$1 Trillion AI Capex Erodes Hyperscaler Credit Quality, Moody's Says

Moody's Ratings warned this week that the breakneck pace of AI investment is straining the balance sheets of the world's largest technology companies. The ratings agency tracked six hyperscalers—Microsoft, Amazon, Alphabet, Meta, Oracle, and CoreWeave—finding that the shift from software to physical infrastructure is pressuring credit quality.

Capital expenditures for physical assets like data centers are projected to reach $785 billion in 2026 before hitting approximately $1 trillion the following year. To fund this buildout, direct debt across the six companies has climbed to roughly $460 billion. Alphabet recently announced an $85 billion equity sale to help finance its ambitions.

Much of the financial burden is being kept off formal balance sheets through long-term data center leases. Moody's noted these lease commitments have ballooned to $1.2 trillion, with more than $820 billion tied to facilities still under construction. The agency treats these obligations as debt-equivalent liabilities that will require heavy rent payments for years to come.

This dynamic breaks a decades-old Silicon Valley playbook. "Previously, these companies relied on asset-light structures centered on software, intellectual property, and scalable cloud services that required modest capital investment," Moody's said. "The transition from asset-light to asset-heavy models requires unprecedented levels of investment and capital raising."

While Alphabet, Amazon, Meta and Microsoft possess some of the strongest corporate balance sheets globally, the immediate risks are concentrated lower down the credit spectrum. Oracle holds a Baa2 rating with a negative outlook, just two notches above junk. CoreWeave sits deep in high-yield territory at Ba3, relying on complex private debt to finance its server fleets.

Moody's also highlighted a structural flaw in the AI boom's revenue assumptions. Hyperscalers have invested billions into pre-IPO labs like OpenAI and Anthropic, which in turn spend heavily on cloud computing from those same tech giants. This circular ecosystem means the industry is heavily dependent on a narrow base of customers sharing identical demand projections.

Robust demand and hundreds of billions in long-term contracts currently shield the sector from downgrade risks. However, the massive upfront costs of AI hardware contrast with revenue that materializes over a longer timeline. "Investors will increasingly focus on these companies' ability to realize an adequate return on investment," Moody's concluded.