Artificial Intelligence in the Shadow Economy: Detecting and Enabling Financial Crime

This paper examines the interaction between artificial intelligence (AI) and financial crime dynamics across 85 countries from 2012 to 2023. By applying fixed-effects panel regression and instrumental variable techniques, we ask whether AI adoption genuinely reduces illicit financial flows and improves anti-money laundering (AML) risk scores. The findings are mixed: while AI seems to support financial integrity in well-governed environments, its impact is far from universal. In weaker institutional settings, AI may be ineffective—or worse, exploited. Our analysis suggests that AI's potential lies not just in the technology itself, but in the regulatory and governance frameworks surrounding it. To be effective, AI tools must operate within transparent, accountable systems. Accordingly, we argue that AI investments should be accompanied by regulatory reform and global cooperation. This study contributes to policy debates by providing empirical cross-country evidence and a practical roadmap for integrating AI into financial governance in a responsible manner.