• Bond markets are growing impatient with AI firms as massive capital expenditures on data centers and infrastructure outpace cash flows, leading to a surge in debt issuance.
  • Credit default swaps for companies like Oracle (ORCL) have spiked sharply, signaling heightened bondholder concerns over default risks tied to heavy AI borrowing without clear returns.
  • Investors are demanding proof of monetization, with forecasts predicting over $300 billion in annual AI-related high-grade bond issuance by 2026, testing market capacity and discipline.

A Shift in Market Sentiment

Bond investors are increasingly skeptical of AI companies' massive capital expenditures, demanding proof of monetization amid rising debt issuance, as warned by Michel Saugne, CIO at La Financière de l'Echiquier. This reflects a shift where bond markets, unlike buoyant equity markets, price higher default risks for firms due to heavy AI borrowing without proportional returns. In recent weeks, credit default swaps for Oracle have spiked sharply since summer 2025, signaling bondholder concerns over its OpenAI exposure and cash burn, while Meta (META), Amazon (AMZN), and Microsoft (MSFT) face milder rises, according to market data.

Big Tech's AI capex forecasts have surged to over $405 billion for 2026, up from initial $250 billion estimates, driven by data center buildouts. Companies issued $75 billion in debt recently for AI projects—double the decade average—pushing capex to 94% of operating cash flows. Efforts to fund this gap have tapped bonds, loans, and private credit, with Bank of America noting an unprecedented debt surge. Hyperscalers like Alphabet (GOOGL), Amazon, Meta, and Oracle borrowed $93 billion in 2025 US investment-grade bonds, accounting for 6% of total issuance; Oracle is deemed riskiest due to high debt-to-earnings ratios and a near-junk rating.

Investor Discipline and Future Implications

Investor discipline is rising, prioritizing revenue proof over hype, as Saugne likens the situation to selling picks and shovels in a gold rush—you need proof there's gold to be found. Without a deal to demonstrate returns, companies could face higher borrowing costs or forced restructuring. JPMorgan forecasts $300 billion in annual AI-related high-grade bonds and $1.5 trillion over five years; Barclays flags this as a 2026 credit supply driver. In a brief statement, Saugne emphasized that bond markets will expect companies to prove they can monetize AI spending, or risk skepticism as they raise debt to fund growth.

Short-term, watch for $300 billion-plus bond issuance in 2026 testing market capacity, with Oracle's debt volume and CDS movements as key indicators. Long-term, this discipline could rein in hype for sustainable AI growth via better risk pricing; revenue shortfalls might spike yields, curbing capex. Related developments include Elon Musk's xAI issuing $5 billion bonds and loans at 12.5%, and CoreWeave raising $3.75 billion high-yield at 9% post-IPO, highlighting the broader trend. Attempts to reach Oracle for comment were unsuccessful, but sources familiar with the matter say internal discussions are ongoing to address bondholder concerns.

*Correction: An earlier version misstated the percentage of operating cash flows; it is 94%, not 95%.