Derivatives
Big Tech's $3 Trillion Off-Balance-Sheet AI Commitments Revealed
According to a Wall Street Journal analysis, nine of the largest technology companies hold around $3 trillion in off-balance-sheet commitments tied to artificial intelligence infrastructure. This figure is approximately five times the $600 billion in combined capital expenditures these firms reported over their most recent 12-month periods.
Breakdown of Commitments
The $3 trillion is divided into two main categories. The first includes unstarted leases, valued between $904 billion and $1.2 trillion, which are commitments to data-center space that companies have secured but not yet utilized. The second category consists of purchase commitments, estimated at $1.52 trillion to $1.9 trillion, covering long-term agreements for chips, data-center construction, and energy procurement.
Alphabet leads with $811 billion in purchase and contractual commitments as of June 30, 2026, while Meta reports $347 billion in leases that have yet to commence.
Accounting Implications
Under current accounting standards, leases that have not started do not appear as liabilities on a company’s balance sheet. This treatment extends to purchase obligations, which are handled differently than outright capital expenditures. Consequently, traditional metrics like debt-to-equity ratios may not fully capture the companies' financial commitments.
The analysis echoes concerns raised in a previous Nikkei report, which estimated $1.65 trillion in off-balance-sheet AI debt across five major companies. Investor Michael Burry has also noted similar risks, suggesting that the real financial landscape for these firms may be more complex than it appears.
Impact on Investors
The stark contrast between reported capital expenditures and off-balance-sheet commitments indicates that the actual investment in AI infrastructure is significantly larger than what is currently disclosed. This discrepancy may lead to a reevaluation of company valuations, as analysts who do not account for these future obligations could overestimate available cash for dividends and other shareholder activities.
FAQ
What are off-balance-sheet commitments in the context of AI investments?
Off-balance-sheet commitments refer to financial obligations that are not recorded on a company's balance sheet. In the context of AI investments, these include unstarted leases for data-center space and long-term purchase commitments for chips and construction.
How much do the largest tech companies have in off-balance-sheet AI commitments?
According to a Wall Street Journal analysis, nine of the largest technology companies hold around $3 trillion in off-balance-sheet commitments related to AI infrastructure.
What are the two main categories of these commitments?
The commitments are divided into two main categories: unstarted leases, valued between $904 billion and $1.2 trillion, and purchase commitments, estimated at $1.52 trillion to $1.9 trillion.
Why do off-balance-sheet commitments pose risks for investors?
Off-balance-sheet commitments can obscure the true financial obligations of a company, leading to potential miscalculations in company valuations. Analysts may overestimate available cash for dividends and other shareholder activities if they do not account for these future obligations.
How do current accounting standards affect the reporting of these commitments?
Under current accounting standards, unstarted leases and purchase obligations do not appear as liabilities on a company's balance sheet. This treatment can result in traditional financial metrics, like debt-to-equity ratios, not fully capturing the companies' financial commitments.