JPMorgan says artificial intelligence companies must demonstrate financial strength and sustainable cash generation to secure financing, as lenders increasingly look beyond AI-related branding when assessing creditworthiness.
“The AI label alone is no longer enough to reassure lenders. Companies must show that they can finance investment, service their debt and eventually turn spending into cash flow,” Justin Biemann, global investment strategist at J.P. Morgan Private Bank, wrote in an Oct. 23 report.
Elon Musk delivered a similar warning on Sunday, saying debt financing is drying up for AI neocloud providers, businesses that depend heavily on data center infrastructure to generate revenue.
The growing pressure could extend to cryptocurrency markets, where enthusiasm for AI has helped support broader risk-taking. Bitcoin and Nvidia both reached market bottoms within weeks of ChatGPT’s launch in late 2022 before recording substantial gains. A deterioration in AI-sector sentiment could therefore weigh on digital assets as investors reassess risk.
Markets are already showing signs of concern about whether large AI investments will translate into meaningful profits. Since Oct. 6, the Nasdaq has declined 6.7%, dropping from 31,360 to 29,263. AI-focused companies, including Nvidia and Oracle, have fallen more sharply than the index.
AI Borrowing Costs Rise as Investors Seek Protection
Credit markets are reflecting these concerns, with the cost of protecting against defaults on debt issued by several AI-related companies reaching record levels.
Investors commonly use credit default swaps (CDS) to hedge against the risk that a borrower fails to meet its financial obligations. Buyers pay an annual premium, known as the spread, in exchange for protection against a qualifying credit event. A spread of 100 basis points, for example, costs $100,000 per year to insure $10 million of debt. Higher spreads indicate that investors perceive greater credit risk.
According to Morgan Stanley, weekly trading in CDS linked to bonds issued by Oracle, SpaceX, Alphabet, Amazon, Meta, Broadcom, and Nvidia averaged more than $9 billion in September. That was a significant increase from the less than $2 billion recorded per week in July.
Companies that depend heavily on debt to finance infrastructure expansion are facing particular scrutiny.
Biemann highlighted an unnamed AI infrastructure company that has accumulated tens of billions of dollars in debt and other obligations. Its five-year CDS spread has climbed to 1,100 basis points, putting the annual cost of insuring $10 million of its debt above $1 million.
Biemann said pricing at these levels points to an implied five-year default probability of around 50% or higher, compared with below 30% a year earlier.
JPMorgan Cautions Against Treating CDS as a Countdown
Biemann referenced Lehman Brothers to illustrate how CDS markets can signal mounting financial pressure before a major failure. The bank’s CDS spreads averaged less than 100 basis points in the third quarter of 2007 before surging beyond 1,400 basis points by mid-September 2008, when Lehman filed for bankruptcy and sent shock waves through global markets.
However, he stressed that elevated CDS spreads should be interpreted as risk indicators rather than precise forecasts of an imminent default.
Trading in protection linked to major AI companies currently appears more consistent with portfolio hedging than widespread expectations of a financial crisis, Biemann said.
Large AI leaders with strong cash flows pay approximately 50 to 120 basis points for five-year credit protection. According to Biemann, those levels remain more consistent with the cost of insurance than with expectations of severe financial distress.
Some established infrastructure borrowers have seen their spreads approach 300 basis points, while credit protection across the broader investment-grade market remains relatively contained.
The main vulnerability lies with companies that require continued access to debt financing while maintaining high investment levels and generating limited cash flow. As lenders become more selective, AI businesses may need to prove that their spending can translate into sustainable revenue and sufficient cash to meet their obligations.
