The use of artificial intelligence (AI) chatbots in customer service shifts part of the interaction between businesses and consumers from human agents to automated systems, raising the question of who bears the loss when a chatbot's statement or decision is wrong. This study analyzes the legal status of AI chatbots, the basis and forms of civil liability of business actors, and their application to digital banking and airline services in Indonesia. It employs normative juridical research using statutory, conceptual, case, and comparative approaches to primary, secondary, and tertiary legal materials, analyzed qualitatively. The findings show that AI chatbots are not legal subjects, so liability rests on the business actor as the electronic system operator and service provider. The basis of liability can be built in layers through breach of contract, tort, Article 1367 of the Civil Code, the Consumer Protection Law, and the Personal Data Protection Law, with sector-specific features: prudential principles and financial-sector consumer protection dominate digital banking, while carrier liability and passenger compensation dominate aviation. Disclaimers that attribute errors to the chatbot cannot displace this liability. The study recommends duties of transparency, escalation to human agents, conversation logging, and a proportionate shift of the burden of proof.
Copyrights © 2026