There’s a particular kind of dread that comes before something breaks. You know it’s coming, you can almost feel it, yet you pretend not to hear it creak.

You can see that firsthand right on X (Twitter) or LinkedIn. Off the back of the latest eyebrow-raising AI investments from OpenAI, AMD, Nvidia, et al, users have been sharing their own follow-the-money flywheels or meme’d stills from the film The Big Short.

And they’re right to do so: The sheer levels of investments in AI and related infrastructure have a poignant precedent in the dot-com bubble and its subsequent crash. The stock market of the late 90s and early 2000s ballooned on the back of billions of dollars in speculative investments in technology that, at the time, there was just no use for.

Sound familiar?

It’s little wonder then, if you cast your mind back to just the start of this year, to a time where one analyst report caused unprecedented levels of unease. TD Cowen’s March report on Microsoft’s purported data center cancellations could be viewed as the first pinprick that caused the initial few ripples into the AI data center bubble.

Add that to the DeepSeek debacle wiping millions of dollars from market caps (which have likely already found their way back by this point), and the precipice the bubble precariously finds itself on doesn’t seem so safe.

So, when a Fortune report featured Harvard economist Jason Furman warning that without data centers, U.S. GDP growth for the first half of 2025 was just 0.1%, it’s increasingly difficult not to feel like the increasingly exacerbated character played by Steve Carell in The Big Short.

At times, it’s even hard to keep track of it all. First, Microsoft invests in OpenAI, which buys chips from Nvidia. Who then invests in OpenAI, but not before selling chips to Oracle. OpenAI then snaps up cloud space from Oracle, which, in turn, invests in OpenAI.

Are you following along? Because now there's AMD, who OpenAI now buys chips from in exchange for 10% of the company. Then along comes CoreWeave, and OpenAI invests billions of dollars into them. And there’s an apparent deal in the works between OpenAI and Google Cloud, and then Pepe Silvia, Pepe Silvia, Pepe Silvia.

After reading and trying to make sense of all of that, and that's just a fraction of the financial shenanigans going on, surely it’s a case not of if, but when. And we might not have too long to wait, either.

Take your mind back slightly further from March's TD Cowen report, to January, when companies were laying off staff left and right. Some of the biggest firms to have reduced their head counts just so happen to be among the biggest data center spenders.

Meta laid off 3,600 employees in February, despite raising both its capex for the year to between $66 billion and $72 billion, and executive bonuses to 200% of base salary. In late 2024, AMD cut 1,000 workers. Google laid off 12,000 of its global workforce in 2023.

Days before Oracle struck its latest mega-deal with OpenAI, the company kicked off yet another round of layoffs, affecting Oracle Cloud Infrastructure (OCI) teams as well as its Health, architects, and other divisions.

Dell, Amazon, and Samsung Electronics have all laid off staff, too. And to top it all off, the embattled Intel has gone from around 100,000 employees in 2022 to approximately 75,000 at the time of writing.

To top it all off, just a few weeks after the infamous TD Cowen report, Microsoft laid off around 6,000 workers – all while planning to spend some $80 billion on data centers this year.

We're a few months away from yet another January of misfortune, but the doom and gloom from the early months of 2026 may just have come sooner than expected. Earlier this week, the hyperscaler reportedly canceled plans for a data center in Caledonia, Wisconsin, only weeks after officially announcing the project.

The tech giant put it down to "community feedback," but you have to wonder...

Wake me up when January ends

It’s said the first month of the year often is the most miserable, what with it being a five-week month, meaning a long time to wait for paychecks, all while those credit card payments for holiday gifts start to rear their ugly head.

With so many jobs already gone of late across the tech sector, it’s hard not to let that misery wash over you. And with the fear of the AI bubble bursting, more cuts are surely on the way.

Of course, we don’t know what’s around the corner. Could more tariffs come to the fore to throw yet another spanner into the works? Remember, DeepSeek came out of nowhere, and look at the damage that it did to some of the firms mentioned above.

Now, I previously examined the concerns around DeepSeek in a piece for a prior publication. But regardless of whether their architectural approach was genuinely achieved on outdated hardware or not, you cannot ignore the impact that one unexpected blip had on the bubble.

Don’t believe me? Or, rather, don’t want to believe me? Jeff Bezos said there's a bubble. JPMorgan Chase CEO Jamie Dimon has warned about it. Even Sam Altman – one of the architects of this cacophony of confusion – admitted we're in one.

When January rolls around, and the axe begins to swing again (and it will), questions have to be asked about how companies are spending such vast sums on technology that, for the most part, isn't ready. And yet, when it comes time to balance the books, it'll be the human talent who will pay the price.