Jason Thomas, head of global research and investment strategy at Carlyle Group, said that there are worrying similarities between how lenders finance data center projects and some of the practices adopted in the mortgage market before the global financial crisis. In an interview with the media on Tuesday, he said that specifically, project financing for artificial intelligence development is often supported by the investment-grade credit qualifications of financial sponsors. He added that lenders are not necessarily betting on the viability and cash flow prospects of the project itself. “But I'm really concerned because this is in some ways reminiscent of the situation from 2005 to 2007,” Thomas said. At the time, lenders believed that they were not evaluating the mortgage loan collateral itself, but rather the bank's ability to include these assets on the balance sheet if necessary. As debt continues to accumulate, this practice eventually becomes unsustainable.

Jason Thomas, head of global research and investment strategy at Carlyle Group, said that there are worrying similarities between how lenders finance data center projects and some of the practices adopted in the mortgage market before the global financial crisis. In an interview with the media on Tuesday, he said that specifically, project financing for artificial intelligence development is often supported by the investment-grade credit qualifications of financial sponsors. He added that lenders are not necessarily betting on the viability and cash flow prospects of the project itself. “But I'm really concerned because this is in some ways reminiscent of the situation from 2005 to 2007,” Thomas said. At the time, lenders believed that they were not evaluating the mortgage loan collateral itself, but rather the bank's ability to include these assets on the balance sheet if necessary. As debt continues to accumulate, this practice eventually becomes unsustainable.