Fraud by generative AI chatbots: On the thin line between deception and negligence

Abstract

The use of generative AI systems is on the rise. As a result, we are increasingly often conversing with AI chatbots rather than with fellow humans. This increasing use of AI systems leads to legal challenges as well, particularly when the chatbot provides incorrect information. In this article, we study whether someone who decides to contract on the basis of incorrect information provided by a generative AI chatbot might invoke the fraud regime to annul the resulting contract in various legal systems. During this analysis, it becomes clear that some of the requirements that are currently being put forward from a public law perspective, such as in the European AI Act, may also naturally arise from existing private law figures. In the same vein, this analysis highlights the interesting intradisciplinary feedback between instruments of public law and other legal domains.