The article focuses on changing the tools of public policy under the influence of artificial intelligence (AI). The aim of the study was to conceptualise the role of legal mechanisms in the transformation of management processes and the formation of new models of interaction between the state, technologies, and society. The methodological basis of the study was a qualitative approach based on policy analysis using elements of case studies and comparative analysis. This approach made it possible to identify key patterns in the interaction of legal regulation, institutional practices and the use of intelligent systems in public administration. The paper analysed the impact of the introduction of intelligent solutions in the executive and legislative branches on the architecture of law and standards of effective governance. The international experience of legal support for national security in the conditions of digital autonomy of systems was considered. Strategic directions for updating public administration are identified to ensure social justice and environmental sustainability. The study found that the introduction of AI contributes to the transformation of the architecture of law, the formation of algorithmic forms of management and the transition to adaptive, risk-oriented models of regulation. It was proven that effective legal regulation of AI ensures increased transparency, accountability, and quality of management decisions, and also minimises the risks associated with the use of algorithmic systems. It was proposed to consider legal regulation of AI not only as a reaction to technological challenges, but as a proactive mechanism for the formation of an innovatively oriented, socially responsible and safe model of public administration. The results of the study can be used to develop effective strategies for the digital transformation of public administration and improve the legal principles of regulating AI
law; tools; sustainable development; effective management; national security
Received 09.02.2026, Revised 04.05.2026, Accepted 28.05.2026 Published 25.06.2026
Retrieved from Volume 19, No. 1, 2026
https://doi.org/10.56318/dg/1.2026.46
Pages 46-58