AI Infrastructure Investment Outpaces Historic Cycles

The rise of artificial intelligence is driving a cycle of physical capital investment that is unprecedented in history. Projections show that spending on data processing infrastructure in the United States will reach 3.6% of GDP between 2025 and 2032. This volume surpasses the 2.2% average recorded during the railroad expansion of the late 19th century.

Tech giants are leading this massive financial movement. Companies such as Alphabet, Amazon, Meta, Microsoft, and Oracle increased their capital expenditures from 97 billion dollars in 2020 to 416 billion in 2025. Estimates suggest this figure could reach 801 billion dollars by 2026.

The ratio between investments and operating cash flow shows a drastic shift. In 2020, the index was 38%, but it is expected to hit 113% in 2026. This means planned spending will exceed the operating cash generation of these organizations by 93 billion dollars next year.

This scenario creates significant macroeconomic risks due to the high concentration of capital. The sustainability of this movement depends entirely on the future profitability of new AI businesses. If return expectations are not met, the sector could face financial crises with direct impacts on lenders and investors.