ON MODELS OF LEGAL REGULATION OF ARTIFICIAL INTELLIGENCE SYSTEMS BASED ON RISK DIFFERENTIATION

Author(s): Fedyunin A.E., Peretyatko N.M.

Rubric: Specific issues of law and law enforcement

DOI: 10.21777/2587-9472-2026-2-70-76

Release: 2026-2 (50)

Pages: 70-76

Keywords: artificial intelligence, legal regulation, risk differentiation, legal regulation model, technology, technological solution, regulatory requirements

Annotation: The relevance of this article stems from the challenges arising in the legal regulation of artificial intelligence systems due to insufficient consideration of the risks associated with the use of this technology. The object of this study is the legal relationships arising in the regulation of artificial intelligence systems. The subject of this study is legal regulation models based on risk differentiation. The research methodology draws on classical methods of scientific analysis, general scientific, and specialized legal methods (formal legal, logical, and analytical), enabling the identification and resolution of issues related to the legal regulation of artificial intelligence systems using a risk-based approach. It is concluded that systems classified as high-risk are subject to more stringent control and monitoring, while low- and minimal-risk systems may meet less stringent requirements. All existing, developing and prospective artificial intelligence systems must comply with social security standards, while a horizontal model of legal regulation based on risk differentiation appears to be the most optimal. In the text of the article, the Russian Federation is abbreviated as RF.

Bibliography: Fedyunin A.E., Peretyatko N.M. ON MODELS OF LEGAL REGULATION OF ARTIFICIAL INTELLIGENCE SYSTEMS BASED ON RISK DIFFERENTIATION // Journal of Legal Sciences. – 2026. – № 2 (50). – С. 70-76. doi: 10.21777/2587-9472-2026-2-70-76