The beginning of scarcity in AI

https://tomtunguz.com/ai-compute-crisis-2026/

by gmays • 3 months ago

197 227 3 months ago

Lapalux• 3 months ago

"The first hit is free....."

stupefy2• 3 months ago

What limits LLM inference accelerators? I heard about Groq (https://groq.com/) not sure how much it pushes away the problem.

vessenes8• 3 months ago

ASML only makes a certain number of machines a year that can do extreme ultra-violet lithography.

Also - turbine blades limit power, according to Elon.

Between them - we cannot chip fabs past a certain rate, and we cannot stand up the datacenter to run these desired chips past a certain rate. Different people believe one or the other is the 'true' current bottleneck. The turbine supply chain scaling looks much more tractable -- EUV is essentially the most complicated production process humans have ever devised.

ls612• 3 months ago

Presumably ASML can increase production if demand is high enough the question is over what time frame. 5 years seems plausible to me but I honestly don't know what that number is.

andai• 3 months ago

Is global compute bottlenecked by one company?

Miraste3• 3 months ago

If only there were some form of cheap, widely manufactured power generation technology that didn't use turbines... Are they really going to wait until 2030 to get more turbines rather than invest in solar?

utopiah3• 3 months ago

Is ASML really the bottleneck? Do you believe anybody but TSMC and few fabs could really use and acquire those machines? I don't know the throughput of a EUV device from ASML but I imagine you need :

- clean room, itself needing the infrastructure for it (size, airCo, filtering, electricity) and the staff to run and maintain that basically empty space - wafers to "print" on, so that's a lot of water and logistic to manipulate them (so infrastructure for clean water and all chemicals) also with dedicated staff - finally staff who would be able to design something significantly better than NVIDIA, Intel, Broadcom, IBM, etc while (and arguably that's the trickiest part IMHO) being able to get it good enough as at a scale that can be manufactured from their own fab.

so I'm wondering who can afford this kind of setup that can only then make use of ASML machines.

vessenes8• 3 months ago

It seems very possible that we have at least five years of real limitations on compute coming up. Maybe ten, depending on ASML. I wonder what an overshoot looks like. I also wonder if there might be room for new entrants in a compute-scarce environment.

For instance, at some point, could Coreweave field a frontier team as it holds back 10% of its allocations over time? Pretty unusual situation.

dist-epoch3• 3 months ago

Jensen just said that if the signal/commitments are there, ASML can scale in 2-3 years.

vessenes8• 3 months ago

With Anthropic buying compute in dark alleys I’d assume that day is coming..

mattas• 3 months ago

This notion that "we don't have enough compute" does not cleanly reconcile with the fact that labs are burning cash faster than any cohort of companies in history.

If I am a grocery store that pays $1 for oranges and sells them for $0.50, I can't say, "I don't have enough oranges."

earthnail• 3 months ago

If there were more oranges you’d pay less to buy them and your economics would work out.

0x3f2• 3 months ago

Not sure if this is a joke or not, but competitive pressure still exists. This only really holds if you're the only orange seller.

vessenes8• 3 months ago

You misunderstand.

"I built a ship to go to the Indies and bring back tea."

"Bro, the ship cost 100,000 pounds sterling and only brought back 50,000 pounds of tea. I don't care if you paid 12,500 pounds for the tea itself, you're losing money."

There is a very rational reason labs are spending everything they can get for more compute right now. The tea (inference) pays 60%+ margins. And that is rising. And that number is AFTER hyper scalars make their margins. There is an immense amount of profit floating around this system, and strategics at the edge believing they can build and control the demand through combined spend on training and inference in the proper ratios.

SpicyLemonZest2• 3 months ago

60%+ margins according to numbers which are not published publicly and have not AFAICT been audited.

Could they be accurate? Sure, I think people who claim this is impossible are overconfident. But I would encourage anyone who assumes they must be right to read a history of the Worldcom scandal. It's really quite easy for a person who wants to be making money (or an LLM who's been instructed to "run the accounts make no mistakes"!) to incorrectly categorize costs as capital investments when nobody's watching carefully.

FloorEgg• 3 months ago

There is a major logic flaw in what you're saying.

'If I am a grocery store that pays $1 for oranges and sells them for $0.50, I can't say, "I don't have enough oranges."'

How about 'if I'm a grocery store and I see no limit on demand for oranges at $.50 but they are currently $1, I can say 'if oranges were cheaper I could sell orders of magnitude more of them'.

Buying oranges for $1 and selling for $0.5 is an investment into acquiring market share and customer relationships and a gamble on the price of oranges falling in the future.

0x3f2• 3 months ago

acquiring market share and customer relationships

The whole setup rests on this, and it seems mythical to me. These guys have basically equivalent products at this point.

lelanthran4• 3 months ago

Buying oranges for $1 and selling for $0.5 is an investment into acquiring market share and customer relationships

It's a delusion that customers are going to remain with the behemoths when a Qwen model run by an independent is $10/m, unlimited usage.

This is not a market that can be locked-in with network effects, and the current highly-invested players have no moat.

eloisant2• 3 months ago

Selling below cost is also called "predatory pricing". Sadly it's legal in US but it's something wealthy companies do to kill competitors and end up with captive customers.

TeMPOraL• 3 months ago

You can if you're exhausting the global production of oranges.

deepseasquid• 3 months ago

The grocery store analogy works if compute is the orange.

But labs arent buying oranges — theyre buying the only orchard on the island, hoping it yields a fruit no ones grown yet. Burning $1B to net $500M isnt "I have too few oranges." Its "Im betting the farm Ill find a new one."

Both can be irrational. Theyre irrational in different ways.

isawczuk• 3 months ago

It's artificial scarcity. LLM inference will soon be commodity as cloud.

There is a 2-3years still before ASIC LLM inferences will catch up.

vessenes8• 3 months ago

I don't think so. GB200 prices are GOING UP. A100s are still expensive. This implies massive utilization and demand, no? These machines are not sitting idle, or prices would drop in the very competitive hyperscaler environment.

Morromist• 3 months ago

Hard to say at this point. I'm sure you can run your LLM chips 24/7 for training and for the public to make weird thirst-trap videos about Judy Hopps but how real is the utilization and demand, really? Maybe very real, maybe not, I don't think we can know yet.

Its like being back in 1850 and you build the world's first amusement park where the rides are free or very cheap. People are like Amusement parks are the next big thing since Steam Boats! And tons of other rich people start to build huge amusement parks everywhere. The people who are skilled at making amusement park rides will increase their prices, and since the first amusement parks are free so they can get the public going to them demand will be huge.

But how sustainable is that? - well obviously we know from history that amusement parks did, in fact, take over the world and most people spent virtually all their time and money at amusement parks - I think the Crimean War was even fought over some religious-based theme park in Israel - until moving pictures came out, so it worked out for them, but for AI?

LogicFailsMe• 3 months ago

so much for all that hardware that was going to be obsolete in 3 years...

observationist• 3 months ago

The problem with this idea is that someone can, and likely will, come up with the next best architecture that leapfrogs the current frontier models at least once a year, likely faster, for the foreseeable future. This means by the time you've manufactured your LLM on an ASIC, it's 4-5 generations behind, and probably much less efficient than current SOTA model at scale.

It won't make sense for ASIC LLMs to manifest until things start to plateau, otherwise it'll be cheaper to get smarter tokens on the cloud for almost all use cases.

That said, a 10 trillion parameter model on a bespoke compute platform overcomes a lot of efficiency and FOOM aspects of the market fit, so the angle is "when will models that can be run on an asic be good enough that people will still want them for various things even if the frontier models are 10x smarter and more efficient"

I think we're probably a decade of iteration on LLMs out, at least, and the entire market could pivot if the right breakthrough happens - some GPT-2 moment demonstrating some novel architecture that convinces the industry to make the move could happen any time now.

dmazin• 3 months ago

Constraints can lead to innovation. Just two things that I think will get dramatically better now that companies have incentive to focus on them:

* harness design

* small models (both local and not)

I think there is tremendous low hanging fruit in both areas still.

com2kid4• 3 months ago

China already operates like this. Low cost specialized models are the name of the game. Cheaper to train, easy to deploy.

The US has a problem of too much money leading to wasteful spending.

If we go back to the 80s/90s, remember OS/2 vs Windows. OS/2 had more resources, more money behind it, more developers, and they built a bigger system that took more resources to run.

Mac vs Lisa. Mac team had constraints, Lisa team didn't.

Unlimited budgets are dangerous.

phist_mcgee• 3 months ago

Perhaps its because american hyperscalers want unlimited upside for their capital?

tasoeur• 3 months ago

Though I do agree with you, I just came back from a trip to China (Shanghai more specifically) and while attending a couple AI events, the overwhelming majority of people there were using VPNs to access Claude code and codex :-/

busfahrer• 3 months ago

Low cost specialized models

Can you elaborate on this? Is this something that companies would train themselves?

jeffhwang• 3 months ago

On the Mac vs Lisa team, I generally agree but wasn't there a strong tension on budget vs revenue on Mac vs Apple II? And that Apple II had even more constrained budget per machine sold which led to the conflict between Mac and Apple II teams. (Apple II team: "We bring in all the revenue+profit, we offer color monitors, we serve businesses and schools at scale. Meanwhile, Steve's Mac pirate ship is a money pit that also mocks us as the boring Navy establishment when we are all one company!")

By the logic of constraints (on a unit basis), Apple II should have continued to dominate Mac sales through the early 90s but the opposite happened.

jackcviers3• 3 months ago

It has been a very bad bet that hardware will not evolve to exceed the performance requirements of today's software tomorrow, just as it is a bad bet that tomorrow someone will rewrite today's software to be slower.

cesarvarela• 3 months ago

Harness is a big one, Claude Code still has trouble editing files with tabs. I wonder how many tokens per day are wasted on Claude attempting multiple times to edit a file.

lpcvoid• 3 months ago

The future is now, I guess

dataviz10002• 3 months ago

What do you mean by harness here?

Ifkaluva• 3 months ago

When you go to the command line and type “Claude”, there is an LLM, and everything else is the harness

codybontecou• 3 months ago

pi vs. claude code vs. codex These are all agent harnesses which run a model (in pi's case, any model) with a system prompt and their own default set of tools.

ElFitz5• 3 months ago

It’s the tool that calls the model, give it access to the local file system, calls the actual tools and commands for the model, etc, and provide the initial system prompt.

Basically a clever wrapper around the Anthropic / OpenAI / whatever provider api or local inference calls.

christkv2• 3 months ago

Could not agree more, this will spur innovation in all aspects of local models is my hunch.

drra2• 3 months ago

Absolutely. Anyone working on inference token level knows how wasteful it all is especially in multimodal tokens.

aldanor• 3 months ago

Yep.

As a recent example in AI space itself. China had scarce GPU resources, quite obvious why => DeepSeek training team had to invent some wheels and jump through some hoops => some of those methods have since become 'industry standard' and adopted by western labs who are now jumping through the same hoops despite enjoying massive computeresources, for the sake of added efficiency.

czk• 3 months ago

"adaptive" thinking

itmitica• 3 months ago

The current inference system is on a down slope.

It remains to be seen what new wave of AI system or systems will replace it, making the whole current architecture obsolete.

Meanwhile, they are milking it, in the name of scarcity.

henry2023• 3 months ago

The US is bound by energy and China is bound by compute power. The one who solves its limitation first will end this “Scarcity Era”.

jakeinspace• 3 months ago

China is installing something like 500 GW of wind and solar per year now. Even if they're only able to build and otherwise access chips that have half the SoTA performance per watt, they will win.

odo12422• 3 months ago

Performance per dollar may be more important than performance per watt here, though

kjkjadksj• 3 months ago

Win what exactly?

CuriouslyC• 3 months ago

The dynamics vastly favor China, part of the reason the US sprinting towards "ASI" isn't totally boneheaded is that the US and its industry needs a hail mary play to "win" the game, if they play it safe they lose for sure.

leptons2• 3 months ago

I'd be fine with a world without AI, honestly. Nobody really wins this race except the very wealthy. And I don't think it's really going to play out the way the wealthy think it will. It's more like a dog catching a car than it is a race.

Miraste3• 3 months ago

China's domestic chips are increasingly close to state-of-the-art. The US electrical grid is... not.

thelastgallon4• 3 months ago

US energy is constrained by the utility monopolies/oligopolies which have to extract more rents, specifically by increasing costs. Their profit is a percentage of cost, these perverse incentives + oligopolies will make it increasingly expensive to make anything (including AI) in US.

hvb23• 3 months ago

Or simply by the fact that increasing production takes time? Any power plant takes years to build?

Years, is like a lifetime for AI at this point...

com2kid4• 3 months ago

To bang on the same damn drum:

Open Weight models are 6 months to a year behind SOTA. If you were building a company a year ago based on what AI could do then, you can build a company today with models that run locally on a user's computer. Yes that may mean requiring your customers to buy Macbooks or desktops with Nvidia GPUs, but if your product actually improves productivity by any reasonable amount, that purchase cost is quickly made up for.

I'll argue that for anything short of full computer control or writing code, the latest Qwen model will do fine. Heck you can get a customer service voice chat bot running in 8GB of VRAM + a couple gigs more for the ASR and TTS engine, and it'll be more powerful than the hundreds of millions spent on chat bots that were powered by GPT 4.x.

This is like arguing the age of personal computing was over because there weren't enough mainframes for people to telnet into.

It misses the point. Yes deployment and management of personal PCs was a lot harder than dumb terminal + mainframe, but the future was obvious.

space_fountain2• 3 months ago

I've seen this claimed, but I'm not sure it's been true for my use cases? I should try a more involved analysis but so far open models seem much less even in their skills. I think this makes sense if a lot of them are built based on distillations of larger models. It seems likely that with task specific fine tuning this is true?

com2kid4• 3 months ago

What are you trying to do?

Write code? No. Use frontier models. They are subsidized and amazing and they get noticably better ever few months.

Literally anything else? Smaller models are fine. Classifiers, sentiment analysis, editing blog posts, tool calling, whatever. They go can through documents and extract information, summarize, etc. When making a voice chat system awhile back I used a cheap open weight model and just asked it "is the user done speaking yet" by passing transcripts of what had been spoken so far, and this was 2 years ago and a crappy cheap low weight model. Be creative.

I wouldn't trust them to do math, but you can tool call out to a calculator for that.

They are perfectly fine at holding conversations. Their weights aren't large enough to have every book ever written contained in them, or the details of every movie ever made, but unless you need that depth and breadth of knowledge, you'll be fine.

rstuart41335• 3 months ago

I've seen this claimed, but I'm not sure it's been true for my use cases?

I'd be surprised if it isn't true for your use cases. If you give GLM-5.1 and Optus 4.6 the same coding task, they will both produce code that passes all the tests. In both cases the code will be crap, as no model I've seen produces good code. GLM-5.1 is actually slightly better at following instructions exactly than Optus 4.6 (but maybe not 4.7 - as that's an area they addressed).

I've asked GLM-5.1 and Opus 4.6 to find a bug caused by a subtle race condition (the race condition leads to a number being 15172580 instead of 15172579 after about 3 months of CPU time). Both found it, in a similar amount of time. Several senior engineers had stared at the code for literally days and didn't find it.

There is no doubt the models do vary in performance at various tasks, but we are talking the difference between Ferrari vs Mercedes in F1. While the differences are undeniable, this isn't the F1. Things take a year to change there. The performance of the models from Anthropic and OpenAI literally change day by day, often not due to the model itself but because of the horsepower those companies choose to give them on the day, or them tweaking their own system prompts. You can find no end of posts here from people screaming in frustration the thing that worked yesterday doesn't work today, or suddenly they find themselves running out of tokens, or their favoured tool is blocked. It's not at all obvious the differences between the open-source models and the proprietary ones are worse than those day to day ones the proprietary companies inflict on us.

dist-epoch3• 3 months ago

Buy new Macs from where? There is a shortage of RAM, SSD, GPUs, and the CPU shortage just started.

ethan_smith• 3 months ago

The mainframe/PC analogy is spot on. And the hardware floor keeps dropping - you can grab a mini PC with 32-64GB RAM for a few hundred bucks and run surprisingly capable quantized models locally. Something like https://terminalbytes.com/best-mini-pcs-for-home-lab-2025/ shows the kind of hardware that's now available at consumer prices. The "scarcity" framing only makes sense if you assume everyone needs frontier-tier models for everything.

dyauspitr2• 3 months ago

That’s nonsense. Local models don’t have any of the nuance in text responses. I find them more akin to GPT 3.5 than even 4.x

paulddraper• 3 months ago

This is wrong along multiple axes.

1. Supply can scale. You can point to COVID/supply-chain shocks, but the problem there is temporary changes. No one spins up a whole fab to address a 3 month spike. Whereas AI is not a temporary demand change.

2. Models are getting more efficient. DeepSeek V3 was 1/10th the cost of contemporary ChatGPT. Open weight models get more runnable or smarter every month. Cutting edge is always cutting edge, but if scarcity is real, model selection will adjust to fit it.

byyoung3• 3 months ago

distillation is an equalizing force

eloisant2• 3 months ago

Distillation doesn't give you an equivalent model.

yalogin• 3 months ago

Does this also mean ram prices are not coming down anytime soon?

stronglikedan• 3 months ago

they already are

dist-epoch3• 3 months ago

yes, and it will keep increasing

i_think_so• 3 months ago

Does this also mean ram prices are not coming down anytime soon?

One person replies "yes". Another replies "no".

This concludes our press conference.

<3 HN

wg06• 3 months ago

There's other side to it too.

Whoever running and selling their own models with inference is invested into the last dime available in the market.

Those valuations are already ridiculously high be it Anthropic or OpenAI to the tune of couple of trillion dollars easily if combind.

All that investment is seeking return. Correct me if I'm wrong.

Developers and software companies are the only serious users because they (mostly) review output of these models out of both culture and necessity.

Anywhere else? Other fields? There these models aren't any useful or as useful while revenue from software companies by no means going to bring returns to the trillion dollar valuations. Correct me if I'm wrong.

To make the matter worst, there's a hole in the bucket in form of open weight models. When squeezed further, software companies would either deploy open weight models or would resort to writing code by hand because that's a very skilled and hardworking tribe they've been doing this all their lives, whole careers are built on that. Correct me if I'm wrong.

Eventually - ROI might not be what VCs expect and constant losses might lead to bankruptcies and all that build out of data centers all of sudden would be looking for someone to rent that compute capacity result of which would be dime a dozen open weight model providers with generous usage tiers to capitalize on that available compute capacity owners of which have gone bankrupt and can't use it any more wanting to liquidate it as much as possible to recoup as much investment as possible.

EDIT: Typos

solenoid09375• 3 months ago

OpenAI has an absurdly high valuation given their cash burn vs RRR.

Anthropic's is far more reasonable.

It makes no sense to lump these two companies together when talking about valuation. They have completely different financial dynamics

wg06• 3 months ago

No matter how low and reasonably Anthropic is valued, don't think $200 Max plans are going to recoup the investment + some return on top because size of the software industry is not that huge and profit margins for AI inference aren't very high either.

christkv2• 3 months ago

It feels like a repeat of the dot com infrastructure buildup that spurred the whole 2005 explosion in affordable hosting and new companies. This will probably leave us massive access to affordable compute in a couple of years.

raw_anon_11114• 3 months ago

For power maybe. But the expected lifetime of GPU hardware is 3 years before they fail completely

classified2• 3 months ago

would resort to writing code by hand because that's a very skilled and hardworking tribe they've been doing this all their lives

Shush, don't tell that to the AI coding acolytes.

2001zhaozhao• 3 months ago

AKA, the beginning of big companies being able to roll over small companies with moar money

(note: I don't expect this to actually happen until the AI gets good enough to either nearly entirely replace humans or solve cooperation, but the long term trend of scarce AI will go towards that direction)

ttul• 3 months ago

Energy scarcity will drive more innovation in local silicon and local inference. Apple will be the unexpected beneficiary of this reality.

piokoch• 3 months ago

Well it's in the books. O(n^2) algorithms are bad in the long run, transformers algorithm has such complexity, so not a big surprise we hit the limits.

utopiah3• 3 months ago

Initially I thought "Well... good for AI companies because they can then charge more" but IMHO that's a very tricky position because it means the cheap wave is behind us.

It's one thing to "sell" free or symbolically cheap stuff, it's another to have an actual client who will do the math and compare expenditure vs actually delivered value.

classified2• 3 months ago

and compare expenditure vs actually delivered value

Which means that the hype production will be driven up another few notches to make people doubt their rational findings and keep them in irrational territory just a tad longer. Every minute converts to dollars spent on tokens.

keiferski5• 3 months ago

We just had a realization during a demo call the other day:

The companies that are entirely AI-dependent may need to raise prices dramatically as AI prices go up. Not being dependent on LLMs for your fundamental product’s value will be a major advantage, at least in pricing.

andersmurphy2• 3 months ago

Yup. Also regardless of price they need to spend more and more as the project collapses under the inevitable incidental complexity of 30k lines of code a day.

It's similar to how if you know what you're doing you can manage a simple VPS and scale a lot more cost effectively than something like vercel.

In a saturated market margins are everything. You can't necessarily afford to be giving all your margins to anthropic and vercel.

prox2• 3 months ago

I also can’t wait for the time when few know how to code. Just like how many folks don’t know html from css when the homebrew website went away.

Their might always be llms, but the dependence is an interesting topic.

finaard• 3 months ago

How is that surprising? We've been taking that into account for any LLM related tooling for over a year now that we either can drop it, or have it designed in a way that we can switch to a selfhosted model when throwing money at hardware would pay for itself quickly.

It's just another instance of cloud dependency, and people should've learned something from that over the last two decades.

keiferski5• 3 months ago

Not so much that it was surprising, rather that we looked at a competitor’s site and noticed that a) their prices went way up and b) their branding changed to be heavily AI-first.

So we thought, hmm, “wonder if they are increasing prices to deal with AI costs,” and then projected that into a future where costs go up.

We don’t have this dependence ourselves, so this seems to be a competitive advantage for us on pricing.

michaelbuckbee• 3 months ago

What's weird though is the bifurcation in pricing in the market: aka if your app can function on a non-frontier level AI you can use last years model at a fraction of the cost.

michaelje• 3 months ago

Absolutely. Pricing exposure is the quiet story under all the waves of AI hype. Build for convenience → subsidise for dependence → meter for margin is a well-worn playbook, and AI-dependent companies are about to find out what phase three feels like.

Hyperscalers are spending a fortune so we think AI = API, but renting intelligence is a business model, not a technical inevitability.

Shameless link to my post on this: https://mjeggleton.com/blog/AIs-mainframe-moment

muppetman• 3 months ago

No shit. People are just figuring this out now?

This is the “Building my entire livelihood on Facebook, oh no what?” all over again.

Oh no sorry I forgot, your laptops LLM can draw a potato, let me invest in you.

lioeters• 3 months ago

Indeed, it was clear from the beginning, "AI" companies want to become infrastructure and a critical dependency for businesses, so they can capture the market and charge whatever they want. They will have all the capital and data needed to eventually swallow those businesses too, or more likely sell it to anyone who wants the competitive advantage.

rybosworld• 3 months ago

Seriously.

We just had a realization during a demo call the other day

These tools have been around for years now. As they've improved, dependency on them has grown. How is any organization only just realizing this?

That's like only noticing the rising water level once it starts flooding the second floor of the house.

strife25• 3 months ago

Marginal costs matter in this world.

zozbot2343• 3 months ago

The companies that are entirely AI-dependent may need to raise prices dramatically as AI prices go up.

It's not that clear. Sure, hardware prices are going up due to the extremely tight supply, but AI models are also improving quickly to the point where a cheap mid-level model today does what the frontier model did a year ago. For the very largest models, I think the latter effect dominates quite easily.

bcjdjsndon4• 3 months ago

There's only so far engineers can optimise the underlying transformer technique, which is and always has been doing all the heavy lifting in the recent ai boom. It's going to take another genius to move this forward. We might see improvements here and there but the magnitudes of the data and vram requirements I don't think will change significantly

chewz5• 3 months ago

We are processing same data for the last 2 years.

Inference prices droped like 90 percent in that time (a combination of cheaper models, implicit caching, service levels, different providers and other optimizations).

Quality went up. Quantity of results went up. Speed went up.

Service level that we provide to our clients went up massively and justfied better deals. Headcount went down.

What's not to like?

lelanthran4• 3 months ago

> The companies that are entirely AI-dependent may need to raise prices dramatically as AI prices go up.
It's not that clear. Sure, hardware prices are going up due to the extremely tight supply, but AI models are also improving quickly to the point where a cheap mid-level model today does what the frontier model did a year ago.

I agree; I got some coding value out of Qwen for $10/m (unlimited tokens); a nice harness (and some tight coding practices) lowers the distance between SOTA and 6mo second-tier models.

If I can get 80% of the way to Anthropic's or OpenAI's SOTA models using 10$/m with unlimited tokens, guess what I am going to do...

CodingJeebus• 3 months ago

You also have to look at how exposed your vendors are to cost increases as well.

Your company may have the resources to effectively shift to cheaper models without service degradation, but your AI tooling vendors might not. If you pay for 5 different AI-driven tools, that's 5 different ways your upstream costs may increase that you'll need to pass on to customers as well.

anonyfox• 3 months ago

in fact I am betting opposite. frontier models are getting not THAT much better anymore at all, for common business needs at least. but the OSS models keep closing the gap. which means if trajectories hold there will be a near future moment probably where the big provider costs suddenly drop shaerply once the first viable local models consistently can take over tasks normally on reasonable hardware. Right now probably frontier providers rush for as much money as they possible can before LLMs become a true commodity for the 80% usecases outside of deep expert areas they will have an edge over as specialist juggernauts (iE a cybersecurity premium model).

So its all a house of cards now, and the moment the bubble bursts is when local open inference has closed the gap. looks like chinese and smaller players already go hard into this direction.

zozbot2343• 3 months ago

Local open inference can address hardware scarcity by repurposing the existing hardware that users need anyway for their other purposes. But since that hardware is a lot weaker than a proper datacenter setup, it will mostly be useful for running non-time-critical inference as a batch task.

Many users will also seek to go local as insurance against rug pulls from the proprietary models side (We're not quite sure if the third-party inference market will grow enough to provide robust competition), but ultimately if you want to make good utilization of your hardware as a single user you'll also be pushed towards mostly running long batch tasks, not realtime chat (except tiny models) or human-assisted coding.

sevenzero3• 3 months ago

This was as clear as the sky when the first llm based businesses popped up. How did you realize this only now?

keiferski5• 3 months ago

Replied here: https://news.ycombinator.com/item?id=47804804

And I don't really mean new businesses that are entirely built around LLMs, rather existing ones that pivoted to be LLM-dependent – yet still have non-LLM-dependent competitors.

bdangubic2• 3 months ago

same as Uber… in the beginning everyone pretty much new that the cost of rides cannot possibly be that cheap and that it is subsudized. once you corner the market etc people just got used to “real” prices to the poibt that now there are often cheaper alternatives than Uber but people still Uber…

sidewndr46• 3 months ago

Not really, the next move is to establish standards groups requiring the use of AI in product development. A mix of industry and governmental mandates. What you view are viewing as COGS instead becomes instead a barrier to entry.

onion2k• 3 months ago

The companies that are entirely AI-dependent may need to raise prices dramatically as AI prices go up

Or they'll price the true cost in from the start, and make massive profits until the VC subsidies end... I know which one I'd do.

andersmurphy2• 3 months ago

We don't know what anthropic's true costs are. So pricing that in is at best a guess.

accrual• 3 months ago

Not being dependent on LLMs for your fundamental product’s value

I think more specifically not being dependent on someone else's LLM hardware. IMO having OSS models on dedicated hardware could still be plenty viable for many businesses, granted it'll be some time before future OSS reaches today's SOTA models in performance.

Cthulhu_2• 3 months ago

That'll be (part of) the big market correction, but also speaking broadly; as investor money dries up and said investors want to see results, many new businesses or products will realise they're not financially viable.

On a small scale that's a tragedy, but there's plenty of analysts that predict an economic crash and recession because there's trillions invested in this technology.

bjornroberg• 3 months ago

I wonder if it could be that they won't because the real mechanism is that AI wrapper pricing power is weak (switching costs near zero) but state of the art models makes it difficult to lower prices due to higher cost.

thih9• 3 months ago

Also: AI dependance could be explicit AI API usage by the product itself, but also anything else, like: AI assisted coding, AI used by humans in other surrounding workflows, etc.

keiferski5• 3 months ago

Yeah that's actually what I initially meant: just dependence on AI as a technology, not purely the usage costs. I didn't spell that out well enough.

lowsong• 3 months ago

Any company that has become dependant on AI will struggle to survive from here on. By the time many teams realise it'll be too late.

throwaway290• 3 months ago

wasnt ai supposed to get us post-scarcity?

PessimalDecimal• 3 months ago

That worked out, for the founders of frontier labs at least.

tim333• 3 months ago

For the first time since the 2000s, technology companies are confronting the limits of their supply chain.

I thought there'd been a shortage of cheap GPUs since ChatGPT took off and also before that in various crypto booms. I'm not sure it's a new thing.

the_gipsy• 3 months ago

But that concerned mostly only gamers and cryptominers. AI is supposed to be replacing traditional software development, which affects everything.

sdevonoes• 3 months ago

It’s time to be AI-independent. It’s like AWS, for most of us, it’s not worth it.

KaiserPro• 3 months ago

one graph, One graph and the author is pinning an entire theory on it?

Infra is always limited, even at hyper scalers. This leads to a bunch of tools dfofr caching, profiling and generally getting performance up, not to mention binpacking and all sorts of other "obvious" things.

malshe• 3 months ago

On X I had seen him mostly posting memes so this post seems par for the course

losvedir• 3 months ago

Infra is always limited, even at hyper scalers

I think maybe infra is limited only at hyperscalers. For the rest of us it's just how much capacity to we want to rent from the hyperscalars.

It's kind of a recent cloud-native mindset, since back in the day when you ran your own hardware scaling and capacity was always top of mind. Looks like AI compute might be like that again, for the time being.

sph• 3 months ago

One graph, about 100 words, AI in the title: Hacker News front page.

Not bad for a coffee break of effort.

chatmasta• 3 months ago

Why is written with an assumption that we have finite hardware production capacity? Industrial processes can scale up, new factories can come online… it will take a while but the whole point of economics is that supply will scale to meet demand. The shortage is a temporary, point-in-time metric.

And that’s not considering the software innovation that can happen in the meantime.

Bengalilol2• 3 months ago

The economic hypothesis that has dominated the past hundred years is that economic growth is infinite because resources are infinite and (almost) free. We all know this is unrealistic and disconnected from our human condition.

Regarding "innovation", I agree with your idea. I even think that the major innovation will be to transpose models locally, using reduced infrastructures that will still be sufficient for the majority of use cases.

Bengalilol2• 3 months ago

... and I have this little idea in the back of my mind: when companies can no longer keep up with demand and people have (albeit more limited and reduced) local capacity, minds will start focusing on techniques (more humble and modest ones) to keep part of the system running locally, without dependency.

I know it may sound ridiculous, but it could actually become a way to break away from the business models that have been developed over the past few decades. Broadly speaking, this even amounts to saying that the biggest victims of AI could be the companies that bet on AI as a service.

Yet I know my vision is way too idealistic but I'm coming to imagine that a human brain, although less efficient in the long run, remains a reliable way to control the resulting costs and could even turn out to be more advantageous and more readily available than its silicon-based counterpart.

20after4• 3 months ago

The human brain is incredibly efficient (Approximately 20W of energy consumption¹). These AI systems use many orders of magnitude more energy than human equivalents.

1. https://pmc.ncbi.nlm.nih.gov/articles/PMC8364152/

bcjdjsndon4• 3 months ago

Neither is this the first time nor are they really confronting it

frigg• 3 months ago

The models have already plateaued, you don't need latest and greatest.

latentframe• 3 months ago

This isn’t really looking like AI scarcity it’s more like compute becoming the bottleneck : when the access depends on chips energy and capital it stops being a pure software game and the winners are often whoever can secure capacity first

siliconc0w• 3 months ago

Definitely feeling this - the subsidized subscription plans are already starting to buckle.

0xbadcafebee2• 3 months ago

This isn't the first time they've dealt with scarcity, there's been supply chain scarcity four times since 2000. Post-dotcom boom, CDMA scarcity, HDD/flash scarcity, Pandemic scarcity.

The scarcity isn't long-term. Like all manufactured products, they'll ramp up production and flood the market with hardware, people will buy too much, market will drop. Boom and bust.

We're also still in the bubble. Eventually markets will no longer bear the lack of productivity/profit (as AI isn't really that useful) and there will be divestment and more hardware on the market as companies implode. Nobody is making 10x more from AI, they are just investing in it hoping for those profits which so far I don't think anyone has seen, other than in the companies selling the AI to other companies.

But more importantly, the models and inference keeps getting more efficient, so less hardware will do more in the future. We already have multiple models good enough for on-device small-scale work. In 5 years consumer chips and model inference will be so good you won't need a server for SOTA. When that happens, most of the billions invested in SOTA companies will disappear overnight, which'll leave a sizeable hole in the market.

topherhunt2• 3 months ago

In 5 years consumer chips and model inference will be so good you won't need a server for SOTA.

Naw man, you crazy. If you tell me that in 5 years, consumer chips will be so good that I can run GPT-5.4-level AI on my phone, I'd find that plausible (I buy cheap phones). If you're telling me that in 5 years we won't need _servers_ because our _phones and/or desktops_ will be powerful enough to run the biggest newest LLMs in existence, I question your judgment, I think that prediction shows a deep uncreativity about how massively compute-hungry SOTA models will get.

The valuable things to do with inference will keep being a server niche because they'll keep being 1-2 OOM more compute-hungry than whatever consumer hardware can handle. Like gaming: my laptop can run games from 2015 at max settings no problem but the games actually worth getting excited about in 2026 still melt a $2k GPU, because whatever headroom the hardware gains, developers immediately spend on ray tracing and Nanite and modelling individual skin cells or whatever. I don't see any plausible reason to expect that the ceiling on "valuable server-side compute" or "inference capacity" will rise any more slowly than the on-device capability is rising.

My assumption is that in 2031, SOTA top-intelligence AI will be hosted on cloud servers like it is today, offering dirt-cheap access to capabilities we can't even dream of today, while your Android will be running some open-source GPT-5+ equivalent.

0xbadcafebee2• 3 months ago

The thing is SOTA has a plateau. All LLMs work on the same principle: input goes in for training, reinforced by humans. There is only so much input (all recorded human knowledge), only so many human tweaks, that can produce only so much increased signal-to-noise in output. The machine can't read your mind, and there is no one truthful answer to most questions, so there will always be a limit on how accurate or correct or whatever any response will get. So at some point, you just can't make a better response. The agent harness, prompts, etc, are the only way to get better, and that's gonna be open source.

Add to that the algorithmic improvements on inference that's making inference faster with more context and higher quality. TurboQuant is just one example, more methods are coming out all the time. So the inference is getting more efficient.

At the same time, hardware can kind of keep getting infinitely better. Even if you can't make it smaller, you can make it more energy efficient, improve multitasking, more GPU cores/RAM or iGPUs, pack in more chips, improve cooling, use new materials... the sky's the limit.

Add all 3 together and at some point you will get Opus 4.7 on a phone with 40 t/s. At that point there's no way I'm paying for inference on a server. You can do RAG on-device, and image/video/voice is done by multi-modals. I want my agent chats replicated, but that's Google Drive. I want the agent to search the web, but that's Google Search. So eventually we're back to just doing what we do today (pre-AI) only with more automation.

The really advanced shit will come in 10 years, when we finally crack real memory and learning. That will absolutely be locked up in the cloud. But that's not an LLM, it's something else entirely. (slight caveat that WW3 will delay progress by 10-20 years)

mystraline• 3 months ago

And folks are just now realizing the SaaS token provider rug-pull?

How convenient, especially since everything has some LLM slop interaction.

But that rug isnt going to pull itself!

AtlasBarfed• 3 months ago

Pay for the latest AI for EXCLUSIVE POWERS ...

Trying to up-tier fractional improvements in something that can't be quantified easily, and !BONUS! with gated access it can't be as easily analyzed by the (low profit) AI analyzers/benchmarkers.

Foster paranoia among top executives that the fractional/debatable improvement is a MUST HAVE to STAY COMPETITIVE in your industry.

Meanwhile, I have not seen any improvement in software in the now almost ?three to four? years than mainstream LLM and AI coding assistance has arrived on the scene.

Although I will hold out the possibility that software has actually gotten far worse for the end user, because AI code is being dedicated to revenue enhancement and dark data collection.

NoSalt• 3 months ago

A few years ago, I purchased a handful of 250GB SSDs from amazon for $17.00 each.

Last year, I purchased a few 8TB hard drives for $80.00 each.

Today, I am sad. ;-(

rafaelero• 3 months ago

Companies who could see it clearly and ignored the "AI is a bubble duh" crowd will ultimately get benefited by the GPUs they already acquired. The companies who acted cautiously will get burned.

cowartc• 3 months ago

The scarcity framing assumes compute is the bottleneck. For most production deployment's Ive seen, the actual bottleneck is evaluation and knowing what to trust.

You can throw cheaper models at a problem all day but, if you can't measure where the model fails on your data, You're just making mistakes faster at a lower cost.

Compute gets cheaper. Reliable evaluation doesn't.

sixhobbits• 3 months ago

There is a lot of demand still coming for sure but I think I'm more optimistic. Ready to eat my hat on this but

- higher prices will result in huge demand destruction too. Currently we're burning a lot of tokens just because they're cheap, but a lot of heavy users are going to spend the time moving flows over to Haiku or onprem micro models the moment pricing becomes a topic.

- data centers do not take that long to build, probably there are bottlenecks in weird places like transformers that will cause some hicups, but nvidia's new stuff is waay more efficient and the overall pipeline of stuff coming online is massive.

- probably we will see some more optimization at the harness level still for better caching, better mix of smaller models for some use, etc etc.

These companies have so much money and they at least anthropic and openai are playing winner takes it all stakes, with competition from the smaller players too. I think they're going to be feeding us for free to win favour for quite a while still.

Let's see though.

mark_l_watson• 3 months ago

I agree and I am amazed at how much money some individuals and also a friend's company burn on token costs. I get huge benefits from this tech just using gemini-cli and Antigravity a few times a week, briefly. I also currently invest about $15/month in GLM-5.1 running Hermes Agent on a small dedicated VPS - fantastically good value for getting stuff done and this requires little of my time besides planning what I need done.

I think the token burners are doing it wrong. I think that long term it is better to move a little slower, do most analysis and thinking myself, and just use AI when the benefits are large while taking little of my time and money to use the tools.

jFriedensreich• 3 months ago

This is probably even the "fun" part of the whole picture. The purely dystopia starts when investment firms just silently grow bigger and bigger data centers like cancer. There will be no press releases, no papers, no chance anyone without billions will even know the details yet alone get access. One day we realise the worlds resources (maybe not as in the paperclip maximiser, but as in memory, energy, GPUs, water, locations) are consumed by trading models and the data centres are already guarded by robot armies. While we were distracted frighting with anthropic and openAI the real war was already over. Mythos is one sign in this direction but i also met a few people who were claiming to fund fairly large research and training operations just by internal models working on financial markets. I have no way to verify those claims but this happened 3 times now and the papers/research they were working on looked pretty solid and did not seem like they were running kimi openclaw on polymarket but actual models on some significant funds. Would be really interested if anyone here has some details on this reality. I would also not be surprised if this is a thing that people in SF just claim to sound dangerous and powerful.

nicbou• 3 months ago

The fun times will also end for consumers soon enough, when the oligopolies are established and investors start asking for returns.

insane_dreamer• 3 months ago

Companies who become dependent on AI to "optimize" their offering/processes are going to be faced with some serious vendor lock-in unless they do it in such a way that they can swap out the foundation model.

skybrian• 3 months ago

That’s a lot of speculation based on one graph. It doesn’t cover whatever Google is doing, for example.

The beginning of scarcity in AI

Related Stories