High-Cost Debt Hits Amid AI Boom; Pressure Mounts on Data Center Companies

Amid the rapidly expanding Artificial Intelligence (AI) sector, companies are now facing the challenge of rising borrowing costs. This week, US Treasury yields hit their highest levels since 2007. This could directly impact companies that are borrowing heavily to build AI data centers and other related facilities. The 10-year US Treasury yield now stands at approximately 5.17%—about one percentage point higher than at the start of the year. Consequently, companies seeking new loans may have to offer higher interest rates to attract investors.

Debt Could Reach Trillions by 2030

According to a June estimate by JPMorgan Chase, approximately $4.1 trillion in debt could be issued for AI-related projects by 2030. Data center firms and other AI-linked companies are rapidly expanding their capacity to meet the continuously rising demand for AI services. In this environment, higher interest rates mean that constructing and operating data centers could become more expensive for these companies than before.

Less Pressure on Major Tech Companies

Major tech companies in the AI ​​sector—such as Amazon, Google, Meta, and Microsoft—plan to spend hundreds of billions of dollars on data centers and AI-related facilities this year. One advantage these companies possess is strong credit ratings, allowing them to secure debt from the market at lower costs compared to newer, smaller players. However, the situation could prove difficult for smaller companies that rely heavily on debt.

Challenges for ‘Neo-Cloud’ Companies

Raising debt could become even more challenging for ‘neo-cloud’ companies involved in AI infrastructure. According to Riley Thompson of Mitsubishi HC Capital America, lenders are now exercising greater caution when selecting projects. He noted that while the market might have previously shown interest in around 50 neo-cloud companies, only about 20 now remain particularly attractive to investors. CoreWeave has also highlighted the risk posed by rising interest rates in its filings; according to the company, every 1% increase in interest rates could raise its interest costs by approximately $30 million.

Rising Costs Impact Oracle Too

Oracle has also relied on debt markets to fund significant investments in AI infrastructure. The company’s shares saw a decline this week following a report claiming it had issued a notice regarding escalating costs associated with its data center project in New Mexico. However, the company maintained that the project is proceeding according to plan.

AI Investment Expected to Continue Despite Costlier Debt

Demand for AI is unlikely to wane despite rising interest rates. Many companies are already expanding capacity to meet future AI computing needs. Experts suggest that while higher interest rates could impact future deals, strong demand for AI means companies are unlikely to halt borrowing altogether. Consequently, the race to build AI infrastructure is expected to continue, though the associated costs for companies may be higher than before.