Use of artificial intelligence algorithms for assessing corruption risks in the activities of law enforcement agencies in accordance with international standards

Abstract

The digitalization of public administration underscores the need for tools to detect corruption risks early in law enforcement agency activities. The use of artificial intelligence algorithms enables the processing of large amounts of information to produce objective assessments of the integrity of official activities, taking into account international requirements for transparency and accountability of law enforcement agencies. The purpose of the study is to substantiate the possibilities of using algorithmic models to identify and assess corruption risks in the law enforcement sector in accordance with international anti-corruption standards. To address the tasks set, a system analysis, a comparative legal method, generalization of the international regulatory framework, and elements of data analytics and machine learning were used. The source basis was international anti-corruption documents, regulatory acts and digital audit practices. The feasibility of using artificial intelligence algorithms for automated detection of corruption risks in law enforcement agency activities by analyzing financial transactions, personnel decisions, official communications and internal control results was substantiated. It has been established that machine learning methods can identify hidden patterns and atypical behavioral models that may indicate potential abuse of official position. The effectiveness of using predictive analytics to anticipate risk situations at their early stages has been proven, which contributes to timely responses and the minimization of negative consequences. The main risk assessment indicators have been determined, including officials' access to information resources, the recurrence of atypical transactions, and the concentration of power and deviations from established decision-making procedures. It has been established that the implementation of algorithmic models increases the accuracy of internal controls, reduces the influence of subjective factors on the assessment process, and provides greater justification for management decisions in the field of corruption prevention. Therefore, integrating algorithmic data analysis systems into anti-corruption monitoring strengthens mechanisms for preventing offenses and increases transparency in law enforcement activities, but it requires harmonizing national procedures with international digital governance standards, improving data processing algorithms and unifying requirements for the use of information resources.