Is Wall Street Repackaging AI Debt the Same Way It Repackaged Subprime Mortgages in 2008?
Jim Rickards, who was once called in to help negotiate an emergency bailout, says "subprime AI" is quietly loading risk into everyday retirement accounts
Washington, D.C., July 31, 2026 (GLOBE NEWSWIRE) -- Jim Rickards says the financial engineering that blew up the housing market in 2008 never really went away. It just found a new home. In a new presentation, the former advisor to the CIA and the Pentagon argues that Wall Street is now packaging the debt behind the AI boom using the same playbook that produced the last great crisis, and that the risk is landing in ordinary Americans' retirement accounts.
The Trick That Blew Up 2008
Rickards starts with a mechanism most people never saw up close. By the mid-2000s, he explains, the market was full of toxic subprime mortgages, so Wall Street bundled bad loans together with good ones, sliced them into tranches, and sold the result as highly rated bonds paying rich yields.
The effect, he argues, was to move the danger off Wall Street's books and onto everyone else's. "Instead of keeping the toxic debt on their own books, Wall Street shifted it to yours," Rickards says. "Without your permission they put your finances on the line."
The Same Blueprint, Now for AI
Most people assume that lesson was learned. Rickards disagrees. "That's exactly what we are doing with AI," he says.
The connection, in his telling, runs through data centers. Making AI work requires enormous physical build-outs, with an estimated $5 trillion in data-center spending in the U.S. alone, and with the companies involved already burning cash, he argues that construction is being financed with massive amounts of debt. That debt, he says, is being structured as real-estate-style deals, where outside investors fund the data center and AI companies pay rent, then bundled into bonds. "Which means that AI is as much a real estate problem as it is a technological one," Rickards says.
The packaging is where the echo gets loud. "In order to sell the debt to finance their risky data center builds they're following the CDO blueprint!" he says, referring to the same instruments at the center of 2008. And the foundation, he warns, is shakier than the housing market ever was: "These loans are essentially backed by an industry that produces NO profits!"
And It's Already in Your Retirement Account
The reason Rickards thinks this matters to people who have never bought a single AI stock is where that debt ends up. Just as subprime bonds spread through the system last time, he argues, this "subprime AI" debt is now doing the same. "This debt is being bought by pension and retirement funds around the world!" he says.
The result, in his view, is quiet but widespread exposure. "These new 'data center bonds' are filling your pension funds, your retirement account, and your 401K's," Rickards says. He frames the conclusion as his analysis rather than a certainty, but his point is that the risk has already been distributed, and that most people holding it have no idea it is there.
About the Presentation
In his new presentation, Jim Rickards explains how the debt behind the AI boom is being packaged, why he compares it to the instruments that triggered the 2008 crisis, and the steps he says investors can take to check their own exposure and protect themselves.
Jim Rickards' presentation is now available online for free. Click here to watch.
About Jim Rickards
Jim Rickards spent decades advising the U.S. intelligence and defense communities before founding the research publication Strategic Intelligence. In 2007 he testified before Congress and handed the U.S. Treasury a plan to head off the coming financial crisis, a warning Washington did not act on before markets collapsed the following year.
Derek Warren Public Relations Manager Paradigm Press Group Email: dwarren@paradigmpressgroup.com
Legal Disclaimer:
EIN Presswire provides this news content "as is" without warranty of any kind. We do not accept any responsibility or liability for the accuracy, content, images, videos, licenses, completeness, legality, or reliability of the information contained in this article. If you have any complaints or copyright issues related to this article, kindly contact the author above.
