$1.65 trillion in "shadow" debt: how Big Tech giants mask the true scale of the AI race
The five largest technology corporations—Alphabet, Microsoft, Amazon, Meta, and Oracle—have accumulated a colossal $1.65 trillion in hidden debt obligations that are not reflected on their official balance sheets. This phenomenon, which I call the "AI financial mirage," is becoming the main threat to investors unaware of the real level of risk.
This is not about classic loans, but about long-term contracts for leasing GPUs, data centers, and participation in joint ventures. Current accounting standards allow such obligations not to be recorded as debt until the facilities are put into operation. As a result, the official debt of $1.35 trillion appears quite manageable, but the real figure of $1.65 trillion remains off the books.
Who is "hiding" how much
The distribution of hidden obligations is extremely uneven. Meta (recognized as an extremist organization in Russia) has "off-balanced" $420 billion—three times its official debt. Oracle's obligations have increased 30-fold in four years. The Bank for International Settlements (BIS) has already dubbed this practice "shadow borrowing."
The problem is compounded by the fact that four of the five companies are reporting financial results within the next two weeks. The official debt will appear moderate, while the $1.65 trillion from the footnotes will not make headlines. As analysts rightly note, the worst-case scenario materializes with weak demand for AI services: facilities will have to be written down, and losses will fall on investors and insurers who financed the construction through private credit and project bonds, unaware of the full extent of the risks.
Cash flow is melting before our eyes
Bank of America has already warned of a cash crisis. The free cash flow of the largest data center operators will fall from $191 billion in 2025 to $19 billion in 2026, and will turn negative by approximately $26 billion in 2027. The reason is record capital expenditures: the five corporations' AI spending has already exceeded $725 billion in 2026, and the sector's debt issuance will reach about $175 billion—six times the average over the past five years.
The giants' strategy is simple: betting on a future monopoly. They aim not just to recoup infrastructure costs, but to become an essential link for everyone using AI, and to profit from every query. Alphabet looks like the favorite: Google Cloud revenue grew 63% year-over-year, exceeding $20 billion per quarter, the order backlog nearly doubled to $462 billion, and the company began selling its proprietary TPU chips to external clients.
My conclusion as an analyst: Big Tech is building AI infrastructure on debt, hiding half of its obligations in footnotes, and sacrificing cash flow for future control over the industry. If AI demand meets expectations, the winner will be the one who owns its infrastructure. If not, the "hidden" trillions will hit investors who were unaware of the full size of the bet. In such conditions, the cryptocurrency market could become a beneficiary of capital outflows from risky corporate securities.