MITATIONS AND WAYS TO OVERCOME THEM IN THE PROCESS OF ARTIFICIAL INTELLIGENCE DEVELOPMENT

Abstract

Background. The growth of technical characteristics of artificial intelligence (AI)-based systems and the expansion of demand from enterprises occur with simultaneously growing development restrictions. The article is devoted to identifying growth barriers, analyzing approaches to reducing their impact, developing recommendations for implementation in organizations based on promising ways to reduce risks and fully utilize the benefits of AI. Materials and methods. The article considers data from statistical resources in the field of AI, expert assessments, and applies a forecasting method based on regression analysis. Results. The article identifies fundamental sources of limitations that complicate the scaling of known solutions, predicts the number of incidents related to AI, and lists the consequences that complicate implementation in business processes. Conclusions. The article formulates recommendations for choosing an AI model, organizational and engineering measures that reduce the risks and costs of providing AI systems.