I think the scepticism around whether the AI firms can provide the return on investment for their capital expenditure plan is completely understandable.
It reminds me of a mini-story within Michael Lewis' Flash Boys - there was a massive project to drill through the solid granite of Pennsylvania mountains in order to lay a new fiber-optic cable so that High-Frequency-trading firms could use it to get their data quicker between Chicago and New York. It was done at huge cost, I think around $300m. The Fiber-optic cable could transmit data between the locations in around 13.1ms - 13.5ms.
However, an alternative option of Microwave towers was setup and installed to transmit data between the same locations. The Microwave Tower could do the same, but around 8.5ms-9ms, ~30% faster. And it didn't cost anywhere near as much.
I worry that not only are those current capital expenditure plans wildly unaffordable, but also that the risk of an innovation rendering all that infrastructure obsolete can't be ruled out. It's a massive gamble.
adverbly 21 hours ago [-]
> risk of an innovation rendering all that infrastructure obsolete can't be ruled out. It's a massive gamble.
It's not just a gamble on technology - it's a gamble on the company too!
You own a company's stock - not a technology's!
In the dot com era, it was the realization that the wrong companies were picked that popped the bubble. Turns out pets.com was never gonna be the billion dollar business. It was that book store Amazon instead. And it was the Google search engine not the other one. And browser A not browser B... And so on... When the expected winner shifts away from one company towards a position where people want to sit on cash and wait and see, you suddenly get a massive rush to the exits and poor pets.com needs to throw in the towel.
We're gonna find out in the next couple of years who the pets.coms are.
jvanderbot 21 hours ago [-]
What would the equivalent be in AI? A dirt cheap inference chip for consumer hardware? 2T sized networks on a cell phone so we don't need data centers?
Your analogy required no technological innovation, just a good look at solution space. Where's the microwave tower for AI that everyone is missing?
lumost 18 hours ago [-]
A breakthrough in grokking/regularization, quantized or sparse training methods, quantum optimization, or more efficient architecture could reduce infrastructure needs. Although any of these could simply advance the parrot frontier without solving the prisoner’s dillema driving infrastructure spend at ai firms.
zipy124 19 hours ago [-]
More efficient architecture/training regimes would be a big one.
1vuio0pswjnm7 19 hours ago [-]
One difference is the HFT firms were certain that increased speed would translate to revenues that exceeded the project's expenditures
The "AI" firms have no such certainty that "AI" translates to revenues that will exceed their expenditures
All they have is endless speculation
Eddy_Viscosity2 22 hours ago [-]
Whenever the stock market goes down there is a default, likely algorithmically produced headline that says "This is the worst its been since the last time it was this bad which was X ago". Meaningless.
Esophagus4 21 hours ago [-]
The worst rainy second Tuesday of the month since 2023…
I see the same thing in sports broadcasting. “No 24 year old rookie player has started the season with this many at bats from the left side of the plate against visiting teams since…”
barbazoo 21 hours ago [-]
That made me laugh, they must get bombarded with stats at any time they can just pick and choose from. I bet to some degree it's also about filling the air with something to say.
Esophagus4 21 hours ago [-]
Agreed - I think they need to fill the air, but I also think the growth of data analytics makes these stats much easier to find now, whereas they were probably more obscure before.
potato3732842 21 hours ago [-]
I think that needlessly specific statistics is an in group joke among the trade. I can't tell, but it sure looks like it.
Esophagus4 21 hours ago [-]
I’d do the same, glad the stats department is having fun at work :)
potato3732842 21 hours ago [-]
I kind of look forward to AI written/approved headlines because we'll get more funny gems from the AI not getting the implications of out of scope context.
biophysboy 21 hours ago [-]
A reference point adds a small amount of meaning
AnimalMuppet 21 hours ago [-]
Worst since April? So, worst in 6 months? That's... not very impressive. It sounds like you should expect this to happen twice a year or so.
I'm not sure that they intended to give that impression...
baq 22 hours ago [-]
usually a good contra signal, anyway
layer8 22 hours ago [-]
Informing about the current value of X is not meaningless.
fleventynine 23 hours ago [-]
It's still above where it was a few weeks ago. I don't see why this is news.
JKCalhoun 22 hours ago [-]
It's the momentum, inertia, that is worrisome.
toomuchtodo 22 hours ago [-]
Bubble running out of steam.
22 hours ago [-]
hexbin010 22 hours ago [-]
Many people trade more than once every few weeks
bobsmooth 22 hours ago [-]
They shouldn't. Day trading is a plague.
lumost 22 hours ago [-]
Everybody is doing some form of rebalancing now. Major institutions have lower trading fees and do this more frequently.
techdmn 22 hours ago [-]
I'm curious about a system where capital gains are 100% for the first... I don't know, let's say a month. Then you ramp down over the course of the next year until it matches the regular income tax rate. I'm less concerned about the specific time periods than I am about the idea that it would be beneficial to society to have our financial systems encourage long-term thinking.
iso1631 5 hours ago [-]
In my tax jurisdiction, Capital Gains is a way lower percentage than Income Tax.
gosub100 21 hours ago [-]
Publicly traded companies are a plague. Have you seen the damage they inflict on the poor?
SoftTalker 21 hours ago [-]
The damage done to the poor are the roadblocks that prevent them from investing in them. Having to have a minimum income to open an IRA, etc.
jfengel 21 hours ago [-]
It's just a correction. Unless it isn't, in which case it might be a massive bubble bursting, followed by recession or even depression.
Everyone wants to know, so it's always news. Even though it usually isn't.
matltc 21 hours ago [-]
I don't think NASDAQ is technically in correction territory yet; I believe that would mean it's down 10% from its high
ajross 21 hours ago [-]
Exactly, the headline sort of paradoxically reflects the desire for news, and not news itself.
But still, the actual stock market behavior right now is PROBABLY (!!) more reflective of random motion than it is of a fundamental shift in investor behavior.
Unless it isn't.
fred_is_fred 23 hours ago [-]
It will go back up when Meta borrows from Amazon to buy capacity from Oracle who buys silicon from Broadcom to fund GPUs from nVidia to let OpenAI enhance their app to make Facebook posts.
CompoundEyes 22 hours ago [-]
They’re all carefree and confident the US government will bail them out on grounds of a national security threat in falling behind.
Too big to fAIl
potato3732842 21 hours ago [-]
I can't think of an industry[1] more deserving of being left high and dry and less able to garner public sympathy than NYC banking and we let them get a bailout.
But everyone hates CA, hates big tech, etc, etc, so maybe the political stars align and this will be the ones who finally set the "no bailout" precedent.
[1] Well actually I can now that I think about it and it's the beltway bandits but that's beside the point.
zerosizedweasle 21 hours ago [-]
I feel like they are just eating people on the bottom of the economy and then expect taxpayers to pay.
And it will go (way) down when the music stops on that particular game of musical chairs.
SaltyBackendGuy 22 hours ago [-]
One of the scariest quotes I read from Why Markets Fail - John Cassidy, is a quote from a big wig at a large financial institution.
While acknowledging being in a bubble the quote was
"While the music is playing, you have to get up and dance".
rufus_foreman 19 hours ago [-]
Chuck Prince from Citigroup. "When the music stops, in terms of liquidity, things will be complicated. But as long as the music is playing, you’ve got to get up and dance. We’re still dancing"
His explanation of the quote:
'"My belief then and my belief now is that one firm in this business cannot unilaterally withdraw from the business and maintain its ability to conduct business in the future,” Mr. Prince said."'
...
'"And if you are not engaged in business, people leave the institution, so it is impossible to say in my view to your bankers we are just not going to participate in the business in the next year or so until things become a little more rational," he said. "You can’t do that and expect to have any people left to conduct business in the future."'
By 1998, Yahoo was the beneficiary of a de facto Ponzi scheme. Investors were excited about the Internet. One reason they were excited was Yahoo's revenue growth. So they invested in new Internet startups. The startups then used the money to buy ads on Yahoo to get traffic. Which caused yet more revenue growth for Yahoo, and further convinced investors the Internet was worth investing in. When I realized this one day, sitting in my cubicle, I jumped up like Archimedes in his bathtub, except instead of "Eureka!" I was shouting "Sell!"
Это к вопросу о эпицикле Оракла-Нвидии.
zerosizedweasle 22 hours ago [-]
For sure. People seem to think that you can just do all the market moving announcements that Altman has done and then say "Oops I don't have the money". Crashing into reality will bring a lot of pain.
ajross 21 hours ago [-]
The problem with that story is the dates. If you sold in 1998 when PG realized the ponzi characteristic of the market[1] you'd have lost ("lost") a ton of money. The nasdaq composite in June of 1998 was about $1800. It reached a peak of $5000 (!!!!) just before the crash.
The advice is correct, but in practice it's only helpful if you can time the crash, which you can't. Cycle-driven run-ups in advance of bubble burst events can be shockingly long.
[1] Which is roughly where we are right now with the AI bubble.
spwa4 20 hours ago [-]
But it crashed down to $1300, so there's still something to be said about it.
nine_zeros 22 hours ago [-]
> By 2026, Nvidia was the beneficiary of a de facto Ponzi scheme. Investors were excited about AI. One reason they were excited was Nvidia's revenue growth. So they invested in new AI startups and cloud hyperscalers. The startups and hyperscalers then used the money to buy Nvidia chips. Which caused yet more revenue growth for Nvidia, and further convinced investors AI was worth investing in.
Fixed this for anyone still not processing the bubble.
rco8786 22 hours ago [-]
I think everyone recognizes the bubble, the question is when does it pop and how.
At the end of the day it's still all about timing the market, which is hard to impossible no matter the conditions.
zerosizedweasle 21 hours ago [-]
Truly life for everyone not partying at the top has been hell, the bubble has frozen everything. This feels endless. Endless hell.
rcbdev 20 hours ago [-]
I am certainly not at the top. My life is fine. What are you doing where it's not?
cindyllm 19 hours ago [-]
[dead]
jauntywundrkind 23 hours ago [-]
Still, up +25% since 6mo ago. Sooo...
zerosizedweasle 23 hours ago [-]
Yeah, there's still room to fall given how inflated it is. How much did those deal announcements add to market caps? And OpenAI can't pay for it so that will need to be taken off the top.
kleiba 21 hours ago [-]
OT: is it possible to change to a career on Wall St. when you're already in your 50s?
rufus_foreman 19 hours ago [-]
Absolutely. In the food delivery industry.
kleiba 4 hours ago [-]
Ha! :)
Rendered at 16:29:07 GMT+0000 (Coordinated Universal Time) with Vercel.
It reminds me of a mini-story within Michael Lewis' Flash Boys - there was a massive project to drill through the solid granite of Pennsylvania mountains in order to lay a new fiber-optic cable so that High-Frequency-trading firms could use it to get their data quicker between Chicago and New York. It was done at huge cost, I think around $300m. The Fiber-optic cable could transmit data between the locations in around 13.1ms - 13.5ms.
However, an alternative option of Microwave towers was setup and installed to transmit data between the same locations. The Microwave Tower could do the same, but around 8.5ms-9ms, ~30% faster. And it didn't cost anywhere near as much.
I worry that not only are those current capital expenditure plans wildly unaffordable, but also that the risk of an innovation rendering all that infrastructure obsolete can't be ruled out. It's a massive gamble.
It's not just a gamble on technology - it's a gamble on the company too!
You own a company's stock - not a technology's!
In the dot com era, it was the realization that the wrong companies were picked that popped the bubble. Turns out pets.com was never gonna be the billion dollar business. It was that book store Amazon instead. And it was the Google search engine not the other one. And browser A not browser B... And so on... When the expected winner shifts away from one company towards a position where people want to sit on cash and wait and see, you suddenly get a massive rush to the exits and poor pets.com needs to throw in the towel.
We're gonna find out in the next couple of years who the pets.coms are.
Your analogy required no technological innovation, just a good look at solution space. Where's the microwave tower for AI that everyone is missing?
The "AI" firms have no such certainty that "AI" translates to revenues that will exceed their expenditures
All they have is endless speculation
I see the same thing in sports broadcasting. “No 24 year old rookie player has started the season with this many at bats from the left side of the plate against visiting teams since…”
I'm not sure that they intended to give that impression...
Everyone wants to know, so it's always news. Even though it usually isn't.
But still, the actual stock market behavior right now is PROBABLY (!!) more reflective of random motion than it is of a fundamental shift in investor behavior.
Unless it isn't.
Too big to fAIl
But everyone hates CA, hates big tech, etc, etc, so maybe the political stars align and this will be the ones who finally set the "no bailout" precedent.
[1] Well actually I can now that I think about it and it's the beltway bandits but that's beside the point.
Edit: People are literally being forced out of the country by cost of living https://www.wsj.com/personal-finance/retirement/middle-class...
While acknowledging being in a bubble the quote was "While the music is playing, you have to get up and dance".
His explanation of the quote:
'"My belief then and my belief now is that one firm in this business cannot unilaterally withdraw from the business and maintain its ability to conduct business in the future,” Mr. Prince said."'
...
'"And if you are not engaged in business, people leave the institution, so it is impossible to say in my view to your bankers we are just not going to participate in the business in the next year or so until things become a little more rational," he said. "You can’t do that and expect to have any people left to conduct business in the future."'
-- https://archive.nytimes.com/dealbook.nytimes.com/2010/04/08/...
To me it looks like history proved him right. The largest institutions were the ones that got bailed out.
By 1998, Yahoo was the beneficiary of a de facto Ponzi scheme. Investors were excited about the Internet. One reason they were excited was Yahoo's revenue growth. So they invested in new Internet startups. The startups then used the money to buy ads on Yahoo to get traffic. Which caused yet more revenue growth for Yahoo, and further convinced investors the Internet was worth investing in. When I realized this one day, sitting in my cubicle, I jumped up like Archimedes in his bathtub, except instead of "Eureka!" I was shouting "Sell!"
Это к вопросу о эпицикле Оракла-Нвидии.
The advice is correct, but in practice it's only helpful if you can time the crash, which you can't. Cycle-driven run-ups in advance of bubble burst events can be shockingly long.
[1] Which is roughly where we are right now with the AI bubble.
Fixed this for anyone still not processing the bubble.
At the end of the day it's still all about timing the market, which is hard to impossible no matter the conditions.