Artificial Intelligence in Customs Compliance and Risk Management: Evidence Gaps and Research Priorities for Sub-Saharan Africa

Abstract

This systematic review examines the integration of Artificial Intelligence (AI) in customs compliance and risk management, within Customs Administrations and how it can be tailored to Sub-Saharan Africa Customs Administrations. The study synthesizes 38 primary sources and identifies a transition from rule-based systems to predictive modelling, particularly for enhancing fraud detection and revenue optimization . Further, for SSA Customs Administrations, findings reveal a significant "revenue paradox" which demands an urgent need for adoption, however, infrastructure deficits and symbolic institutional adoption often hinder the realization of adoption and use of AI in customs compliance and risk management. The review also highlights critical evidence gaps regarding regional scalability and the psychological dimensions of stakeholder trust. The study emphasizes the necessity of localized frameworks to ensure sustainable digital transformation. The paper concludes with administrative recommendations for enhancing operational transparency and algorithmic sovereignty in emerging digital economies.