Uman National University | today: 12/28/2025
| Author(s) |
Бондаренко Н. В., , , Яковенко О. Р., , , Щербатюк О. П., , , |
||
|---|---|---|---|
| Category | Economics | ||
| year | 2025 | issue | Issue 107 part 2 |
| pages | 98-110 | index UDK | 336.71:004.8 | DOI | 10.32782/2415-8240-2025-107-2-98-110 (Link) |
| Abstract | The purpose of the study is to systematize current approaches to the use of AI, SupTech and RegTech technologies in banking supervision, assess their role in risk-based supervision, and identify challenges associated with the implementation of intelligent systems. The objectives include: analyzing the technological foundations of SupTech tools, examining the international experience of using AI in supervisory practices, evaluating the dynamics of AI/ML adoption by central banks in 2022–2025, and determining constraints that affect the reliability and transparency of intelligent models. The methodological framework combines a system approach, comparative analysis, economic-statistical methods, and logical generalization. System analysis was applied to define the structure and functions of SupTech technologies. Statistical methods were used to study trends in cashless payments and data volumes. Comparative analysis made it possible to examine supervisory practices of regulators in Australia, the EU and Singapore, as illustrated by SupTech platforms such as MAI, Athena, Virtual Lab and integrated monitoring systems. Logical and analytical tools were applied to evaluate risks and limitations associated with AI-based models. The results demonstrate that AI significantly enhances supervisory efficiency by enabling real-time data monitoring, anomaly detection, automated compliance checks, risk modelling, and early warning mechanisms. The international comparison confirms that leading regulators actively leverage various AI technologies, including machine learning, natural language processing, sentiment analysis and Generative AI. Moreover, the empirical data from 2022–2025 show consistent growth in the use of AI/ML instruments by central banks – from 50% in 2022 to 73.5% in 2025 – indicating a systemic shift toward digital supervision. At the same time, the study highlights critical challenges such as algorithmic bias, model instability, and limited interpretability, which may undermine regulatory transparency and decision reliability. The conclusions emphasize that AI is becoming an essential component of modern supervisory architecture, enabling the transition to risk-based, data-driven and forward-looking supervision. However, the effectiveness of AI depends on the development of robust data governance systems, model validation procedures, ethical standards, and transparent regulatory frameworks. Strengthening these elements will allow AI technologies to support financial stability, enhance market discipline, and improve the resilience of supervisory institutions in a rapidly changing digital environment. | ||
| Key words | SupTech; RegTech; artificial intelligence; machine learning; banking supervision; risk-based supervision; financial stability; compliance; digitalization | ||