Evolutionary trends in the development of artificial intelligence in corporate resource management and strategic decision-making

Abstract

The study analyzes the evolution of artificial intelligence application in corporate resource management and strategic decision-making, clarifying its transition from local optimization to integration into the core of strategic management. The aim is to quantitatively and qualitatively assess the impact of Al on financial and non-financial performance indicators of companies and to identify factors of successful implementation. The methodology combines comparative financial analysis and KPI calculation (ROI, NPV, cost and revenue dynamics), correlation-regression models, and a case study of 5 leading companies based on data from 50 Fortune 500 multinational corporations for 2018-2023, with triangulation of sources (unconsolidated statements, annual reports, public cases, and interviews). The results confirm the systemic effect of Al: an average reduction in total operating costs by 16,21%, with the most significant decreases in logistics and supply chains (-19,28%), administrative functions including HR (-17,65%), and quality control (-12,91%). There is high differentiation in return on investment across functions: ROI in marketing and sales - 378,09%, in financial management and risk analysis-315,98%, in production-236,76%, in R&D-176,20%. A strong positive correlation is shown between the depth of Al integration and strategic KPIs: an increase in the Customer Satisfaction Index (CSI) from 72.4 to 94.6 points, a reduction in time-to-market (TTM) from 18,3 to 8,2 months, and a steady growth of the Employee Productivity Index (EPI). The comparison of forecasted and actual effects in new revenue streams shows that the actual performance exceeded forecasts by +6,78% overall, with +20,99% in personalized services, +8,59% in niche market development, and -5,66% in dynamic pricing, highlighting behavioral limitations of purely algorithmic strategies. The discussion indicates a transition to data-driven management and prescriptive analytics, the need for organizational transformation (culture, skills, solution architecture), and a hybrid human-machine model. The limitations are related to industry heterogeneity, dependence on public data, and the need for longitudinal measurement. Prospects lie in developing Al management maturity metrics, risk management models, and ethical frameworks for application.