly within STEM (Science, Technology, Engineering, and Mathematics) disciplines. In Africa, however, the adoption of AI in STEM education remains fragmented and underexplored. This systematic review offers a novel and comprehensive synthesis of 46 peer-reviewed studies to assess how AI is being integrated into STEM education across African higher education institutions. Using the PRISMA framework, the review applies both thematic analysis and a PESTEL (Political, Economic, Social, Technological, Environmental, and Legal) lens to discover patterns, regional disparities, and systemic barriers. The study reveals three key innovations: first, it integrates a wide range of theoretical perspectives including Diffusion of Innovations, Constructivist Learning Theory, Cognitive Load Theory, and Postcolonial Theory to interpret the socio-technical and pedagogical dynamics of AI adoption. Second, it develops a strategic, context-sensitive framework to guide the equitable and sustainable implementation of AI in STEM education, aligned with the UN Sustainable Development Goals (SDGs 4 and 9). Third, it critiques Eurocentric approaches to AI adoption and calls for a decolonized, locally adaptive model of AI integration. Major findings include infrastructure deficits, insufficient lecturer training, ethical and policy gaps, and the digital divide all of which hinder AI’s transformative potential in African STEM education. Yet, the increasing use of AI tools like ChatGPT, Intelligent Tutoring Systems, and LMS platforms post-2020 signals a turning point. This review advances a nuanced, Africa-centered roadmap for AI in STEM education and contributes original theoretical and strategic insights to the global discourse on educational innovation.

Abstract licence: CC BY-SA

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.