Functions of civil-law liability in the context of AI development: traditional aspects and new horizons

This article provides a comprehensive analysis of the functions of civil liability in the context of the rapid development and widespread adoption of artificial intelligence (AI) systems. It explores the essence and purpose of civil liability in the digital era. The study examines how the established functions of objective law (regulatory and protective) and civil liability (compensatory and preventive) manifest and acquire new significance in interactions with AI systems. Special attention is given to the interplay between the regulatory and protective functions, with the European Union’s AI Act serving as a prominent example of their implementation. The article demonstrates that the regulatory function in the AI Act serves as a prerequisite for the protective function, which ultimately aims to effectively safeguard individuals’ health, safety, and fundamental rights from the potentially harmful impacts of AI. For the first time, the traditional understanding of the regulatory function of law in the AI era is expanded and refined by identifying three new forms of its embodiment, each responding to contemporary technological challenges: Innovative - legal mechanisms should not hinder but encourage the development of safe and ethically compliant AI systems by promoting best practices and transforming liability from a reactive tool into a mechanism for fostering responsible Coordinative - ensures the establishment of uniform requirements for the functioning of the digital market, harmonizing liability rules for AI-related harm at the international level, which is critical for cross-border AI systems and promotes predictability and stability in civil Ethico-legal - upholds fairness and legal order in civil relations, encouraging developers and operators of AI to act ethically, fostering accountability, building trust in AI tools, and integrating ethical principles into legal mechanisms, particularly through defining conditions for compensation and circumstances allowing exemption from liability. The study confirms that the compensatory function remains the most relevant and foundational in the context of civil liability involving AI. It emphasizes the critical interaction between the regulatory function of objective law and the preventive function of civil liability, manifested through mandatory risk management systems, data quality requirements, and enhanced «AI literacy» among stakeholders to minimize potential risks before they arise. The article also furthers the discussion on the punitive function of civil liability.