Legal Aspects of Using Artificial Intelligence in Cybersecurity: Regulation and Ethical Dilemmas

The article is devoted to the analysis of legal aspects of artificial intelligence (AI) implementation in the field of cybersecurity. In the context of rapid technological development, AI has become an integral tool in identifying and countering cyber threats through anomaly detection systems, automated firewalls, machine learning–based log analysis, threat intelligence, and incident response automation. However, the deployment of such technologies reveals a number of unresolved legal and ethical dilemmas that require urgent regulatory attention. The research identifies three core groups of legal issues: determination of liability, challenges of system autonomy, and the transparency of AI decision-making. The absence of a legal framework for assigning responsibility in cases where AI malfunctions or acts autonomously complicates the process of legal evaluation. The article emphasizes that Ukrainian legislation lacks a definition of “autonomous agent” or “electronic subject,” making it difficult to apply traditional norms of civil and administrative law. The problem of autonomy is particularly relevant, as modern AI systems can modify their behavior independently through machine learning processes. This raises questions about human oversight and the balance between innovation and regulatory control. Additionally, many AI systems function as “black boxes,” making it difficult – even for developers – to understand the logic behind certain decisions. This threatens the protection of human rights, including the right to information and due process. To address these challenges, the article proposes several legal and policy measures, such as the introduction of a dedicated law on AI, adoption of explainable AI principles, implementation of algorithmic audit standards, and development of ethical guidelines for AI developers and users. Keywords: artificial intelligence, cybersecurity, information law, legal responsibility, autonomous systems, artificial intelligence ethics, algorithmic transparency, legal regulation.