Вплив АI-автоматизації на архітектуру систем забезпечення безпеки програмних рішень

Abstract

The article examines the impact of the active integration of artificial intelligence systems on the architecture of security assurance for software solutions in the context of a reduced human role in IT companies. The increasing use of AI tools in development, testing, and security monitoring processes is transforming traditional approaches to building secure systems. Particular attention is paid to the risks of excessive automation, the emergence of new attack vectors (prompt injection, data poisoning), as well as the necessity of implementing Zero Trust principles for AI agents. The study substantiates the need to revise architectural security models taking into account the autonomy and scalability of AI components.