Odpowiedzialność za szkody wynikające z działania systemów sztucznej inteligencji
Liability for harm resulting from the operation of artificial intelligence

Summary/Abstract: Artificial intelligence (AI) increasingly accompanies us in our professional and private lives. Alongside its undeniable benefits, it can also be a source of harm. The purpose of this article is to present the issue of legal liability for damage resulting from the operation of AI systems. The problem is significant because research indicates that liability for damage caused by AI constitute one of the key external obstacles to the adoption of artificial intelligence. The article identifies major challenges in determining liability, including: autonomy, constant adaptation, limited predictability, and lack of transparency. Using primarily the dogmatic-legal method, it outlines the general EU approach to regulating AI systems, which is based on risk analysis and management of the entire product life cycle, and presents the EU legal framework for liability for AI-related harm. The framework provides a two-track procedure for pursuing claims: under the liability of economic operators for defective products and under national tort liability systems based on the principle of fault. The article contributes to the discussion on the optimal regime of liability for harm caused by AI systems and evaluates the impact of these rules from the perspective of innovative enterprises and users. The study identifies key problem areas, including: partial discretion in classifying AI systems as high-risk, the regulatory focus on product safety risks with insufficient attention to other fundamental rights, and the lack of harmonization of national tort regimes resulting in different evidentiary standards adopted by national courts.

• Details
• Contents