SCIENTIFIC ASPECTS OF ENSURING SECURITY AND TRANSPARENCY OF ARTIFICIAL INTELLIGENCE ALGORITHMS IN THE CONTEXT OF DIGITAL TRANSFORMATION OF PUBLIC ADMINISTRATION

Abstract

The article examines the scientific aspects of ensuring security and transparency of artificial intelligence algorithms in the context of digital transformation of public administration. It is substantiated that the implementation of artificial intelligence technologies enhances the efficiency of administrative processes, but at the same time is associated with increasing security risks, including cyber threats, data breaches, algorithmic bias, and lack of transparency in decision-making. The main security threats to AI algorithms are systematized and their multi-level structure is identified. An analytical model for assessing the level of AI implementation in public administration bodies is proposed, along with an algorithmic model for ensuring security and transparency that integrates technical and governance aspects of AI systems.