The financing frenzy of AI data centers has raised concerns in the market, with Bank of America warning that tech giants borrowing to expand could become a potential source of credit risk.

date
06:00 20/05/2026
avatar
GMT Eight
With the continuing trend of AI data center construction heating up, Wall Street is increasingly worried that large-scale borrowing for expansion may be laying the groundwork for the next credit market shock.
With the AI data center construction boom continuing to heat up, Wall Street is becoming increasingly concerned that large-scale borrowing and expansion may be laying the groundwork for the next credit market shock. The latest survey by Bank of America shows that the capital expenditure of AI mega-scale cloud providers has rapidly become one of the top potential credit risks for global investors. Survey data shows that among global fund managers surveyed in May, about 34% of respondents believe that AI-related capital expenditures are most likely to trigger a future systemic credit event, a doubling from April's 17%. Although the U.S. private credit market remains the largest source of concern, at 42%, it is significantly lower than last month's 57%. Since the beginning of last year, technology companies have raised over $300 billion through the U.S. bond market for AI infrastructure construction, and investment banks expect the scale of financing to continue to increase significantly in the coming months. The market's concern is that technology companies are borrowing at an unprecedented rate to invest in AI construction, but there is still significant uncertainty about whether these investments will produce sufficient returns in the future. David De Boltz, Managing Director of Leveraged Finance at JPMorgan, said, "The growth in financing scale has expanded exponentially." He pointed out that almost all funds in the market are flowing into the AI field, and investors are constantly evaluating how much capital needs to be set aside for these transactions. The Bank of America survey was conducted between May 8 and 14, with more than 150 global fund managers participating. In addition to AI capital expenditures, 6% of respondents believe that U.S. consumer credit is the most likely area to trigger credit risk; 4% are concerned about the Japanese government bond market; and another 2% believe that cryptocurrencies and stablecoins may become potential sources of risk. In terms of broader market tail risks, 40% of respondents listed "inflation re-igniting" as the biggest threat; 20% are worried about geopolitical conflicts; 18% are concerned about disorderly spikes in bond yields; 11% believe the AI bubble may burst; and 6% are focused on private credit risks. However, despite the increasing vigilance of the market towards AI financing risks, De Boltz said that current lending institutions are still very interested in AI-related projects. He pointed out that capital is now more inclined to flow towards companies that directly benefit from AI development, rather than software companies that may be disrupted by AI. De Boltz bluntly stated, "All funds are now flowing into AI."