Introduction. The article discusses the problems of the artificial intelligence development effectiveness in connection with the risks of its implementation and functioning. An economic approach to analyzing the use of AI involves comparing the positive and the negative aspects of its development. Only an economic process in which the benefits exceed the costs can be considered effective. Such a comparison is impossible without analyzing the risks of developing artificial intelligence technologies. Theoretical analysis. Currently, increasing the efficiency of the market economy requires a multiplicative effect from the digitization of all its spheres. In this regard, it is especially important to develop and constantly improve the use of AI. Thus, along with the traditional markets of resources – land, labor and capital, the digital technology market starts to occupy a priority place in the activities of modern businesses, especially in connection with the production and use of artificial intelligence. The study examines the main risks of AI development, including concentration and monopolization, inequality in development and application, loss of jobs, ethical issues, etc. Some risks are presented in the form of economic and institutional traps: dependence on selected strategies, pricing, asset dematerialization, scale, etc. The problems of AI development effectiveness in macroeconomic and sectoral aspects are investigated. Results. It is concluded that the trajectory of AI development is determined by a delicate balance between bold innovations and strict responsibility: excessive enthusiasm for rapid breakthroughs can trigger financial bubbles, economic crises and general disillusionment with technology, similar to the well-known “dotcom crisis”. According to the author, government regulation should play an important role in minimizing the risks of AI development and increasing its effectiveness.