Problems and Prospects of Legal Regulation of Public Relations connected with the Use of Neural Networks

Abstract

The level of digitalization has increased significantly in the current century, the speed of the Internet has increased by many times, and it is now possible to access it from different parts of the world. Today, artificial intelligence occupies a special place in the digital technology market, which is already an indispensable tool in many sectors of the economy of developed countries. The purpose of the study is to identify the main urgent problems of using neural networks, as well as to form proposals for their legal regulation. The study uses the formal logical method, the comparative legal method, analysis and synthesis, methods of induction, deduction, and abstraction. It has been established that artificial intelligence cannot yet distinguish a joke from a real command or user request, respectively, further development of these technologies is impossible without the implementation of an analog function of cognitive thinking. It is concluded that self-regulation can be the best way to regulate the use of neural networks, since the legal system of continental law, to which the Russian Federation belongs, is quite rigid and often may not have time to regulate the rapidly developing field of artificial intelligence. Self-regulation is able to provide an opportunity to convey proposals on the legalization of effective rules for organizing the activities of IT market participants, to create an effective mechanism for guaranteeing the quality and safety of artificial intelligence based on the joint property liability of members of self-regulating organizations. At the same time, it requires the adoption of legal norms on liability for the illegal use of neural networks, as was done in the United States and China. In the near future, deepfakes created on the basis of neural network technologies may become a threat to national security and cause harm to thousands of citizens.