THE PROCEDURE FOR IDENTIFYING THE RISKS OF INTRODUCING ARTIFICIAL INTELLIGENCE INTO PUBLIC ADMINISTRATION

Abstract

Identifying the risks of implementing artificial intelligence (AI) in public administration is a key step in ensuring the effectiveness and security of this process. This procedure involves a systematic analysis of potential threats that may arise when integrating AI technologies into administrative processes, including decision-making, service provision and data management. The first stage is to identify areas of application of AI, such as automation of routine processes and analysis of large volumes of data. Next, technical, ethical, social and legal risks should be assessed.Technical risks may include errors in algorithms and system failures, while ethical risks include data confidentiality issues and algorithm bias. Social risks may manifest themselves in reducing citizens’ trust in public administration, and legal risks relate to the inconsistency of AI implementation with applicable legislation. To identify risks, it is advisable to use methods such as SWOT analysis and expert survey.The aim of the article is to study modern theoretical approaches to risk assessment and propose a model for the procedure for identifying risks of implementing AI in public administration.The research methodology involves the use of a combination of qualitative and quantitative methods, including literature analysis and expert survey. The scientific novelty of the work lies in the systematic analysis of risks associated with the implementation of AI and the development of new approaches to their management. Risk management measures require regular revision to adapt to new conditions and technologies.Conclusions. The proposed model of the risk identification procedure allows for the effective identification and minimization of potential threats, which will contribute to increasing trust in state institutions and increasing the efficiency of public services.