Artificial Intelligence: Modern Approaches to Multilateral Regulation

The subject of the research is the causes, prerequisites and the state of regulation of artificial intelligence (AI) in multilateral formats. The aim of the work is to identify approaches to AI regulation at the multilateral level and analyze the current status of the regulatory process. The relevance is due to the rapid development of AI technologies and the various risks associated with this process. The research methodology includes the use of general scientific methods, namely statistical and expert assessments, analysis of empirical data, classification, structuring and systematization of objects. As a result of the research, the level of the global AI technology market has been determined, and conceptual approaches to multilateral regulation in various structures have been identified. The main directions of AI regulation within the framework of international organizations (UN, OECD, Council of Europe, and WTO), the EU and non-integration type associations, as well as informal discussion clubs (Hiroshima Process, Global Partnership on AI) are considered. It is concluded that despite the mass of solutions and documents being developed in multilateral formats, the «rules of the game» for AI developers and distributors lag behind the development of technologies in this area. It has been established that decisions taken at the multilateral level are not binding on subjects of the global economy, with the exception of regulatory acts of integration groupings. There has been an increase in the regulatory activity of international structures in recent years due to an increase in the level of commercialization of AI technologies, as well as innovative products produced using AI or containing it as an integral part. The role of multilateral regulation of AI in shaping global economic development trends is outlined.