We’ve been following this conversation here in The Editors’ Blog about what AI is, what it can do and whether it can possibly live up to its hype. It’s worth considering the question from what we might loosely call not only the supply side but also the demand side. From a very broad perspective, the history of the last three decades has been one defined by the fact that you have too much capital chasing too few productive investments. That’s the driver of the recent history of cycles of booms, bubbles, over-investment and busts: the Asian Financial Crisis, the Dot Com Bust, the Global Financial Crisis, perhaps soon the AI Bubble Collapse, along with many smaller of regional ones.
Want to keep reading?
Join TPM to get full-access to our reporting, newsletters, and other member-only features.
We’re seeing article after article now restating in slightly different ways the same basic point about the AI infrastructure boom: how will this level of investment ever be recouped? This new article in the Journal is entitled: Will America Spend 9% of Its GDP on AI? The Industry Is Counting on It. (It references the same Brookings study we discussed a week ago.) This is more or less the same story as I flagged in that Bain report in last night’s post. A market of $6 trillion for AI by 2031; 9% of GDP in 2032. The numbers may be based on slightly different global modelings — just how much investment, rates of return, etc. But they’re mostly different ways of stating the staggering growth that will be required to cover current levels of capital expenditure. The Journal notes that 9% of GDP in 2032 (one year later than the Bain & Co report focuses on) will mean Americans are spending as much on AI as food, half as much as they will spend on health care. The Journal article reminds us that this isn’t simply about adoption and new products. You have to factor in what will likely be rapidly falling prices as well as competition from new entrants with possibly far lower sunk costs and debt.
Getting into the particulars involves so many variables it’s hard make sense of it without an economic degree. But the overall story is the same as the one in the Bain study. Silicon Valley has committed the economy to unimaginable levels of investment in a very promising but still largely unproven technology. And now we have to go on a national crash course of maniacal innovation just to get out from under that spending overhang or the whole economy craters. Of course, if AI just does a million amazing things and makes all our lives happier and more fun … well, we’ll have to really thank them. But this seems uncertain.
A friend sent me this new Bain & Company study on the growth required to match the current levels of capital expenditure on AI, essentially breaking down the cost of the data center and other associated build outs. It’s an interesting document, not simply for the number-crunching and the predictions, but as a document in the more literary and analytic sense. Needless to say, Bain isn’t coming at this from any “boo capitalism!”/”this is insane lol” viewpoint. But the conclusions aren’t that far from the “this is insane lol” position. It’s very much, “hey! we can do this” but also, “um … wow, there’s quite a lot to do.”
The broad stroke numbers are these. The study says that to keep up with the current levels of capital expenditures AI will need to become a $6 trillion dollar market by 2031. (To give some perspective on that number, a recent Gartner study predicted that global IT spending will be just under $6.4 trillion this year. So devices, hardware, software, services, kind of everything.) It then posits that we can see the beginnings of, with some level of out-year predictability, something on the order of a $1.5 trillion market — which would be made up of a greatly expanded consumer market (Claude apps, ChatGPT, etc.) of between $200 and $400 billion and then an enterprise market (software, sales, marketing, business optimization) of between $1 and $1.4 trillion. So we’re left with about $4.5 trillion to go. That’s where the crash course in innovation comes in.
Want to keep reading?
Join TPM to get full-access to our reporting, newsletters, and other member-only features.
This is the weekly, free edition of The Backchannel. If you’d like to receive the free edition of The Backchannel in your inbox, you can subscribe here.
In my recent writing about LLMs and AI, I’ve mainly focused on what it can do, what it is, what its potential dangers are. I haven’t focused as much on the pretty substantial evidence that we’re in the midst of an AI bubble. The entire U.S. economy is heavily dependent on the AI boom. Much of the rest of the economy is in a slump. That boom is based on cheap money and very high expectations for AI profits, just as the Fed is facing irresistible pressure to raise interest rates, which is to say, raise the cost of money. This whole question of an AI bubble gets talked about a lot. You probably know the basic outlines. What’s caught my attention is how just in the last week or so these questions, verging into assumptions about unsustainable spending, are bleeding into the tech and the Beltway political press.
This afternoon I saw an ad being run by a SuperPAC supporting Dan Osborn (“Nebraska Values PAC”) against Sen. Pete Ricketts (R) of Nebraska. It’s a brutal ad about a brutal crime. (You can watch it here.) A woman identified as “Jody S.” speaks over a montage of a crime scene imagery explaining that in 1993 an intruder broke into her home, tied her to her bed and violently raped her while her children listened to everything. A man named John Arias was arrested, pled guilty, was convicted and sentenced to prison. We then fast-forward to then-Gov. Pete Ricketts casting the deciding vote as the member of a pardon board that ultimately pardoned Arias. Jody S. notes that Arias told her she would never be safe from him and that he was now off the state sex offender registry and allowed to own a gun. She also notes that she wasn’t even informed about the pardon until after it happened.
I tend to be hard-hearted about political ads. I watched this one a few times to make sure I had the details right. It’s hard to watch; in a way it’s more hard to listen to. The ad has stark black and white imagery and has audio of Ricketts announcing, in an uncomfortably chipper tone, his vote during the hearing: “But with that you’ve got your pardon.”
This morning the full D.C. Circuit Court of Appeals heard oral arguments on whether the Trump administration will be subject to a contempt of court inquiry for not stopping and turning around the deportation flights in the original Alien Enemies Act case, which began way back in March 2025, when the Trump administration sent more than 200 Venezuelan men to El Salvador’s CECOT prison.
Each side — the Trump DOJ and the ACLU — was given 30 minutes, but the oral arguments ended up lasting nearly three hours.
The top line: The en banc court is likely to rule against the Trump administration and allow U.S. District Judge James Boasberg to proceed with a contempt of court inquiry.
Rufo often gets treated as a “respectable” conservative. His latest podcast episode — framing Indian-Americans as a problem in need of solving — is more evidence that he shouldn’t be.
