Wait so AI data centres can’t be built because of the limited availability caused by other AI data centres.
Which would lock all the AI CEOs in a big room and throw away the key.
It seems clear to me that cloud LLMs have the same scaling problems and demand concerns that cloud gaming did.
Oh well, maybe regular people will have to still run software locally after all. 🤷
Everyone was so expecting this, and yet…
Please let this be the beginning of the bubble burst 🤞
Also, what do you mean data centers aren’t all dark rooms lit exclusively by RGB lighting?!? I feel lied to…
It’s the beginning of reality biting for sure.
I think the big one will be when companies like openAI and anthropic have to file audited books in order to IPO (which they both want to do).
I thought they were Disco Balls set to spin forever with Shoebody bop playing in the background
“Looks like RAM is back on the menu, boys!”
To address shortages, companies are turning to global markets. As a result, Canada, Mexico, and South Korea became the biggest suppliers of high-power transformers for AI data centers to AI data centers.
God I wish democracy meant that we could vote on decisions like this
God I wish democracy meant that we could vote on decisions like this
You can! Only problem is that it’s one vote per dollar instead of one vote per person.
If you don't remember the ballot you cast It's printed on every receipt you were passed Each time you selected our products and services We were elected in each of your purchasesCyberstream by the Stupendium
it’s one vote per dollar instead of one vote per person.
Damn that is well-said
The bubble is actively breaking. Next couple months are going to be awesome. Along with oil prices making people think about getting an EV.
oracle slashed 30k jobs recently, yea its definitely getting worst. They might try to move the datacenters to India, bangladesh, SE asia where theres less resistance against regulation. most likely exacerbated by thier purchase of WB which netflix probably instigated them to overpay, and a cancellation fee of billions to netflix.
All that venture capital. Gone into rich people’s pockets.
It’s debt stacked on debt. The money isn’t real, just the consequences of decades of credit expansion.
Now we either deal with a cascading wave of defaults (a la '29, '73, '86, '08, and '23) or we rush in with state credit to bail out all the private lenders.
But there’s very little real money in real pockets. At the end of the day, it’s borrowing power that makes you a billionaire.
The venture capital is just piles of money in other rich peoples pockets. As I understand it, the AI bubble is funded by massive amounts of saudi oil money anyway.
Its kinda like the jokers cash bonfire, but its Sam Altman pleading for more of it while eagerly dousing it in gasoline.
and ME countries likely isnt getting much influence out of trump, and white house, as they claimed too. they are paying for nothing. like they were already investing in sportswashing, EA, and other ventures at least that has value to SAUDIs.
its all a pissing contest between people we really aught to hang from the rafters… The consequences of their pissbaby behaviour is ours to bare.
We should be making soup out of them.
Makes sense. It takes years to complete construction projects and budgets always overrun. AI is a fad for most companies trying to get into it.
Think about the dotcom bubble. We went from having all of these businesses that shot to the moon for valuation… and then it collapsed. Feels exactly the same. A few will survive to become metas and googles because it is not a worthless technology, but most of them are going to be worthless companies when investors decide it’s time to run.
markets are now starting to punish tech companies that overextend on AI, a realisation has rapidly kicked in that Al is far from profitable.
Winter is coming
Think about the dotcom bubble. We went from having all of these businesses that shot to the moon for valuation… and then it collapsed.
Amazon and Google and basically all of FAANG/MAMA beg to differ.
And basically every company still has a website and ecommerce in general is still massive.
The Dot Com bubble was brutal. But it was not a collapse of the industry. It was much more of a correction after so many companies blew up to sell services like “rate my dog” and many of the smaller ecommerce sites were absorbed by larger ones.
And that is likely what will happen with generative AI. I expect a MUCH bigger bloodbath for openai but anthropic seem to have scaled a lot better (although they also are going into their IPO with a massive data breach…). But expect most of the smaller companies to similarly get gutted.
But
Makes sense. It takes years to complete construction projects and budgets always overrun. AI is a fad for most companies trying to get into it.
Data Centers exploding… honestly have little to do with AI. No… not the way they explode in the UAE… metaphorical exploding. The reality is that data centers are GOOD money. Basically every company needs data hosting and offloading internal infrastructure is… honestly a really smart play. Same with the massive push for streaming of basically everything so that nobody owns anything and subscribe to everything. You want regional data centers and… this is how you get them.
Amazon and Google and basically all of FAANG/MAMA beg to differ.
But it was not a collapse of the industry.
I mean they don’t beg to differ at all and it absolutely collapsed. The result of which is the reason why FAANG as an acronym exists. FAANG is the resulting consolidation of the industry made possible by its widespread collapse.
It collapsed into… 4 or 5 of the most powerful companies in human history.
And that is the thing to understand. These bubbles are less about widespread collapse and more about consolidation. And it tends to be the smaller companies that suffer, not the biggies who were driving much of it to begin with.
That’s a bubble, quite literally. Collapsing into other companies. Do you think when people say bubble they mean that all companies related to LLMs will go under forever?
You’re just describing bubbles. That a few survived is not the strong point you think it is.
Data centers existing makes sense, but this specific aggressive AI data center buildout (with special-purpose hardware) doesn’t: the two AI companies you mentioned, OpenAI and Anthropic, aren’t making a profit, and they don’t appear to have a viable path to one. OpenAI claims it’ll be wildly profitable in just a few years, but they don’t go into how.
“aren’t making a profit” gets into the mess that is book keeping and is a giant rabbit hole people actively avoid because it is just easier to get angry at stupidity rather than complex malfeasance.
But what makes something an “AI data center” outside of the branding?
The reality is that it is a shit ton of computers connected to a really fast internet connection. Preferably through a properly managed set of switches but you do you. And the reason that we still mostly use GPUs for “AI” rather than highly specialized hardware (although, nvidia DID just buy groq a few months back…) is for that reason. They might do linear algebra of quarter precision floats REALLY well but they also do linear algebra of single and double precision floats pretty well too. And the CPUs and mobos (that are mostly optimized for data movement to offload to said GPUs) are no slouches either.
Which is what most of these companies are planning for. openai is, arguably, really fucking stupid. Whereas anthropic have shown decent signs of “diversifying” as it were. And nvidia… if we lived in a world where they could get enough RAM I think they would be fine. As it stands… Jensen (and a LOT of people) are kinda fucked and I expect to see a hard pivot over the next 12 months.
Because if we banned ALL generative AI tomorrow? The people who think you can’t use a computer without installing litellm first are gonna be fucked. But everyone else will just put other workloads on there and be… “fine” is a strong word but they won’t go bankrupt. And the data centers themselves will still be incredibly valuable.
I wish GPUs in AI data centers (or worse, the ones purchased and not installed yet) were more general-purpose than they appear to be. That’s the part that makes them AI data centers: the optimized hardware.
I do agree things are complex. And I like reading about the intricacies of that complexity. The overall picture is still a pretty bad one, though.
Ehhhh.
Yes, there are some fairly revolutionary(-ish) chips. Those are few and far between because they tend to be hyper specialized. Inference but not training or only optimized for a very small input matrix (common for edge computing like cameras).
By and large? They really ARE “traditional” GPGPUs that are optimized to hell and back for vector operations and linear algebra. And a lot of the gains there come from multiplying their floating point performance by 2-4 (depending on if half or quarter precision). They aren’t as good for double precision as something optimized for it but basically only a very small subset of users need that. There will be no issues repurposing the hardware in these data centers.
And the rest is data movement which has always been the real problem.
I don’t think most companies will find much value in that though. I know that none of the infrastructure I work with uses heavy calculations, and if we tried to jam it in, we’d be making solutions looking for problems.
An email server doesn’t need a GPU, neither does a file server, or a website, or an e-commerce platform.
Suppose they could rent it out as supercomputers but I don’t think the return on cost is going to be that good.
U.S. still depends on China, despite years of onshoring efforts.
Reminds me of that time when one president signed a bill to move chip production to the US then the next president raided the chip-making facility and arrested all those chip-making people for trying to show the locals how to make chips.
An absolute disgrace
wasnt that HYUNDAI cars?.
That was the biggest failure so far. The other ones seem to be smaller
one president signed a bill to move chip production to the US then the next president raided the chip-making facility and arrested all those chip-making people for trying to show the locals how to make chips
Chip making production has begun to migrate largely as a consequence of the Trump tariffs. The Biden plan to send Intel a few billion in kickbacks in exchange for a chip fab industry that wasn’t the laughing stock of the planet only enriched shareholders and executives.
I’m not sure what you’re referring to. Maybe Operation Low Voltage, which was a raid on a Hyundai battery plant in Georgia? I can’t find anything about a raid on a chip fab plant.
In fairness to Trump leadership, a bunch of these workers were, in fact, overstayed on their visas. Although, as usual, Trump sent guys in with shotguns to do what a sternly worded letter would have just as easily accomplished.
But these are the two faces of Western Capital. The bailout and the beat down. Liberals throw money at the problem. Conservatives slap you around on the thinnest pretexts. Neither president seems to have benefited the US domestic consumer.
More so both are factors towards onshoring happening. Same with PVs and batteries. Though Trump’s other policy undermind a lot of that progress as well.
The CHIPs act also favored union work and even funneled some money to worker owned shops. That so far seemed to be toothless in keeping this admin aligned there
Would you mind elaborating a little more? I live under a rock but this seems interesting
The Hyundai engineers, technicians, and equipment installers were not teaching anyone how to make chips.
Makes you wonder what they’re going to do with all this hardware they’re buying? I hope that they (the ultra rich) are the ones to hold the bag for once in their lives.
Just kidding, they’ll probably get a bailout from our (US) corrupt government.
If I recall, the hardware isn’t actually “bought”.
Altman just signed super-not-binding pinky promises with Hynix & Samsung that he’d mop up all of their DRAM inventory.
And now that data center buildouts are getting shitcanned, we’re seeing Micron stock tumble, because they threw it all away for that sweet, sweet AI payout.
It’s a planning thing. My company has sewed up power capacity contracts with the utility providers very early in the process. We have outstanding capacity contracts available for future build out commitments. It’s the folks late to the party that are starting projects without a plan and then failing.
This is disconcerting. China’s not going to slow down on their build out of infrastructure.
And for those who spin this as a positive. AI is not all LLMs. Real diseases are being cured by the complex modeling, real world tangible products, like airplanes and ships, are being designed safer.
I speculate that there’s a good chance that the modeling will eventually help to resolve the climate issue too, rather than continue to contribute to it. Physics models become more robust for simulating nuclear fusion; logistics models for transportation and energy distribution too.
Living in a tourist town, I don’t want a large data center in my backyard either, but there are plenty of places that do and where it makes sense to do so both from a logistical and resource perspective.
We get behind the curve here and it’s going to be near impossible to catch up, and when the smart people can’t play ball with the newest toys, that leads to brain drain.
Oh fuck off with your whitewashing. Majority of these AI installations are LLMs. Look at what Oracle, Grok, OpenAI, Microsoft and more have been trying to build off. It’s all an infrastructure race that is hurting us all (physically too if you look at how they’re being powered).
The sooner it collapses, the better.
These data centers are NOT being used for research. Universities and research companies generally have their own computer clusters for that.
These are being designed and built for LLMs and LLMs only.
Plus, the US has already lost. We’ve cut off our allies, destroyed our trade partnerships, made the economy unsustainable, and also caused the very brain drain you mention here. We’re cooked for decades at the least.
Let the prices fall.
Let the prices hit the floor
Let the prices hit the floor
Let the prices hit the…FLOOOOOOOOOOOOOOOOOR
