Tech giants' AI borrowing spree rivals sovereign demand for bond market cash
Major technology companies are issuing record amounts of debt to finance AI data centers, putting them in the same investor pool as sovereign borrowers. Four large US tech firms plan roughly $650 billion in AI infrastructure spending in 2026, while hyperscaler capital expenditure estimates range from over $800 billion to $1.1 trillion. AI-related corporate debt issuance reached about $450–500 billion in the first nine months of 2026, and analysts expect it may climb further in 2027.
The borrowing push spans Amazon, Alphabet, Meta, Microsoft and Oracle. Four of the largest US technology firms alone plan roughly $650 billion of AI infrastructure spending in 2026, while hyperscaler capital expenditure estimates for the year run from over $800 billion to $1.1 trillion.
Corporate borrowing tied to AI reached about $450–500 billion in the first nine months of 2026, more than double earlier years. More than $120 billion of data-center spending has been moved off company books through separate entities and leases, and undisclosed lease and purchase obligations exceed $1.5 trillion. Analysts expect further growth, with one projection near or above $1 trillion in 2027.
Competition for bond market cash could affect more than tech firms. Pension funds, insurers and asset managers may face tougher choices about which bonds to hold, while governments might pay more to borrow if demand weakens. If AI infrastructure costs keep rising, consumers could eventually feel the effects through pricing or slower investment elsewhere. The opacity of off-balance-sheet commitments may also make it harder for savers and regulators to judge risk, though the full consequences remain uncertain.