Methodological Problems of Information Development–Analytical Infrastructure for Assessing the State and Forecasting the Sphere of Artificial Intelligence

— The article examines the directions for improving the information and analytical infrastructure in the field of artificial intelligence (AI) and the development of its individual elements. It is proposed to form a comprehensive data collection system capable of providing government bodies, business and society with high-quality information about the current and forecast conditions of the object. The need to institutionalize the concept of “artificial intelligence” for the purposes of government monitoring is proven. An analysis of the main parameters of the state of the artificial intelligence market, the most relevant from the point of view of modern analysts, is given, on the basis of which it is concluded that the global AI market has become one of the important factors in GDP growth. Analysis of the AI segment in Russia shows that in the coming years it can reach statistically significant volumes, and therefore it is necessary to actively include and expand data on AI in the national information and analytical infrastructure, in particular in the state statistical observation system. Recommendations are given regarding the methodological elaboration of the features and specifics of the development of the information infrastructure of AI. The most significant challenges facing this area are discussed: formalization of the definition of AI, development of a unified measurement and monitoring infrastructure, problems of reflection in statistical accounting, adaptation of existing statistical observations in order to obtain up-to-date data on its current and forecast state. It is proven that the measurement infrastructure and monitoring system for AI should not only reflect its contribution to achieving strategic goals, but also be specified in accordance with the current institutional framework for implementing the innovation economy model as a whole.