Liability of Medical AI System Developers: Comparative Legal Analysis of International Regulations

This study presents a comparative legal analysis of liability regulation for developers of artificial intelligence-based medical systems in key jurisdictions: the European Union, the United States, and Asian countries (China, Japan). The research identifies two dominant liability models: strict liability (EU, China) and mixed liability (USA, Japan), analyzing their advantages and disadvantages. The study identifies key legal challenges, including the opacity of AI systems' decision-making processes, difficulties in determining responsible entities, and challenges in establishing causation. Specific mechanisms for harmonizing international regulatory approaches are proposed: establishing a differentiated liability regime based on risk levels, requirements for algorithm explainability, easing the burden of proof for victims, and implementing mandatory liability insurance. The work demonstrates the necessity of balancing innovation incentives with patient safety as a key principle for effective regulation of medical AI systems.