<![CDATA[Artificial intelligence (AI) tutors are gaining traction in African education to facilitate personalised learning, feedback, evaluation, and educational access. However, the evidence available is dispersed in terms of country, level of education, technology and context of implementation. This systematic literature review aimed to aggregate evidence regarding types, educational impacts, implementation conditions and sustainability of adaptive AI tutors and personalised learning systems in African education in the period 2019- March 2026. The review was designed to combine the Scientific Procedures and Rationales for Systematic Literature Reviews with the PRISMA 2020 reporting guidelines. A total of 312 records were found in the databases searched, 54 of which were duplicates, 258 were screened, and 26 studies were included in the final analysis. The accompanying research studies included secondary, primary, basic, vocational, teacher, engineering, and higher education settings, and focused on the use of intelligent tutoring systems, adaptive learning platforms, mobile AI tutors, educational robotics, generative AI, predictive analytics, and AI-supported assessment. Overall, the evidence suggested gains for learner engagement, personalised pacing, formative feedback, academic outcomes, and teaching efficiency, but the extent and ability to generalise findings across studies were highly variable. The need for greater connectivity, a reliable electricity supply, devices, teacher preparation, funding, data privacy concerns, algorithmic bias, and cultural and linguistic localisation were recurrent barriers to implementation. The review shows that the interaction of adaptive technological functions, active participation of the learner, the facilitation of the teacher and the enabling institutional conditions pave the way for the emergence of educational results based on Programmed Logic for Automatic Teaching Operations and Constructivist Learning Theory. This study brings an Africa-centred Science and Technology framework depicting the mobile compatibility, contextual localisation, ethical regulation and human-in-the-loop model as the most likely features of adaptive AI tutors that will support equitable learning.]]>
Copyrights © 2026
