The use of modeling to predict the development of artificial intelligence in information technology: economic aspects
TL;DR
The present study focuses on the systematization of mathematical approaches to modeling the dynamics of AI development as a technological phenomenon, including logistic and multilogistic models, methods based on the analysis of scientific publications and approaches using deep learning to predict the growth of technology.
Abstract
This article examines the theoretical and methodological aspects of applying mathematical modeling to predict the development of artificial intelligence in information technology, with an emphasis on analyzing various approaches to formalizing the dynamics of technological progress, from classical growth curves to multilogistic models and deep learning methods for predicting technology development trajectories. The relevance of the research is due to the fundamental contradiction between the need for sound forecasting of AI development trajectories for making strategic decisions in the field of information technology, innovation management and the formation of scientific and technical policy, on the one hand, and the lack of unified, widely recognized quantitative methods capable of adequately describing the historical evolution of AI and making reliable extrapolations about its future trends, on the other hand. Unlike works focused on predicting individual technical characteristics of models or on expert surveys on the timing of a technological singularity, the present study focuses on the systematization of mathematical approaches to modeling the dynamics of AI development as a technological phenomenon, including logistic and multilogistic models, methods based on the analysis of scientific publications and approaches using deep learning to predict the growth of technology.