Civil Liability for Artificial Intelligence Damages:Towards a Renewed Theory of Compensation

Abstract

The rapid proliferation of artificial intelligence technologies across economic, social, medical, and industrial domains generates unprecedented legal challenges that traditional civil liability frameworks are ill-equipped to address. AI systems increasingly make autonomous decisions and produce unforeseeable outcomes, complicating the application of classical standards premised on personal fault, causal attribution, and identifiable liability. This study critically examines the inadequacy of existing legal frameworks in addressing AI-generated harms, drawing on emerging legislative developments in the European Union and comparative legal systems. Through a comparative analytical methodology, the research evaluates various proposed civil liability theories and their suitability for ensuring fair compensation for AI damage victims. The study concludes by advocating a renewed compensation theory grounded in risk distribution and strict objective liability, complemented by specialized compensation funds and mandatory AI risk insurance.