When the man who helped finance the borrower tells you the borrower is fine, you are not hearing analysis. You are hearing an investor defend his position.
Jon Gray, Blackstone's president, says CoreWeave can handle the financing pressures of building AI infrastructure. Blackstone has been a lender to CoreWeave and a giant of the data-center business. Gray's case rests on strong demand, tight supply, rising compute prices, and older chips holding value better than expected. That last point is the one to watch. A great deal of this debt is secured by GPUs, and the assumption that those chips will keep their worth is what makes the borrowing look prudent. If it's wrong, the collateral is a warehouse of depreciating silicon. Gray also conceded that higher energy costs and Treasury yields are squeezing the sector. That is a polite way of saying the model works only while money stays cheap and power stays plentiful.
Now look at the rest of the story. Jensen Huang complains of a "scarcity of capital." His company sells the chips, and it has put money into customers who buy them, so the scarcity he laments is partly of his own design. S&P says the big cloud providers carry more AI-related debt than anyone had recognized. Put those together and you have an industry that insists it is both indispensable and starved for cash, borrowing at a pace that surprises the rating agencies.
Then there is the paragraph about France. Investors are fretting over its debt and deficits, and Moody's is weighing whether to cut a sovereign rating. A government that funds hospitals, schools and pensions gets a public examination. Its ministers are told to find savings, and bond traders watch like hawks. Meanwhile a private borrowing boom of enormous scale is described as "navigating financing pressures," and the people running it get a vote of confidence from their own financiers.
The double standard is the story. Public debt that buys things citizens can use is a moral failing to be corrected. Private debt that buys speculative capacity, with returns hoped for years out, is treated as visionary. When the public version goes wrong, voters get austerity. When the private version goes wrong, we have a pattern: the losses spread through banks, private-credit funds and pensions, and eventually someone says the system is too important to let fail.
The costs also land on people who never signed the loan documents. Gray names energy costs as a challenge. The question is who pays them. In regions where data centers cluster, utilities build generation and transmission for a handful of enormous customers, and ordinary ratepayers are asked to share the bill. Higher Treasury yields, which Gray also flags, are partly the market pricing in all this appetite for capital. That raises borrowing costs for governments, homebuyers and small businesses. The AI build-out competes with them for credit, and so far it is winning.
None of this means AI is a fraud or that CoreWeave will collapse. The technology is real, and some of this investment will look wise. But the Left's job is to ask what happens if the optimists are wrong, and who is stuck holding the bag. Wall Street's answer is always the same: trust us, the fundamentals are strong. We heard that about mortgage securities and about every boom that ended with public rescue.
If AI infrastructure is going to be a public necessity, treat it as one. That means transparency on leverage, honest disclosure of how chip collateral is valued, and rules ensuring that data-center power demand is paid for by the companies creating it. It also means no quiet assumption that taxpayers are the backstop. We cannot lecture France about its deficits while waving through a private debt binge because a Blackstone executive says it's fine.
How it may affect me
If you live near a large data-center cluster, ask your state utility commission how new power and transmission costs are allocated; ratepayers can end up sharing them. Also check whether your pension or retirement plan holds private-credit funds, since AI infrastructure debt is increasingly financed there. Rising Treasury yields can also keep mortgage and loan rates higher for longer.
