The largest AI infrastructure companies, Alphabet, Amazon, Meta, Microsoft and Oracle, have collectively added roughly $350 billion in debt over the past five years to finance an unprecedented expansion of AI data centers, Bloomberg reports.
This surge in borrowing reflects their belief that AI will become a major driver of future growth, with massive investments in data centers and advanced chips expected to support rising demand for AI services.
While these companies remain financially strong overall, the scale of investment has significantly increased their borrowing costs. Combined annual interest expenses have more than doubled since 2019 to over $10 billion, although they remain manageable for most firms given their strong cash generation.
However, some companies are beginning to show financial strain. Amazon recently reported negative free cash flow, Oracle’s leverage has increased and its credit rating was downgraded due to concerns about the pace of AI-related spending.
Investor sentiment has become more cautious as questions grow about whether these enormous capital expenditures will generate sufficient returns. A recent $25 billion Amazon bond offering received a weaker-than-expected reception, suggesting debt investors may be reaching the limits of their appetite for financing the AI build-out. Equity investors have also become more selective, with concerns focused on the timing and scale of future AI profits.
The five companies are expected to spend as much as $725 billion this year, primarily on AI data centers and high-performance chips. Company executives remain confident that strong customer demand and continued shortages of AI computing capacity will justify the investment.
However, analysts caution that while demand for AI infrastructure appears robust, the long-term return on these investments remains uncertain. The rapid buildup of debt also highlights the risks of relying heavily on borrowing to fund large-scale technological transitions, particularly if expected AI revenues take longer than anticipated to materialize.
