The CURRENT INVESTMENT CYCLE IN ARTIFICIAL INTELLIGENCE IN THE UNITED STATES: A TECHNOLOGY BUBBLE?

Abstract

The recent expansion of artificial intelligence (AI) in the United States has been accompanied by an investment cycle of considerable magnitude, driven by large technology corporations with a high capacity to mobilize capital. This process relies on digital infrastructures developed during the mass adoption of the internet in the 1990s, as well as on the financial strength of companies established as key players in both stock markets and corporate debt. The growing centrality of these firms raises questions about the nature of the technological cycle and its implications for financial stability, especially under the hypothesis of a technological bubble. This study adopts a historical-financial comparative methodology, based on the examination of valuation indicators, market data, and institutional reports, with attention to the recent period (2022–2025) and its contrast with the dot-com bubble (1995–2000). Three main dimensions are examined: stock market concentration around the technology sector linked to AI, the accelerated expansion of investment in digital infrastructure, and the dynamics of financial interconnection among large companies in the sector. These dynamics are compared with historical episodes of overvaluation, particularly the dot-com bubble of the late 1990s, with the aim of identifying, in the current cycle of AI expansion, traits compatible with such processes. The results suggest that, although there are solid industrial foundations in the development of AI, the magnitude of investment and market concentration raise questions about the sustainability of the current cycle.