en

Law Journal of the National Academy of Internal Affairs

  • Received 01.07.2025,
  • Revised 11.10.2025,
  • Accepted 25.11.2025
Download article Download article
Volume 15, No. 4, 2025
  • model explainability; federated learning; graph neural networks; graph autoencoders; reduction of false positives; cryptocurrency operations; privacy of transactional data
  • https://doi.org/10.63341/naia-chasopis/4.2025.75
  • Pages 75-87

The aim of the study was to examine the effectiveness of applying artificial-intelligence algorithms in the financial-monitoring system. The methodology included a comparative analysis of the practices of the United States, the European Union and Ukraine, a case analysis of international financial incidents (United States, European Union, Ukraine), and an assessment of the regulatory framework. It was established that the regulatory basis for a rule-based system is enshrined in the international standards of the Financial Action Task Force and implemented in the legislation of the European Union, the United States and Ukraine, which ensures transparency of control while simultaneously reducing adaptability to new schemes. In the United States, legal norms ensure strict reporting and sanctions, yet these norms demonstrated critical gaps in rule-based monitoring. In the European Union, multi-level directives strengthened centralised supervision while preserving the problem of bureaucratic inertia. In Ukraine, cryptocurrency Anti-Money Laundering still remained limited. It was identified that in the 2024 judgments of the High Anti-Corruption Court there were recorded cases of using fractional land deals totalling more than 3.1 million dollars, as well as large-scale organised schemes that rule-based systems did not detect. The 2024 statistics (1.75 million financial reports, UAH 12.1 billion of seized assets) demonstrated the scale of Ukraine’s Anti-Money Laundering system but revealed the need to reduce false positives and strengthen analytics. Alignment with international practice showed that the effectiveness of future anti-money-laundering solutions for Ukraine is possible only if the regulatory framework is combined with AI models that meet the requirements of the Financial Action Task Force and the European Union. The practical significance lay in applying the results by banks, regulators and law-enforcement bodies of Ukraine to reduce false positives, detect complex schemes and adapt monitoring systems to international standards

References

  1. Ajagbe, S.A., Majola, S., & Mudali, P. (2025). Comparative analysis of machine learning algorithms for money laundering detection. Discover Artificial Intelligence, 5, article number 144. doi: 10.1007/s44163-025-00397-4.
  2. Altman, E., Blanuša, J., von Niederhäusern, L., Egressy, B., Anghel, A., & Atasu, K. (2023). Realistic synthetic financial transactions for antimoney laundering models. In A. Oh, T. Naumann, A., Globerson, K., Saenko, M., Hardt & S., Levine (Eds.), Advances in neural information processing systems. Red Hook: Curran Associates, Inc.
  3. Bakry, A.N., Alsharkawy, A.S., Farag, M.S., & Raslan, K.R. (2024). Automatic suppression of false positive alerts in anti-money laundering systems using machine learning. The Journal of Supercomputing, 80, 6264-6284.
  4. Balusamy, S., Rengasamy, R., & Aravind, J. (2025). Protecting financial transactions and cryptocurrency networks from fraud using AI-powered blockchain technology. In Proceedings of the 2025 global conference in emerging technologies (pp. 1-6). Pune: IEEE.
  5. Chen, Y., Chen, Z., & Amin, H.U. (2025). LG-VGAE: A local and global collaborative variational graph autoencoder for detecting crypto money laundering. Knowledge and Information Systems, 6, article number 586.
  6. Di Gennaro, M., Panebianco, F., Pianta, M., Zanero, S., & Carminati, M. (2025). Amatriciana: Exploiting temporal GNNs for robust and efficient money laundering detection. arXiv. doi: 10.48550/arXiv.2506.00654.
  7. Dumitrescu, B., Baltoiu, A., & Budulan, S. (2022). Anomaly detection in graphs of bank transactions for anti money laundering applications. IEEE Access, 10(96). doi: 10.1109/ACCESS.2022.3170467.
  8. Eurojust. (2024). International operation leads to seizure of 2 000 fake works of art with potential losses of EUR 200 million. Retrieved from https://surl.li/vvgdgf.
  9. European Banking Authority. (2025). Opinion and report on money laundering and terrorist financing risks affecting the EU’s financial sector. Retrieved from https://surl.lt/mprhoo.
  10. Financial Action Task Force (FATF). (2025). International standards on combating money laundering and the financing of terrorism & proliferation. The FATF Recommendations. Paris: FATF.
  11. FinCEN. (2024). FinCEN assesses record $1.3 billion penalty against TD Bank, N.A. and TD Bank USA, N.A. U.S. Department of the Treasury. Retrieved from https://surl.lu/uhpaml.
  12. FinCEN. (2025). FinCEN year in review FY 2024 [Infographic]. Retrieved from https://www.fincen.gov/system/files/shared/FinCEN-Infographic-Public-2025-508.pdf?utm_source=.
  13. Fortunenko, A., Shkondin, O., Stadnyk, O., & Hontaruk, I. (2025). Ukraine: Evolving antimoney laundering framework expands corporate criminal liability. Retrieved from https://surl.li/mjofyb
  14. Gandhi, H., Tandon, K., Gite, S., Pradhan, B., & Alamri, A. (2024). Navigating the complexity of money laundering: Anti–money laundering advancements with AI/ML insights. International Journal on Smart Sensing and Intelligent Systems, 17(1).
  15. Horobets, N., Reznik, O., Maliyk, V., Vyhivskyi, I., & Bobrishova, L. (2025). Artificial intelligence technologies in banking: Challenges and opportunities for anti-money laundering in the context of EU regulatory initiatives. Journal of Money Laundering Control, 28(4-5), 593-608. doi: 10.1108/JMLC-03-2025-0041.
  16. Jensen, R.I.T., & Iosifidis, A. (2023). Qualifying and raising anti-money laundering alarms with deep learning. Expert Systems with Applications, 214, article number 119037. doi: 10.1016/j.eswa.2022.119037.
  17. Johannessen, F., & Jullum, M. (2023). Finding money launderers using heterogeneous graph neural networks. arXiv. doi: 10.48550/arXiv.2307.13499.
  18. Kalabukhova, S. (2025). Methods of analysis in the financial intelligence system. Finance of Ukraine, 4, 46-57. doi: 10.33763/finukr2025.04.046.
  19. Karthikeyan, G.K., & Bhowmik, B. (2025). Enhancing money laundering detection in bank transactions using GAGAN: A graph-adapted generative adversarial network approach. International Journal of Data Science and Analytics, 20, 6301-6331. doi: 10.1007/s41060-025-00823-x.
  20. Konstantinidis, G., & Gegov, A. (2024). Deep neural networks for antimoney laundering using explainable artificial intelligence. In Proceedings of the 2024 IEEE 12th International Conference on Intelligent Systems (pp. 430-435). Red Hook: IEEE.
  21. National Police of Ukraine – Zaporizhzhia Region. (2025). Zaporizhzhia police uncovered a man who misappropriated over UAH 3,500,000 in cryptocurrency. Retrieved from https://surl.li/ylxqgt.
  22. National Police of Ukraine. (2023). Over UAH 1 billion laundered through resale of securities on the stock exchange: Law enforcement officers served notices of suspicion to members of a criminal group. Retrieved from https://surl.li/upjfav.
  23. National Police of Ukraine. (2024). Over UAH 200 million laundered via VAT schemes: law enforcement exposed organizers using enterprises employing persons with disabilities. Retrieved from https://surl.li/czvubg .
  24. Ovsianiuk, D. (2024). Intelligence cycle as the basis of analytical activity in combating drug-related crime. Law Journal of the National Academy of Internal Affairs, 14(2), 95-104. doi: 10.56215/naia-chasopis/2.2024.95.
  25. Ovsianiuk, D., & Ustymenko, O. (2024). Exchange of information as a form of international cooperation in combating drug trafficking. Novum Jus, 18(1), 181-216. doi: 10.14718/NovumJus.2024.18.1.7.
  26. Pocher, N., Zichichi, M., Merizzi, F., Shafiq, M. Z., & Ferretti, S. (2023). Detecting anomalous cryptocurrency transactions: An AML/CFT application of machine learning-based forensics. Electronic Markets, 33, article number 37. doi: 10.1007/s12525-023-00654-3.
  27. Poon, C.-H., Kwok, J., Chow, C., & Choi, J.-H. (2025). LineMVGNN: Anti-money laundering with line-graph- assisted multi-view graph neural networks. AI, 6(4), article number 69. doi: 10.3390/ai6040069.
  28. Pousette, T., & Rosendal, A. (2024). Explainable anti-money laundering. (Master’s thesis, Chalmers University of Technology, Gothenburg, Sweden).
  29. Purohit, M., & Barot, H. (2024). Data-driven forensics: Unveiling the hidden depths of financial crime. In S.K. Shandilya, D. Sujay & V.B. Gupta (Ed.), Advancements in cybercrime investigations and modern data analytics. Boca Raton: CRC Press.
  30. Rana, S., Aggarwal, S., Jagirdar, S.S., & Jain, S. (2025). Insights in banking analytics and regulatory compliance using AI. Retrieved from https://www.igi-global.com/gateway/book/358948.
  31. Reite, E.J., Karlsen, J., & Westgaard, E.G. (2025). Improving client risk classification with machine learning to increase anti-money laundering detection efficiency. Journal of Money Laundering Control, 28(1), 93-107. doi: 10.1108/JMLC-03-2024-0040.
  32. Singh, V. (2025). Policy optimization for Anti-Money Laundering (AML) compliance using AI techniques: A machine learning approach to enhance banking regulatory compliance. International Journal of Engineering Research & Technology (IJERT), 14(4).
  33. State Financial Monitoring Service of Ukraine. (2023). Methodological recommendations for analysing money laundering trends related to modern forms of slavery in Ukraine. Retrieved from https://surl.lt/aglghq.
  34. State Financial Monitoring Service of Ukraine. (2025). Report of the State Financial Monitoring Service of Ukraine for 2024. Retrieved from https://fiu.gov.ua/assets/userfiles/0350/2025/REPORT2024.pdf.
  35. U.S. Department of Justice, Office of Public Affairs. (2025a). Nevada man sentenced for over $11 M COVID-19 relief fraud and money laundering scheme. Retrieved from https://surl.li/gmfdzu.
  36. U.S. Department of Justice, Office of Public Affairs. (2025b). Founder and CEO of AML Bitcoin sentenced to seven years in prison for multi-million-dollar fraud scheme. Retrieved from https://surl.li/dhzpjy.
  37. United Nations Conference on Trade and Development (UNCTAD). (2023). First-ever official data on illicit financial flows now available. Retrieved from https://surl.lt/xtcjng.
  38. Wang, S., Wang, P., Wu, B., Zhu, Y., Luo, W., & Pan, Y. (2024). Structural entropy minimization combining graph representation for money laundering identification. International Journal of Machine Learning and Cybernetics, 15, 3951-3968. doi: 10.1007/s13042-024-02129-z.
  39. Wójcik, F. (2024). An analysis of novel money laundering data using heterogeneous graph isomorphism networks: FinCEN Files case study. Econometrics. Ekonometria. Advances in Applied Data Analysis, 28(2), 32-49.