Analysis of the main techniques and tools to combat money laundering: a review of the literature

PurposeThe purpose of this paper is to systematically review and evaluate recent anti-money laundering (AML) research, focusing on methodological shifts toward machine learning and network analysis, and identify key challenges and future directions for effective and ethical AML.Design/methodology/approachThis is a systematic review that follows Preferred Reporting Items for Systematic Reviews and PRISMA guidelines. An analysis of 45 studies (2017–2024) was conducted via Google Scholar using structured content analysis with a bi-dimensional framework (methodology and contextual applicability).FindingsAML research shows a paradigm shift from statistics to machine learning and network analysis. Mixed methods are increasingly important. Key challenges include cryptocurrencies, balancing detection with privacy and model interpretability/scalability. The literature shows significant variation in methods and results across operational contexts, but few studies offer direct comparisons of their relative effectiveness. Network analysis effectiveness depends on regulatory context and data sharing. The reviewed studies reveal ongoing discussion and varied approaches regarding model complexity versus practical applicability in diverse settings. Similarly, a debate on the factors influencing network analysis effectiveness emerges, frequently pointing to the critical roles of regulatory frameworks and data-sharing capabilities, though without a unified consensus on optimal implementation across all contexts.Research limitations/implicationsThis study reveals the need for research into adaptable models, context-specific solutions, privacy-preserving analytics and the interplay between AML evolution and criminal adaptation.Practical implicationsThis study recommends layered analytics, data-sharing frameworks and interpretability for FIs; graduated compliance, sandboxes and outcome metrics for regulators; and interoperable, tailored and simplified solutions for tech providers.Social implicationsThe broader implications of this research extend beyond immediate technical considerations to encompass significant societal impacts. The findings of this study reveal complex interactions between surveillance capabilities, privacy considerations and regulatory effectiveness. The analysis suggests that successful AML frameworks must balance competing demands: the need for comprehensive monitoring systems against legitimate privacy concerns and the requirement for transparent processes against the complexity of modern financial networks. These considerations have substantial implications for policy development and regulatory frameworks.Originality/valueThis review applies a unique bi-dimensional framework for analysis. It peaks at the contextual variance influencing network analysis effectiveness, questions a straightforward “complexity equals superiority” assumption by pointing to practical implementation factors and notes the emerging potential of hybrid models. Additionally, this study identifies critical gaps, such as the co-evolution of methods and counter-methods, and various implementation barriers.