Bond Market Anxiety Grows Over AI Capital Spending

Growing anxiety is evident in the bond market as investors scrutinize the substantial capital expenditures by technology companies for artificial intelligence (AI) infrastructure.

Alphabet's recent upward revision of its capital expenditure forecast for AI buildouts has fueled concerns that other hyperscalers, including Google, Amazon, and Meta, may follow suit. This trend is reflected in widening credit spreads for these companies, as fixed-income investors demand higher compensation for lending.

A significant driver behind the increased capital spending is the rising cost of energy. The rapid construction of large data centers by tech giants across the U.S. entails considerable energy consumption. GE Vernova CEO Scott Strazik anticipates the current inflationary environment to persist, partially due to heightened geopolitical tensions, with oil prices already exceeding significant levels.

According to analysts at Mizuho, these capital expenditure increases are testing investor limits. Companies once perceived as "capital fortresses" are now facing a dramatic rise in AI-related costs. Analysts project that these hyperscalers collectively will spend more on capital expenditures next year than they generate in free cash flow.

The situation has triggered intense discussions between bond and equity investors. Oracle's five-year credit default swap (CDS) has reached a multi-year high, often serving as a proxy for AI debt concerns. Barclays credit analyst Andrew Keches noted that Oracle's CDS appeal extends beyond company-specific fundamentals, reflecting its role as a liquid hedge against AI capex, OpenAI's execution, and broader data center spending narratives. S&P Global recently downgraded Oracle's credit rating to BBB-, just one notch above junk status.