Wall Street sees AI bubble coming and is betting on what pops it

https://www.bloomberg.com/news/articles/2025-12-14/wall-street-sees-an-ai-bubble-forming-and-is-gaming-what-pops-it

by simonpure • 7 months ago

51 76 7 months ago

https://archive.ph/2G0uD#selection-1219.0-1219.64

kryllic• 7 months ago

If there wasn't an ongoing de facto recession, I would wager that the overvaluation of companies _would_ be at or near dot-com bubble levels. These AI companies have plenty of venture capital, but consumers are probably not as influential as they were back then. I agree we likely won't see a dotcom-like crash, but there will still be fallout that will take months to settle.

omgJustTest2• 7 months ago

IMO what you are describing is distributed choice vs concentrated choice.

It's one of my main arguments against a crash: why would one (1) or a few choose to do that?

finghin• 7 months ago

Unless I misunderstand your question, isn’t the obvious answer just short-term profit?

No investor has thus far invested something they can’t yet cash out.

miohtama• 7 months ago

I sincerely hope this would kill Oracle with their karma

oldjim7982• 7 months ago

This can't happen soon enough. AI is a vacuum sucking up a lot of capital and attention that could and should be put to much much better uses

barbazoo• 7 months ago

Like what? Do you have any insight into what money might be spent on instead?

ishtanbul2• 7 months ago

Real infrastructure and housing

dwa3592• 7 months ago

Healthcare research to start with.

schnebbau• 7 months ago

We could go back to crypto and NFTs being the capital darlings?

dymk2• 7 months ago

Those did not suck up nearly the capital and attention that AI is. They were like GameStop and AMC.

tester756• 7 months ago

On the other hand we've never invested as much in hardware & software optimization as we do now

palmotea2• 7 months ago

On the other hand we've never invested as much in hardware & software optimization as we do now

Fat lot of good that does me: all the client-side software I use is as bloated as ever.

s1mplicissimus• 7 months ago

I tried figuring out how to bet on the AI bubble popping since the circle investments - that's by the way the one thing the article does not explain.

Can someone from the dear HN financially savvy users clarify what kinds of specific bets could be placed to that avail?

ebiester• 7 months ago

Just remember that the market can stay irrational longer than you can stay solvent.

Consider diversification in your portfolio. Maybe you divert a little more away from the US market, for example, which is heavily dependent on 7 stocks largely tied to AI.

Ekaros• 7 months ago

Also check how diversified the instruments you choose are. Sometimes they are lot less than one will think they are. Mainly due to those 7 stocks being big component in them also.

nairboon• 7 months ago

It really depends on what exactly you want to bet on and on what timeframe. More short term bet? Puts on the AI companies or an AI ETF. Do you assume that the rest of tech stays up even if AI pops? Then you could short some AI ETF and hedge with long QQQ. (=betting that the AI subset of Nasdaq will underperform relative to the Nasdaq.

infecto6• 7 months ago

We know over the long term it is exceptionally hard to beat the market, timing the market is near impossible. Everyone keeps talking about a bubble but we don’t know how big of one it is or when/if it will pop.

You are better off being in the market than betting on an idea that you don’t know will even happen or when.

I definitely think there is over enthusiasm in the space but at the same time I am not convinced that the demand for compute has let off yet.

My take is always you could build up some cash reserve in treasuries or somewhere like that and deploy it if a pop does happen. You will miss out on the potential growth but if you wanted to participate that is one way imo.

spwa42• 7 months ago

But isn't that in the past?

Look at stocks: everything is synchronizing, for years now. Either something like 85% of stocks all go up, with a predictable difference between them (meaning, e.g. META moves about double GOOG does, whichever way it goes, up or down), or 85% of stocks all go down. SPY, VOO. And in fact the only ones that make a move to speak of are the MAG7. It isn't just that they're the fastest to rise, they're the only ones that beat the market.

Zoom out and you'll see that in recent years you can include even non-stock-market assets in this argument. Housing ... same (of course there I understand), Gold, surprisingly, same.

And that's ignoring the warnings European authorities are issuing these days. It's pretty public information at this point that European authorities expect open ("kinetic" if you will) hostilities between Germany, France and Russia to open somewhere between March 2026 and Jan 2027. That will crash the stock market. That will crash the housing market. That will probably even crash the gold market, AI or no AI. Imho, that will crash the value of fucking Trump tower. The places these warnings are coming from are very serious and not known for joking on these matters (like the German chancillary, which if anything is far too conservative, or the French department of health, which has literally never issued a warning like this)

square_usual• 7 months ago

You are better off being in the market than betting on an idea that you don’t know will even happen or when.

This is true if you're willing to wait forever, but if you have near- and medium-term goals, you should not be investing money in the market if you believe there will be a crash. I have such goals and I'm putting my money into treasuries instead of putting more into what I believe is a very overvalued market.

electroly• 7 months ago

Just knowing that the bubble will pop at some point in the future isn't enough to trade on. You'll get trounced if this is the only piece of information you have. To a first approximation, everybody knows the bubble will pop. The question is: when and how?

AaronAPU• 7 months ago

Not a single person “knows” the bubble will pop.

Many think it will.

francisofascii• 7 months ago

ProShares UltraShort Technology (REW) You are betting against Nvidia, Microsoft, Apple, etc. Has a .95% expense ratio. https://www.proshares.com/our-etfs/leveraged-and-inverse/rew

eudamoniac• 7 months ago

Long dated puts on NVDA

rsynnott• 7 months ago

Generally it's not enough to know that something is a bubble; to bet against it you kind of need to know _when_ it collapses. For every Michael Burry there are a lot of investors who correctly identified that things were broken, but were sufficiently off on the timing that they made a loss on their bets anyway.

Now, granted, personally if I had, say, a lot of Oracle stock, I'd probably be getting rid of that. But unless you conveniently already have a bunch of bubble stock, there's not really a remotely safe way to play.

leviliebvin• 7 months ago

What is missing from the picture with all these articles is the numbers. LLMs already have a few solid use cases as translators, general document processors, coding helpers ...etc. So the first question is, to what extent does this demand support the investment? Would it be enough if basically every SP500 corp provided paid LLM access to their employees? Or is the investment so big, that people are betting on less solid applications, like Agentic AI, with some non-trivial automation?

gmerc• 7 months ago

Why pay much when you can get bare metal China LLM without the premium?

lostmsu• 7 months ago

Fully utilizing hardware is not trivial.

sleepyguy2• 7 months ago

Listened to Tom Lee on a Prof G Podcast and he made a lot of sense about it all.

https://youtu.be/2J_IGuA-IdY?si=uTptx9R-HMhjD9LH&t=1200

Worth listening to the entire podcast but this is a snip where he speaks about AI valuations. Somewhere in that podcast he brings up the fact that Costco is trading at double future earnings to NVIDIA, let that sink in.....

baal80spam• 7 months ago

Somewhere in that podcast he brings up the fact that Costco is trading at double future earnings to NVIDIA, let that sink in.....

English is not my first language, so let me make sure I understand: with this sentence you are suggesting that NVIDIA is undervalued, correct?

sleepyguy2• 7 months ago

Well according to that metric, Costco has better growth prospects than NVIDIA.

In my opinion NVIDIA has better prospects at future growth than Costco, but the market hasn't priced it that way.

g023• 7 months ago

I'm thinking that the bubble will be the vortex caused by an abundance of power that becomes freely available locally due to the AI datacenters moving to space.

OutOfHere• 7 months ago

Never take Wall Street at their word. They just want to take profit and buy back cheaply. It's not the bubble they're inventing; it's the pop.

Wall Street sees AI bubble coming and is betting on what pops it

Related Stories