Display omitted •Science history and progress in the application of AI in MBRs are identified.•The advances in AI techniques for MBRs are reviewed and critically discussed.•Artificial neural network is the most implemented AI techniques in MBR.•Future outlook of the implementation of AI in MBRs are discussed.•Recommendation for future studies in this field are presented and discussed. Efforts are currently in progress to commercialize membrane bioreactor (MBR) technologies already developed at laboratory and pilot scale. To attain this goal, the efficiency of MBRs needs to be high, and they should be sustainable, reliable and cost-effective. Adoption of artificial intelligence (AI) is anticipated to have a positive impact on these criteria. This paper covers the AI-based models used in the treatment of wastewater from various sources, and discusses the advantages and disadvantages of each model. The existing gaps to push for the commercialization of MBR technologies are discussed to provide state-of-the-art insights for future research. The conclusions and discussions presented in this review show that AI models are useful to predict the performance of MBR technologies to recover clean water from polluted sources. However, further efforts are still needed to reach an excellent match between the predictions made by the AI-based techniques and the experimental results to deal with high strength and highly polluted effluents. This can be achieved through modification and/or integration of the existing AI-based methods. Also, the development of appropriate variables to optimize the performance of MBRs, and improving their efficiency to deal with recalcitrant pollutants such as contaminants of emerging concern(CECs) are among the priorities to promote the application of MBR technologies in real-scale applications.