Financial markets are showing signs of increasing stress in funding the global artificial intelligence (AI) investment boom, with borrowing risks rising for companies involved in building AI infrastructure, according to a Jefferies research report.
The report highlighted a sharp increase in credit risk indicators for companies such as Oracle and CoreWeave, raising concerns over whether financial markets will continue to support the massive investments required for AI expansion.
In its GREED & fear research report titled "K-shaped stresses, AI tremors and the Houthis," Jefferies' Christopher Wood said investors should closely monitor borrowing conditions as companies continue to spend heavily on AI infrastructure.
The report said investors need to remain "laser focused on credit spreads for any evidence that the market is no longer willing to fund the ongoing AI capex boom".
Credit spreads are indicators of the risk associated with lending to companies. A rise in these spreads generally signals that investors are demanding greater compensation for taking on credit risk.
According to the report, Oracle's five-year credit default swap (CDS) spread increased from 186 basis points on September 11 to 288 basis points, while CoreWeave's spread surged from 749 basis points to 1,118 basis points during the same period.
"These are the highest levels seen so far since the AI capex cycle got underway in early 2023," the report noted.
It also highlighted that the yield on CoreWeave's bond maturing in 2032 had reached 14.8 per cent, indicating elevated financing costs for the company.
The development is significant as expanding AI infrastructure requires substantial investments in data centres, computing capacity and related technology. Higher borrowing costs could make financing such projects more challenging, particularly for companies dependent on external funding.
However, the report does not suggest that funding for AI projects has stopped. Instead, it identifies rising credit risk as an early warning sign that investors should monitor.
Jefferies also highlighted the possibility of slowing earnings growth among AI-related companies. Citing a separate research report titled "Diversifying Beyond AI," published on October 13, it said consensus earnings growth for global AI stocks is expected to moderate from around 100 per cent in 2026 to 30 per cent in 2027 and 22 per cent in 2028.
The report noted that global AI stocks trade at 23 times forward earnings, compared with 13 times for non-AI stocks, reflecting a considerable valuation gap.
The findings suggest that although investment in AI remains substantial, rising financing risks and expectations of slower earnings growth could increase pressure on companies to justify their spending.