ANALYSIS OF CYBER ATTACKS USING MACHINE LEARNING ON THE INFORMATION SECURITY MANAGEMENT SYSTEMS
A.V. Habrylchuk, Віталій Сусукайло, Yevhenii Kurii, Sviatoslav Vasylyshyn · Komp'ûternì sistemi ta merežì · 2025
The article analyzes how modern machine learning algorithms are integrated into cyber threats, changing traditional cyberattack approaches. Artificial intelligence allows attackers to automate systems compromise and adapt their actions to real-time defense mechanisms. Detecting such attacks is one of the biggest challenges, as traditional cyber defense tools cannot always adequately respond to the speed and dynamism of threats created with the help of artificial intelligence. The article also examines the risks associated with using AI threats, including privacy compromise, damage to the reputation of organizations, and financial losses. The article proposes protection measures based on international standards, such as ISO 27001, to counter these challenges. In particular, it emphasizes the importance of implementing access controls, threat monitoring, ensuring data integrity, using cryptography, and conducting regular security audits. It also emphasizes the need to develop new tools to detect threats and prevent manipulations carried out using AI. Keywords: EvilProxy, PassGAN, DeepLocker, FaceSwap, Respeecher, ISO 27001:2022.