Financial Resilience of TNCs in the Context of Growing Investments in Artificial Intelligence

Анотація

The article presents a comprehensive study of the transformation of financial architecture and an assessment of the financial resilience of leading TNCs (Alphabet, Microsoft, and Meta) amid growing investments in artificial intelligence. Based on a retrospective analysis of financial statements from 2020 to 2025 and forecast modeling for 2026, a critical increase in capital expenditures and depreciation charges was identified, driven by the accelerated obsolescence of high-tech assets. A fundamental shift in the investment paradigm was observed: the transition from a «low capital expenditure – rapid growth» model to a capital-intensive model with high energy consumption and depreciation burden. A dangerous imbalance was revealed between the exponential growth of infrastructure costs and the deflationary pressure on end-service prices due to fierce competition. The authors substantiate the conception of an «amortization trap», according to which the pace of technological obsolescence of hardware outstrips capital payback cycles. The calculated ROIAI indicator, currently ranging between 0.33–0.49, demonstrates the inability of the current monetization of AI solutions to fully offset the costs of deploying corresponding capacities. The authors note that in the absence of revenue growth from the AI segment in the next 12–18 months, the market can expect a systemic revaluation of assets. This will lead to a fundamental correction of market multiples for Big Tech companies to the levels of traditional industrial corporations. The year 2026 will likely become a critical point for a change in the prevailing paradigm of investment in artificial intelligence. The amortization trap emerges as one of the most important financial problems of the current decade, and its resolution will play a decisive role in shaping the dynamics of leadership within the global digital economy in the coming years.

Класифікація

Ідентифікатори

Схожі роботи