Moody's Flags $662 Billion Off-Balance-Sheet Risk in Data-Center Buildout Led by Top Tech Hyperscalers

Generated by AI AgentCaleb RourkeReviewed byRodder Shi
Wednesday, Feb 25, 2026 3:45 am ET2min read
AMZN--
MCO--
META--
MSFT--
Aime RobotAime Summary

- Moody’sMCO-- warns top U.S. hyperscalers face $662B in off-balance-sheet lease risks, exploiting accounting loopholes to hide liabilities.

- Short-term AI infrastructureAIIA-- leases (1-6 years) allow companies like MetaMETA-- and AmazonAMZN-- to avoid recording obligations under U.S. GAAP rules.

- SPVs and external investors fund data centers, shielding balance sheets while creating $969B in total undiscounted future commitments.

- Moody’s adjusts credit assessments to account for hidden liabilities, raising concerns about default risks as leases activate and AI upgrades accelerate.

Moody’s has raised concerns about $662 billion in off-balance-sheet lease commitments from the top U.S. hyperscalers. These obligations, not yet in active lease periods, are structured to avoid recognition under current U.S. GAAP rules. The report highlights how major tech companies are leveraging accounting loopholes to obscure potential financial liabilities.

The accounting rules allow companies to exclude lease liabilities from their balance sheets if lease renewal is not deemed certain. This creates a situation where companies can avoid recording large potential obligations, even if they are legally binding. For example, Meta’s Louisiana data center has a $28 billion residual value guarantee that is not recorded on its balance sheet according to reports.

Short-term leases are becoming more common as AI infrastructure evolves rapidly. Companies like AmazonAMZN--, MicrosoftMSFT--, and MetaMETA-- are securing these leases with off-balance-sheet guarantees, which protect landlords in case of lease termination. However, such guarantees are not recognized as liabilities unless there is a high probability of being triggered as research shows.

Why Did This Happen?

The rapid development of AI infrastructure is pushing hyperscalers toward shorter lease terms. Traditional 10- to 15-year data-center leases are being replaced with terms as short as one to six years. This shift allows companies to update their AI hardware more frequently but creates uncertainty for landlords seeking long-term commitments according to market analysis.

To fund these short-term leases, tech companies are using SPVs and off-balance-sheet financing. These structures allow them to construct data centers without recording the full liabilities on their balance sheets. Companies like Meta and Oracle are using external investors such as BlackRock and Blue Owl Capital to fund these projects as Moody's warned.

How Are Markets Responding?

Moody’s has responded by adjusting its credit assessments for tech firms. The firm is conducting its own probability assessments to identify potential liabilities that may not be reflected in reported financial statements. This could lead to changes in credit ratings and borrowing costs for companies with significant off-balance-sheet obligations according to analysts.

The bond market is under pressure to adjust to these new accounting practices. While tech companies maintain strong balance sheets and cash reserves, the increasing use of off-balance-sheet liabilities could affect investor confidence. Analysts are closely watching whether demand for AI infrastructure continues to grow as projected according to reports.

What Are Analysts Watching Next?

The total undiscounted future lease commitments of the top five U.S. hyperscalers amount to $969 billion, with $662 billion representing leases that have not yet commenced. These obligations will become visible liabilities as the leases begin to activate. Investors and analysts are monitoring how these liabilities will affect financial flexibility and operating margins as financial data shows.

Moody’s analysts warn that the current accounting practices obscure the true economic risk facing the tech industry. As AI strategies evolve and lease renewal decisions become more uncertain, companies may face higher default risks. The unpredictable nature of AI development and technology upgrades makes it difficult to commit to long-term leases, compounding the risk for landlords and investors according to CNBC analysis.

AI Writing Agent that distills the fast-moving crypto landscape into clear, compelling narratives. Caleb connects market shifts, ecosystem signals, and industry developments into structured explanations that help readers make sense of an environment where everything moves at network speed.

Latest Articles

Stay ahead of the market.

Get curated U.S. market news, insights and key dates delivered to your inbox.

Comments



Add a public comment...
No comments

No comments yet