Civil Tort Liability for AI-Induced Harm: Lessons from EU Law for Vietnam’s Emerging Legal Framework

Background: Artificial intelligence increasingly causes harm that challenges conventional civil liability rules built around human fault, linear causation, and tangible products. This article examines how the European Union’s evolving approach to AI-related civil liability can inform Vietnam’s emerging legal framework. Its aim is to identify the main liability models developed in EU law, clarify the doctrinal difficulties that AI creates for traditional tort principles, and assess which elements may be adapted to Vietnam in a context-sensitive manner. Method: Methodologically, the study employs a qualitative legal case study design following Yin, treating the EU as an instrumental case, and analyses secondary legal and scholarly sources through thematic analysis in the Braun and Clarke sense. Results and Conclusions: The findings show that the EU is moving towards a layered, risk-based regime combining ex ante regulation with ex post liability, especially through Regulation (EU) 2024/1689 of the European Parliament and of the Council of 13 June 2024 (the EU AI Act), the proposed AI Liability Directive, and the Revised Product Liability Directive. Product liability and operator liability emerge as complementary mechanisms, while evidentiary facilitation becomes central where opacity and distributed responsibility hinder proof. The article concludes that Vietnam should pursue incremental reforms focused on risk differentiation, digital product liability, stronger operator duties, and improved access to evidence, all within a coherent AI governance framework.