BACKGROUND AND AIM: In the last two decades, the proportion of internet users has greatly increased worldwide. Data regarding internet addiction (IA) are lacking in Africa compared to other continents. This systematic review and meta-analysis aimed to estimate the pooled prevalence of IA in African countries. METHODS: We systematically sought relevant articles in PubMed/MEDLINE, EMBASE, PsycINFO and Cochrane database published before September 25, 2021. The risk of bias was assessed using the Joanna Briggs Institute tool, and we estimated the pooled prevalence of IA using a random-effects meta-analytic model. We followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. RESULTS: We included 22 studies (13,365 participants), and collected data from Egypt, Ethiopia, Morocco, Nigeria, South Africa, Tanzania and Tunisia between 2013 and 2021. The mean age of participants ranged from 14.8 to 26.1 years, and the most used tool for IA screening was the Young's 20-item Internet Addiction Test. The pooled prevalence rate of IA was 40.3% (95% CI: 32.2%-48.7%), with substantial heterogeneity. The pooled prevalence for Northern Africa was 44.6% (95% CI: 32.9%-56.7%), significantly higher than the prevalence in sub-Saharan Africa, which was 31.0% (95% CI: 25.2%-37.1%). The risk of bias was moderate for most studies, the certainty was very low, and we found no publication bias. DISCUSSION AND CONCLUSIONS: Four in every ten individuals was considered to have IA in Africa. Further research with methodological optimization seems needed, especially for IA screening tools and the representativity of some subregions.