Incitement to suicide by or with the use of artificial intelligence

Abstract

Artificial intelligence-based technologies are used in finance, insurance, and medicine, and are also accessible to individuals. However, new risks are emerging with the development of artificial intelligence (AI). Evidence shows that AI systems can, in fact, persuade a potentially suicidal person to attempt to take his or her own life. This is a new threat in the field of cybersuicidology. This article aims to examine the attribution of liability under current law, either for the autonomous action of an AI (inciting or assisting suicide) or for its use for this purpose by a human user. The cases analysed illustrate the possibility of using AI as a tool in the hands of the user, or of its autonomous actions posing a threat to potentially suicidal persons. In the first case, the user may incur liability under the Criminal Code. In the second case, where the danger arises from the autonomous peration of AI, the criminal law proves inadequate and does not take into account the rapid development of the technology and the risks arising from it, making it impossible to identify the entity responsible for the actual operation of AI. At present, in most cases, the responsibility for the autonomous operation of AI should be attributed to the manufacturer, particularly when there is a failure to provide adequate oversight of the system, or when the system is released without adequate safety filters. The AI manufacturer would then be liable under Article 415 of the Polish Civil Code.