Liability for Breach of Contract by Autonomous AI in Cross-Border Contracts

The development of Artificial Intelligence (AI) has generated new challenges in civil law, particularly regarding breach of contract in cross-border transactions. The autonomous decision-making capacity of AI systems raises questions concerning their legal status and the allocation of contractual liability. This study aims to analyze liability for breaches of contract caused by AI under Indonesian civil law and to examine the determination of the applicable law (lex causae) in cross-border contracts from the perspective of private international law. Using a normative juridical method with conceptual, case-based, and comparative approaches, this research draws upon the Indonesian Civil Code, the Electronic Information and Transactions Law, the Copyright Law, and relevant legal literature. The findings show that autonomous AI cannot be recognized as a legal subject in the Indonesian legal system due to the absence of free will and moral consciousness, so liability remains attributed to the human or legal entity operating the system under the doctrine of vicarious liability. The case of Quoine v B2C2 confirms that contracts executed through algorithms remain legally binding on human or corporate parties rather than AI systems. From a private international law perspective, the study affirms that the use of AI does not alter the fundamental principles for determining lex causae, which continue to rely on connecting factors such as lex loci contractus, lex loci solutionis, and party autonomy.