The article claims that AI services are currently over-utilised. Well isn't that because customers are being undercharged for services? A car when in neutral will rev up easily if the accelerator pedal is pushed even very slightly, because there's no load on the engine. But in gear the same engine will rev up much less when the accelerator is pushed the same amount. Will there be the same overutilisation occurring if users have to financially support the infrastructure, either through subscriptions or intrusive advertising?
I doubt it.
And what if the technology to locally run these systems without reliance on the cloud becomes commonplace, as it now is with open source models? The expensive part is in the training of these models more than the inference.
> Will there be the same overutilisation occurring if users have to financially support the infrastructure, either through subscriptions or intrusive advertising?
> I doubt it.
I agree. Right now a lot of AI tools are underpriced to get customers hooked, then they'll jack up the prices later. The flaw is that AI does not have the ubiquitous utility internet access has, and a lot of people are not happy with the performance per dollar TODAY, much less when prices rise 80%. We already see companies like Google raising prices stating it's for "AI" and we customers can't opt out of AI and not pay the fee.
At my company we've already decided to leave Google Workspace in the spring. GW is a terrible product with no advanced features, garbage admin tools, uncompetitive pricing, and now AI shoved in everywhere and no way to granularly opt out of a lot of it. Want spell check? Guess what, you need to leave Gemini enabled! Shove off, Google.
I am totally on the other end of the spectrum. For $20 a month, the amount of value I get from ChatGPT is incredible. I can talk to it in voice mode to help brainstorm ideas, it teaches me about different subjects, it (+ claude code) helps me write boilerplate code so I can spend more time doing things I enjoy.
I'm going through the process of buying a home, and the amount of help its given is incredible. Analyzing disclosures, loan estimates, etc. Our accountant charges $200 an hour to basically confirm all the same facts that ChatGPT already gave us. We can go into those meetings prepped with 3 different scenarios that ChatGPT already outlined, and all they have to do is confirm.
Its true that its not always correct, but, I've also had paid specialists like real estate agents and accountants give me incorrect information, at the cost of days of scheduling, and hundreds of dollars. They also aren't willing to answer questions at 2am in the morning.
> At $249/month the market adoption will crash resulting in somewhere in the middle pricing that the market can bear
Or much like what is going to happen with Alexa, it just dies because the cost of the service is never going to align with “what the market can bear”. Even at $75/mo, the average person will probably stop being lazy and just go back to doing 10 minutes worth of searching to find answers to basic questions.
I feel like I can get all of that for free already. Not sure why I would pay a monthly subscription when I'm already getting Gemini across the Google ecosystem.
Lol what. Analyzing disclosures? What information of use could it possibly synthesize from a three column table with checkmarks in [no disclosure]? For your sake I sincerely hope you aren't using ChatGPT when you should be getting inspections.
Inspections are fairly worthless if you are remotely handy and competent at basic stuff. I don’t need someone to go through and catalog the make and models of all the appliances, and I can visually inspect my own water heater and furnace pretty trivially.
You can get “real” inspections done but they cost thousands of dollars and take a full or more day to do with multiple subject matter experts. Almost no one does this.
Waiving inspection other than for major material defect is what I’ve done for all my real estate transactions. I’m not putting in an offer to nickel and dime someone over trivial bullshit like a busted GFCI circuit. My offer simply accounts for the trivial odds and ends I’ll have to take care of. Plus I’d much rather get the work done myself since I don’t trust a seller to do anything but the bare minimum.
Every one of my friends who have had five figures or more of surprise repair work on homes they purchased all had an inspection done. None of those could have found the various hidden damages for those buildings short of destructive stuff like pulling drywall out or lifting up shingles from a roof. Don’t worry though, the inspectors found stuff like a bathroom faucet with a crack in the knob.
Liability for what? The only real liability in my state is for outright misrepresentation or fraud via failing to disclose. The disclosure form covers anything material I'd care about. Even then - good luck actually proving anything short of exceptional circumstances.
If you look at the standard offer document for waiving inspection it's pretty easy to walk it back. You're simply waiving a contingency - you can typically still inspect the property itself. I'm sure if you get way off the beaten path you are correct, but almost no one is engaging in totally non-standard contracts where I'm at.
I'm curious what liability you think would apply for an inspection that misses whatever it may be that ends up in dispute after the transaction closes - since the whole point in the inspection is finding that beforehand? If I find a material defect in the foundation after I close - it won't matter if I had an inspection or not. Unless I can prove the seller knew about it and failed to disclose.
And if I ever sell any properties - I will be pretty loath to sell to anyone demanding an inspection contingency. They are almost always used for nickel and dime BS that I really don't have time for. If you walk the place, get your inspector to do so too, and come up with a punch list and still want to make an offer, discount it appropriately and fix it yourself after you close. It's nearly always either pointless or used as a negotiation tool after the fact due to the fact buyers can expect a seller to not want to walk away from the middle of a transaction (sunk cost/time). I'd much rather take an offer at 5% less up-front than deal with someone wasting 30-45 days on the market and my time dealing with trivial items.
> Will there be the same overutilisation occurring if users have to financially support the infrastructure, either through subscriptions or intrusive advertising? > I doubt it.
Yea, I think this is wrong. The analogy is more like the App Store, in that there is very little to do currently other than a better Google Search with the product. The bet is that over time (short time) there are much more financially valuable use cases with a more mature ecosystem and tech.
Unlike the smartphone adoption era where everything happened rather rapidly, we're in this weird place where labs have invented a bunch of model categories, but they aren't applicable to a wide variety of problems yet.
The dial up -> broadband curve took almost a decade to reach penetration and to create the SaaS market. It's kind of a fluke that Google and Amazon came out of the dial up era - that's probably what investors were hoping for by writing such large checks.
They found chat as one type of product. Image gen as another. But there's really not much "native AI" stuff going about. Everyone is bolting AI onto products and calling it a done day (or being tasked with clueless leadership to do it with even worse results).
This is not AI. This is early cycle WebVan-type exploration. The idea to use AI in a given domain or vertical might be right, but the tools just don't exist yet.
We don't need AI models with crude APIs. We need AI models we can pull off the shelf, fine tune, and adapt to novel UI/UX.
Adobe is showing everyone how they're thinking about AI in photoshop - their latest conference showed off AI-native UX. And it was really slick. Dozens of image tools (relighting, compositing, angle adjustment) that all felt fast, magical, and approachable as a beginner. Nobody else is doing that. They're just shoving a chat interface in your hands and asking you to deal with it.
We're too early. AI for every domain isn't here yet.
We're not even in the dialup era, honestly.
I'd expect the best categories of AI to invest in with actually sound financials will be tool vendors (OpenRouter, FAL, etc.) and AI-native PLG-type companies.
Enterprise is not ready. Enterprise does not know what the hell to do with these APIs.
I've got 4 different chat windows open on 4 free plans. That's not counting the free IDE-autocomplete I use.
In any given day I never have no access to free LLM help.
Since all the models are converging onto the same level of performance, I mostly can't even tell responses from ChatGPT and Claude apart.
> Right now a lot of AI tools are underpriced to get customers hooked, then they'll jack up the prices later.
Good luck with that. I mean it.
The ChatAI TAM is now so saturated with free offerings that the first supplier to blink will go out of business before they are done blinking.
I see people (like sibling reply to parent) boasting about the amount of value they get from the $20/m subscription, but I don't see how that is $20 better than just using the free ChatAIs.
The only way out of the red for ChatAI products is to add in advertising slowly; they have to boil the frog. A subscription may have made sense when ChatGPT was the only decent game in town. Subscriptions don't make sense now - I can get 90% of the value of a ChatAI for 0% of the cost.
> The article claims that AI services are currently over-utilised. Well isn't that because customers are being undercharged for services?
Absolutely, not only are most AI services free but even the paid portion is coming from executives mandating that their employees use AI services. It's a heavily distorted market.
And a majority of those workers do not reveal their AI usage, so they either take credit for the faster work or use the extra time for other activities, which further confounds the impact of AI.
This is also distorting the market, but in other ways.
We're talking miraculous level of improvement for a SOA LLM to run on a phone without crushing battery life this decade.
People are missing the forest for the trees here. Being the go to consumer Gen AI is a trillion+ dollar business. How many 10s of billions you waste on building unnecessary data centers is a rounding error. The important number is your odds of becoming that default provider in the minds of consumers.
I used ChatGPT for every day stuff, but in my experience their responses got worse and I had to wait much longer to get them. I switched to Gemini and their answers were better and were much faster.
I don’t have any loyalty to Gemini though. If it gets slow or another provider gives better answers, I’ll change. They all have the same UI and they all work the same (from a user’s perspective).
There is no moat for consumer genAI. And did I mention I’m not paying for any of it?
It’s like quick commerce, sure it’s easy to get users by offering them something expensive off of VC money. The second they raise prices or offer degraded experience to make the service profitable, the users will leave for another alternative.
> The important number is your odds of becoming that default provider in the minds of consumers.
I haven't seen any evidence that any Gen AI provider will be able to build a moat that allows for this.
Some are better than others at certain things over certain time periods, but they are all relatively interchangeable for most practical uses and the small differences are becoming less pronounced, not more.
I use LLMs fairly frequently now and I just bounce around between them to stay within their free tiers. Short of some actual large breakthrough I never need to commit to one, and I can take advantage of their own massive spends and wait it out a couple of years until I'm running a local model self-hosted with a cloudflare tunnel if I need to access it on my phone.
And yes, most people won't do that, but there will be a lot of opportunity for cheap providers to offer that as a service with some data center spend, but nowhere near the massive amounts OpenAI, Google, Meta, et al are burning now.
As a regular user, it becomes increasingly frustrating to have to remind each new chat “I’m working on this problem and here’s the relevant context”.
GenAI providers will solve this, and it will make the UX much, much smoother. Then they will make it very hard to export that memory/context.
If you’re using a free tier I assume you’re not using reasoning models extensively, so you wouldn’t necessarily see how big of a benefit this could be.
In fact it's apparently $5.2 trillion by 2030 [0] (out of $6.7T total data center spend; meaning all of "traditional IT needs" are less than a quarter of the total). That's the total if you add up all of the firms chasing this opportunity.
I do wonder, if you (and the commenter you replied to) think this is a good thing, will you be OK with a data center springing up in your neighbourhood, driving up water or power prices, emitting CO2? Then if SOTA LLMs become efficient enough to run on a smartphone will you be OK with a data center bailout coming from your tax dollars?
My hot (maybe just warm these days) take is, the problem with voice assistants on phones is they have to be able to have reasonable responses to a long tail or users will learn not to use them, since the use cases aren’t discoverable and the primarily value is talking to it like a person.
So voice assistants backed by very large LLMs over the network are going to win even if we solve the (substantial) battery usage issue.
Why even bother with the text generation then? You could just make a phone call to an LLM with a TTS frontend. Like with directory enquiries back in the day. Which can be set up as easily as a BBS if you have a home server rack like Jeff Geerling makes youtube videos about.
Yes, over-utilization is a natural response to being undercharged. And being undercharged is a natural result when investors are throwing money at you. During bubbles, Silicon Valley often goes to "lose money, make it up with scale". With the vague idea that after you get to scale, THEN you can figure out how to make money. And fairly consistently, their idea for how to make money is "sell ads".
Past successes like Google encourage hope in this strategy. Sure, it mostly doesn't work. Most of of everything that VCs do doesn't work. Returns follow a power law, and a handful of successes in the tail drive the whole portfolio.
The key problem here doesn't lie in the fact that this strategy is being pursued. The key problem is that it is rare for first mover advantages to last with new technologies. That's why Netscape and Yahoo! aren't among the FAANGs today. The long-term wins go to whoever successfully create a sufficient moat for themselves to protect lasting excess returns. And the capabilities of each generation of AI leapfrogs the last so well that nobody has figured out how to create such a moat.
Today, 3 years after launching the first LLM chatbot, OpenAI is nowhere near as dominant as Netscape was in late 1997, 3 years after launching Netscape Navigator. I see no reason to expect that 30 years from now OpenAI will be any more dominant than Netscape is today.
Right now companies are pouring money into their candidates to win the AI race. But if the history of browsers repeats itself, the company that wins in the long-term would launch in about a year from now, focused on applications on top of AI. And its entrant into the AI wars wouldn't get launched until a decade after that! (Yes, that is the right timeline for the launch of Google, and Google's launch of Chrome.)
Investing in silicon valley is like buying a positive EV lottery ticket. An awful lot of people are going to be reminded the hard way that it is wiser to buy a lot of lottery tickets, than it is to sink a fortune into a single big one.
> Today, 3 years after launching the first LLM chatbot, OpenAI is nowhere near as dominant as Netscape was in late 1997.
Incorrect. There were about 150 millions Internet users in 1998, or 3.5% of total population. The number grew 10 times by 2008 [0]. Netwcape had about 50% of the browser market at the time [1]. In other words, Netscape dominated a small base and couldn’t keep it up.
ChatGPT has about 800 millions monthly users, or already 10% of total current population. Granted, not exclusively. ChatGPT is already a household name. Outside of early internet adopters, very few people knew who Netscape or what Navigator was.
Furthermore my point that the early market leaders are seldom the lasting winners is something that you can see across a large number of past financial bubbles through history. You'll find the same thing in, for example, trains, automobiles, planes, and semiconductors. The planes example is particularly interesting. Airline companies not only don't have a good competitive moat, but the mechanics of chapter 11 mean that they keep driving each other bankrupt. It is a successful industry, and yet it has destroyed tons of investment capital!
Despite your quibbles over the early browser market, my broader point stands. It's early days. AI companies do not have a competitive moat. And it is way to premature to reliably pick a winner.
Netscape in 1997/1998 had about 90% of the target market.
OpenAI today has about 30% of the target market, maybe? (seeing as how every single Windows installation comes with copilot chat already, it's probably less. Every non-tech user I know has already used copilot because it was bundled and Windows prompted them into using it with a popup. Only one of those non-tech users even heard of OpenAI, maybe 50% of them have heard that there are alternatives to Copilot, but they still aren't using those alternatives)
Not many people buy Windows, they buy laptops that happen to have Windows installed. IMO this is a worthwhile distinction because most people don’t really care about operating systems anyway, and would happily (I suspect, at least) use an Open Source one if it came installed and configured on a device that they got in a store.
Installing an OS is seen as a hard/technical task still. Installing a local program, not so much. I suspect people install LLM programs from app stores without knowing if they are calling out to the internet or running locally.
Besides the fact that this article is obviously AI generated (and not even well, why is there mismatches in british/american english? I can only assume that the few parts in british english are the human author's writing or edits), yes "overutilization" is not a real thing. There is a level of utilization at every price point. If something is "overutilizated" that actually means it's just being offered at a low price, which is good for consumers. It's a nice scare word though and there's endless appetite at the moment for ai-doomer articles.
Sorry but it's highly suspect to be spelling the same word multiple different ways across paragraphs. You switch between using centre/center or utilization/utilisation? It is a very weird mistake to make for a human.
I mix British and American English all the time. Subconsciously I type in British English but since I work in American English, my spell checkers are usually configured for en-US and that usually means a weird mix of half and half by the time I've fixed the red squiggles I notice.
> You sometimes see this with real live humans who have lived in multiple counties.
Also very common with... most Canadians. We officially use an English closer to British English (Zed not zee, honour not honor) however geographically and culturally we're very close to the US.
At school you learn "X, Y, Zed". The toy you buy your toddler is made for the US and Canadian market and sings "X, Y, Zee" as does practically every show on TV. The dictionary says it's spelled "colour" but most of the books you read will spell it "color". Most people we communicate with are either from Canada or the US, so much of our personal communication is with US English.
But also there are a number of British shows that air here, so some particularly British phrases do sneak in to a lot of people's lexicon.
See a similar thing in the way we measure things.
We use celsius for temperature but most of our thermostats default to Fahrenheit and most cookbooks are primarily in imperial measures and units because they're from the US. The store sells everything in grams and kilograms, but most recipes are still in tablespoons/cups/etc.
Most things are sold in metric, but when you buy lumber it's sold in feet, and any construction site is going to be working primarily in feet and inches.
If anything I expect any AI-written content would be more consistent about this than I usually am.
One of my least favorite things to come from AI is labelling any writing someone doesn't like as "obviously AI generated". I've read 3 of these kinds of comments on HN just today.
As non native English speaker I mix British and American English all the time, and you should hear me speaking. I mix in some novel accent too. Anyway, the author answered in a sibling reply.
Will the OpenRouter marketplace of M clouds X N models die if the investor money stops? I believe its a free and profitable service, offered completely pay as you go.
I don't. This is simply the "drug dealer" model where the first hit is free. They know that once people are addicted, they will keep coming back.
The question of course is, will they keep coming back? I think they very much will. There are indications that GenAI adoption is already increasing labor producitivity labor improvements at a national scale, which is quite astounding for a technology just 3 years old: https://news.ycombinator.com/item?id=46061369
Imagine a magic box where you put in some money and get more productivity back. There is no chance Capitalism (with a capital "C") is going to let such a powerful growth machine wither on the vine. This mad AI rush is all about that.
OpenAI has 800,000,000 weekly users but only 20,000,000 are paying while 780,000,000 are free riding. Should they by accident under provision then they could simply remove the freebee and raise the prices for the paying clients. But that is not what they want.
IMHO the investors are betting on a winner-takes-it-all market and that some magic AGI will be coming out of OpenAI or Anthropic.
The questions are:
How much money can they make by integrating advertising and/or selling user profiles?
What is the model competition going to be?
What is the future AI hardware going to be - TPUs, ASICs?
Will more people have powerful laptops/desktops to run a mid-sized models locally and be happy with it?
The internet didn't stop after the dotcom crash and the AI wont stop either should there be a market correction.
>OpenAI has 800,000,000 weekly users but only 20,000,000 are paying while 780,000,000 are free riding.
By itself, this doesn't tell us much.
The more interesting metric would be token use comparison across free users, paid users, API use, and Azure/Bedrock.
I'm not sure if these numbers are available anywhere. It's very possible B2B use could be a much bigger market than direct B2C (and the free users are currently providing value in terms of training data).
> How much money can they make by integrating advertising and/or selling user profiles?
I would say if executed well the revenue per user could be at least an order of magnitude more than Google search ads as the ads could be much more convincing and the information density is higher in chat.
I just thought about it, and honestly, from my surroundings of people aged 12-70s, across multiple continents, I can’t think of anyone who isn’t using some sort of LLM once a week.
American software engineer here, and I have only used LLM tools ~3 dozen times: tried at work a few times, and been unimpressed/actively frustrated; a few times to ask questions I would be normally survey blog posts about (social/lifestyle things); and the rest has been generating images with my kids.
But currently, aside from generating "creative media", I'd say I'm pretty much opposed to LLM tools. They have yet to demonstrate any value to me at work or with respect to the areas of research I am interested in, and given the kind of statistical mechanism that they are, I do not believe they are capable of doing so.
Interesting take, because I'm the opposite of it. My biggest use case is getting into a completely new topic, as it's the most frictionless starting point for most of the queries. Then I look around based on the rudimentary knowledge that I can gather from LLMs. However, I'm completely opposed to any sort of creative media created by LLMs and try to avoid it as much as I can (music, images, and etc.).
Also, it has become the natural workflow for me to throw bunch secondary priority work stuff to Claude and let it do its things, while I focus on the important stuff.
My point is, everyone finds a way to use it. Some are opposed to specific things, others are using other parts.
These takes feel like a failure compared to my daily usage, which is literally non-stop for ten hours a day. Want to construct a niche jq, curl, or find linux command? Can't remember the parameters for a function? Don't want to leave your terminal to search for something? ctrl+I and type in readable english.
I wonder if this is counting the "AI Summary" that is gratuitously included in a standard web search now?
That's the only AI I use anywhere near weekly. I have tried claude a few times, it was useless at helping me with my questions. I haven't really been back.
The thing that makes AI investment hard to reason about for individuals is that our expectations are mostly driven by a single person’s usage, just like many of the numbers reported in the article.
But the AI providers are betting, correctly in my opinion, that many companies will find uses for LLM’s which are in the trillions of tokens per day.
Think less of “a bunch of people want to get recipe ideas.”
Think more of “a pharma lab wants to explore all possible interactions for a particular drug” or “an airline wants its front-line customer service fully managed by LLM.”
It’s unusual that individuals and industry get access to basically similar tools at the same time, but we should think of tools like ChatGPT and similar as “foot in the door” products which create appetite and room to explore exponentially larger token use in industry.
> Think more of “a pharma lab wants to explore all possible interactions for a particular drug”
Pharma does not trust OpenAI with their data, and they don't work on tokens for any of the protein or chemical modeling.
There will undoubtedly be tons of deep nets used by pharma, with many $1-10k buys replacing more expensive physical assays, but it won't be through OpenAI, and it won't be as big as a consumer business.
Of course there may be other new markets opened up but current pharma is not big enough to move the needle in a major way for a company with an OpenAI valuation.
My claim is that there will exist some company which pharma is willing to trust for AI research…they presumably trust Microsoft with their email today.
But my bigger claim is that ~half the Fortune 500 will be able to profitably deploy AI with spends in the tens or hundreds of millions per year quite soon. Not that pharma itself is a major contributor to that effect.
But for 'AI' to be a winner-take-all market, it seems that the winner would have to be using customer data to improve the 'AI'. Not only do you have to believe that one of these (relatively) under-capitalized upstarts can corral the money, but also that they can convince (enterprise) customers to 'fork over' their proprietary data to only one provider, and also that the provider can then charge a monopoly rent.
Those all seem possible, but I wouldn't assign greater than a 50% probability to any of them, and the valuations seem to imply near-certainty.
When I'm building out a new feature, I can churn through millions of tokens in Claude code. And that's just me... Now think about Claude code but integrated with Excel or datadog, or whatever app could be improved through LLM integration. Think about the millions of office workers, beyond just software engineers, who will be running hundreds of thousands or millions of tokens per day through these tools.
Let's estimate 200 million office workers globally as TAM running an average of 250k tokens. That's 50 trillion tokens DAILY. Not sure what model provider profit per token is, but let's say it's .001 cents.
I find it absurd to pay for tokens I cant control, predict or even check in any reasonable way. It is literally amounts to "pay whatever random money the company asks you to pay" kind of contract.
> Are they unprofitable because they don't profit on inference, or because they reinvest all of the profit into scaling up?
They are scaling up using VC money, not revenue. As far as profit on inference goes, it's hard to separate it out from training: they cannot, at any given time, simply stop training because that would kill any advantage they have 6 months down the line.
For all practical purposes, you can't look at their inference costs independent of the training cost; they need to keep spending on both if they want to continue doing inference.
> Remember how long Amazon was unprofitable, by choice.
That was a very different scenario - AMZ was not spending their revenue on land-grabbing, they were spending their revenue on long-lived infra, while AI companies now are spending VC investment, not revenue, on land-grabbing.
The difference between spending your revenue on short-lived infra (training a new model, acquiring GPUs) and long-lived infra is that with long-lived infra, at any time, even after 10+ years, you can stop expanding your infra and keep the resulting revenue as profit.
With short-lived infra (models, GPUs), you can't simply stop infra spending and collect profit from the revenue, because the infra reached end-of-life and needs to be replaced anyway.
> “an airline wants its front-line customer service fully managed by LLM.”
This has been experimented on before by many companies over the recent years, most notably Klarna which was among the earliest guinea pigs for it and had to later on backtrack on this "novel" idea when the results came out.
But if I'm a pharma lab, I don't want to rely on a statistical engine that makes mistakes to answer those questions, I want to query a database that is deterministic.
Sam Altman says that he thinks scientific research is a huge opportunity for AI to contribute as it more fully develops and I think he’s right.
Since I’m not a scientific researcher, I have no idea if he’s just blowing smoke but I think it’s reasonable to think of a purpose-built system which has an LLM component being used by a team to do something useful.
This. LLM is NOT the tool for a pharma lab - properly trained ML is the right tool. Heck, English is probably not even the right LANGUAGE to use for discussing chemical interactions.
OpenAI has first mover advantage with respect to raising money but not wrt tech dominance.
Consensus theory: If AGI then superintelligence.
AI CapEx plans are not ROI based. Rather, they are the cost of "how do I remain competitive in the race to attain AGI" coupled with conveniently deep pockets. The money is being spent because the spenders can afford it and they see it as an existential risk as much as a profit opportunity.
Maybe OP's conclusion about the headline question blunts some political opposition to data centers, but that's not the salient issue.
The issue is this: America is betting a meaningful chunk of GDP that AGI is possible. This is The Manhattan Project 2.0.
- Lead Time vs. TCO vs. IRS Asset Deprecation: The moment you get it fully built, it's already obsolete. Thus from a CapEx point of view, if you can lease your compute (including GPU) and optimize the rest of the inputs for similar then your CapEx overall is much lower and tied to the real estate - not the technology. The rest is cost of doing business and deductible in and of itself.
- The "X" factor: Someone mentioned TPU/ASIC but then there is the DeepSeek factor - what if we figure out a better way of doing the work that can shortcut the workflow?
- AGI partnerships: Right now, you see a lot of Mega X giving billions to Mega Y because all of them are trying to get their version of Linux or Apache or whatever at parity with the rest. Once AGI is settled and confirmed, then most all of these partnerships will be severed because it then becomes which company is going to get their AI model into that high prestige Montessori school and into the right ivy league schools - like any other rich parent would for their "bot" offspring.
So what will it look like when it crashes? A bunch of bland empty "warehouses" with mobile PDU's once filling all their parking lot space gone. Whatever "paradise" that was there may come back... once you bulldoze all that concrete and steel. The money will do something else like a Don McLean song.
You're not quite thinking things through there man. Once the elites who built these follies have gone, the mob will go shopping for building materials. I wouldn't be surprised even if people end up living in these datacentres once they become derelict. They have AC after all.
Yes or no conclusions aside (and despite its title, the article deserves better than that), the key point is I think this one: “But unlike telecoms, that overcapacity would likely get absorbed.”
Stylistically, this smells like it was copy and pasted from straight out Deep Research. Substantively, I could use additional emphasis on the mismatch between expectations and reality with regards to telco debt-repayment schedule.
There's an enormous amount of unused, abandoned fiber. All sorts of fiber was run to last mile locations, across most cities in the US, and a shocking amount effectively got abandoned in the frenzy of mergers and acquisitions. 2 trillion seems like a reasonable estimate.
Giant telecoms bought big regional telecoms which came about from local telecoms merging and acquiring other local telecoms. A whole bunch of them were construction companies that rode the wave, put in resources to run dark fiber all over the place. Local energy companies and the like sometimes participated.
There were no standard ways of documenting runs, and it was beneficial to keep things relatively secret, since if you could provide fiber capabilities in a key region, but your competition was rolling out DSL and investing lots of money, you could pounce and make them waste resources, and so on. This led to enormous waste and fraud, and we're now on the outer edge of usability for most of the fiber that was laid - 29-30 years after it was run, most of it will never be used, or ever have been used.
I so desperately wish it weren't abandoned. I hate that it's almost 2026 and I still can't get a fiber connection to my apartment in a dense part of San Diego. I've moved several times throughout the years and it has never been an option despite the fact that it always seems to be "in the neighborhood".
That has nothing to do with fiber, it’s all about politics and a regulatory environment where nobody is incented to act. Basically, the states can’t fully regulate internet and the Federal government only wants to fund buildouts on a pork barrel basis. Most recently rural.
At the local level, there is generally a cable provider with existing rights of way. To get a fiber provider, there’s 4 possible outcomes: universal service with subsidy (funded by direct subsidy), cherry-picked service (they install where convenient), universal service (capitalized by the telco) and “fuck you”, where they refuse to operate. (ie. Verizon in urban areas)
The private capitalized card was played out by cable operators in the 80s (they were innovators then, and AT&T was just broken up and in chaos). They have franchise agreements whose exclusivity was used as loan collateral.
Forget about San Diego, there are neighborhoods in Manhattan with the highest population density in the country where Verizon claims it’s unprofitable to operate.
I served on a city commission where the mayor and county were very interested in getting our city wired, especially as legacy telco services are on the way out and cable costs are escalating and will accelerate as the merger agreement that formed Spectrum expires. The idea was to capitalize last mile with public funds and create an authority that operated both the urban network and the rural broadband in the county funded by the Federal legislation. With the capital raised with grants and low cost bonding (public authority bonds are cheap and backed by revenue and other assets), it would raise a moderate amount of income in <10 years.
We had the ability to get the financing in place, but we would have needed legislation passed to get access to rights of way. Utilities have lots of ancient rights and laws that make disruption difficult. The politicians behind it turned over before that could be changed.
The worst part is it'd probably cost less than $100 of fiber and labor to splice something into your building, maybe $200-400 of gear to light it up, and you'd have a 10 gbps pipe back to some colo. It's more economical to run new fiber in most places these days, even if the local ISP knows exactly where all the old abandoned legacy lines are run, because of subsidization and basically scamming. The big companies like Lumen keep their knowledge regionally compartmented, legally shielded, and deliberately obfuscated, because if it became known that existing fiber was already run to a place they claim they can't serve, they can't get access to yet more funding for their eternal "service for the underserved" government money grift.
I stumbled on old maps that showed a complete coverage of fiber in my municipality, paperwork from a company that was acquired, and which in turn merged, then was bought out by one of the big 5 ISPs. When local officials requested information regarding existing fiber, this ISP refused and said any such information was proprietary. They later bid on and won contracts to run new fiber (parallel to existing lines which they owned, which still had more than a decade of service life left in them at that point).
I estimated that only around 10-15% of the funding went toward actual labor and materials, the remainder was pure profit. The local government considered it a major victory, money well spent.
For infrastructure, central planning and state-run systems make a lot of sense - this after all is how the USA's interstate highway system was built. The important caveat is that system components and necessary tools should be provided by the competitive private sector through transparent bidding processes - eg, you don't have state-run factories for making switches, fiber cable, road graders, steel rebar, etc. There are all kinds of debatable issues, eg should system maintenance be contracted out to specialized providers, or kept in-house, etc.
The GDP 1995-2000 (inclusive) was about $52T. So that assertion would mean that about %3.8 of the US' economic activity was laying fiber. That seems like a lot, but in my ignorance doesn't sound totally impossible.
Don’t think looking at power consumption of b200s is a good measure of anything. Could well be an indication of higher density rather than hitting limits and cranking voltage to compensate
Yes, one of NVidia's selling points for the b200 is that performance per watt is better than before. High power consumption without controlling for performance means nothing.
In the telecoms case, just trowing in extra dark fiber just in case was an almost negligeable cost in light of digging new connections. So who cares if in year 1 you have just 4 fibers lit of the 244 you trew into the trench?
This is not the case for AI data centers at all. The compute is a major cost in the whole build budget. Having that installed and not used is financial ruin.
I am still a little skeptical about utilisation rates. If demand is so extreme, wouldn't we see rental prices for H100/A100 prices go up or maintain? Wouldn't the cost for such a gpu still be high (you can get em 3k used).
On "runpod community cloud" renting a 5090 costs $0.69/hour [1] and it consumes about $0.10/hour electricity, if running at full power and paying $0.20/kWh.
On Amazon, buying a 5090 costs $3000 [2]
That's a payback time of 212 days. And Runpod is one of the cheaper cloud providers; for the GPUs I compared, EC2 was twice the price for an on-demand instance.
Is there a way in which this is good for a segment of consumers? When the current gen of GPUs are too old, will the market be flooded with cheap GPUs that benefit researchers and hobbyists who therwis would not afford them?
GPUs age surprisingly gracefully. If a GPU isn't cutting edge, you just tie two or more of them together for a bit more power consumption to get more or less the same result as the next generation GPU.
if there's ever a glut in GPUs that formula might change but it sure hasn't happened yet. Also, people deeply underestimate how long it would take a competing technology to displace them. It took GPUs nearly a decade and the fortunate occurrence of the AI boom to displace CPUs in the first place despite bountiful evidence in HPC that they were already a big deal.
source: Nvidia Shields still shipping with Maxwell GPUs from 2014 and many university labs still running on Pascal GPUs from 2016. But stupid facts don't matter, this one hit you wrong in the feels, amIRight?
* The GPUs in use in data centers typically aren’t built for consumer workloads, power systems, or enclosures.
* Data Centers often shred their hardware for security purposes, to ensure any residual data is definitively destroyed
* Tax incentives and corporate structures make it cheaper/more profitable to write-off the kit entirely via disposal than attempt to sell it after the fact or run it at a discount to recoup some costs
* The Hyperscalers will have use for the kit inside even if AI goes bust, especially the CPUs, memory, and storage for added capacity
That’s my read, anyway. They learned a lot from the telecoms crash and adjusted business models accordingly to protect themselves in the event of a bubble crash.
We will not benefit from this failure, but they will benefit regardless of its success.
Why not? It’s 40A at 240V, or 25% of the continuous load rating of a 200A 240V single-phase service.
If someone can afford an 8 GPU server, they should be able to afford some #6 wire, a 50A 2P breaker, and a 50A receptacle. It has the same exact power requirements as an L2 EV charger.
In the US, you don’t need a neutral for a 240V 50A circuit on a residential single-phase service, there are two line conductors (120V to ground) both connected to a 2-pole single-phase breaker, line-to-line between the two is 240V.
You would need a neutral if it was a 208/120V three-phase service.
Neutrals and grounds are sized per the NEC, neutrals are the same size as the line conductors and equipment grounds are sized off of a table.
#6 conductors and #10 ground is what the NEC calls for.
I just haven't seen server rooms that don't demand a doubly sized neutral when one is required.
I live the NYC 208V life doing mostly resi, though.
Quick search of the spec for that is 6 power supplies, 2 of which are redundant. Looks to use a neutral to me. Says it uses C19/C20 connectors
edit: wait most ranges use 14-50R outlets and need a neutral ran. I am calling your statement into question. Surely harmonics and 120V internal draws cause non-zero neutral current. And I'm sure GPUs have harmonics being semiconductor flavored.
My bad, NEMA 14-50R is a 4-wire receptacle with a neutral.
I learned something new today, 200% neutrals are not required by the NEC but can help with non-linear loads, and certain transformers that mitigate harmonics need 200% neutrals.
I'm thinking you'd have a hard time fitting it in a smaller residential unit purely because of the load calculations. You'd probably have to add 3000W worth of AC units or heat pumps. Considering it's 4+ space heaters worth of output, I think you could fit it on a 200A service. 150A if you're bold and use optional calcs (these assuming you have electric cooking). To your point, basically the same as if you have an EV but with some cooling in there.
I just put in a dedicated 50 amp circuit myself and my car charges from ~25% to full in about 6-7 hours. But I wanted to present the lazy worst case scenario. The warrior's FUD here is that there isn't enough easily available lithium for everyone in just California alone to have one.
The 2001 telecoms crash drove benefits for companies that came later in the availability of inexpensive dark fiber after the bubble popped. WorldCom, ICG, Williams sold off to Verizon, Level 3, Teleglobe, and others. That in turn helped future Internet companies gain access to plentiful and inexpensive bandwidth.
Cable telephony companies such as Cablevision Systems, Comcast, Cox Communications, and Time Warner, used the existing coaxial connections into the home to launch voice services.
Rail and fiber deprecates on multiple decade timescales. AI data centers close to tulips. Even assuming we manage to make data center stretch to 10 years, these assets won't be around long enough to support ecosystem of new companies if the economics stops making sense. Ultimately the only durable thing is any power infra that gets built, vs rail and fiber where inheritance isn't just rail networks or fiber but like 1000s of kilometers of earthwork projects to build out massive physical networks.
Data centers last decades. Many or the current AI hosting vendors such as Coreweave have crypto origin. Their data centers were built out in 2010s, early 2020s.
Many of legacy systems still running today are IBM or Solaris servers at 20, 30 year old. No reason to believe GPU won’t still be in use in some capacity (e.g. interference) a decade from now.
Skeletons of data centers and some components (i.e. cooling) have long shelf life, but they're also ~10% of investment. Plurality of fiber and rail went towards building out linear infrastructure where improvements can be milked at nodes to improve network efficiency (better switches etc).
VS plurality of AI investment, i.e. trillions are going towards fast deprecating components where we can say with relative confidence will likely be net negative stranded assets in terms of amoritization costs if current semi manufacturing trends continues.
Keeping some mission critical legacy systems around is different than having trillions that makes no financial sense to keep on the books, i.e. post bubble new gen hardware will likely not have scarcity pricing or better compute efficiency (better apex and opex), there is no reason to believe companies will legacy GPUs around at scale if every rack loses them money relative to new hardware. And depending on actual commercialization compute demand, it can simply make more economic sense to retire them than keep them going.
Both. Semi vs concrete, depreciating vs durable assets. Durable linear assets you upgrade switches to improve fiber/rail which is where most of the investment was in. GPUs you replace the racks which is where most of the investment is in. Either way it cannot be stretched the same way materially and most important economically, i.e. new chips with better power efficiency = running old chips is literally losing money squatting on an data center slot. There is very little reason to believe new chips will cost more than legacy (current chips) for the simple reason much of currnet chips are acquired on scarcity pricing, i.e. Nvidia margins went from 50% to 70%. That 20% is massive capex premium that is not going to be competitive if bubble pops and Nvidia has to sell new hardware on commodity pricing, new hardware that is in all likelihood also going to be more compute efficient in terms of power (opex). Even if you stretch existing compute past 3-5 to 10 years it is still closer to tuplids than rail or fiber in terms of economically productive timescale.
TLDR old durable infra tends to retains positive residual value because they're not easy to replace economically/frequently, old compute has negative residual value because they are easy to replace economically/frequently.
This is indeed true, but doesn't fiber have a far longer lifetime than GPU heavy data centers? The major cost center is the hardware, which has a fairly short shelf life.
Well you still get the establishment of 1) large industrial buildings 2) water/electricity distribution 3) trained employees who know how to manage a data center
Even if all of the GPUs inside burn out and you want to put something else entirely inside of the building, that's all still ready to go.
Although there is the possibility they all become dilapidated buildings, like abandoned factories
The building and electrical infrastructure are far cheaper than the hardware. So much so that the electricity is a small cost of the data center build out, but a major cost for the grid.
Of the most valuable part is quickly depreciating and goes unused within the first few years, it won't have a chance for long term value like fiber. If data centers become, I don't know, battery grid storage, it will be very very expensive grid storage.
Which is to say that while there was an early salivation for fiber that was eventually useful, overallocation of capital to GPUs goes to pure waste.
I'm sure there are other "emerging" markets that could make use of the GPUs, I heard game streaming is relatively popular so you can play PC games on your phone for example. I'd guess things similar to that would benefit from a ton of spare GPUs and become significantly more viable.
>The building and electrical infrastructure are far cheaper than the hardware.
Maybe it's cheaper if we measure by dollars or something, but at the same time we lack the political will to actually do it without something like AI on the horizon.
Nuclear is not very controversial, there are tons of places that would be very happy to have additional reactors, namely those with successful reactors right now. It's just super expensive to build and usually a financial boondoggle.
AI companies are saying they are trying to build nuclear because it makes them sound serious. But they are not going to build nuclear, solar and storage is cheaper more flexible and faster to build. The only real nuclear commitment is Microsoft reopening an old nuclear reactor that had become uneconomic to operate. Building anything new would be a five+ year endeavor, if we were in a place with high construction productivity like China. In the US, new nuclear is 10 years away.
But as soon as Microsoft restarted an old reactor, all their competitors felt like they had to sound as serious, so they did showy things that won't result in solving their immediate needs. Everybody's renewable commitments dwarf their nuclear commitments.
AI companies can flaunt expensive electricity at high cost for high investor impact precisely because electricity is a small cost component of their inputs. It's a hugely necessary input, and the limiting factor for most of their plans, but the dollar amount for the electricity is small. The current valuations of AI assume that a kWh put towards AI will generate far far more value than the average kWh on the grid.
Welp Gemini got me. Using G3 to improve what I write, generate specific images, and use NotebookLM to dive into some research materials. Tried to do a bit each day with my free credits, but hit the limit too often. G2.5 was not nearly as useful. So I upgraded my baselevel Google workspace plan. Recently spoke to someone who is also using G3 a lot with good results. YMMV re: G3, but Google hooked me, and now I pay more. However I think it is worth it for what I do. G3 is my helpful, nerdy work mate. I never plan to use agenic AI. Not using ChatGPT much if any at all anymore. Sorry Sam.
I'm curious about the telecom collapse. How was it not in the pipeline that much better use of the fibres was around the corner? Surely they would have known that people were looking into it with some promise?
Simultaneous claims that 'agentic' models are dramatically less efficient, but also forecasts efficiency improvements? We're in full-on tea-leaves-reading mode.
No, because at least dark fiber is useful. AI GPUs will be shipped off to developing nations to be dissolved for rare earth metals once the third act of this clown show is over.
> You can already use Claude Code for non engineering tasks in professional services and get very impressive results without any industry specific modifications
After clicking on the link, and finding that Claude Code failed to accurately answer the single example tax question given, very impressive results! After all, why pay a professional to get something right when you can use Claude Code to get it wrong?
This seems to be either LLM AI slop or a person working very hard to imitate LLM writing style:
The key dynamic: X were Y while A was merely B. While C needed to be built, there was enormous overbuilding that D ...
Why Forecasting Is Nearly Impossible
Here's where I think the comparison to telecoms becomes both interesting and concerning.
[lists exactly three difficulties with forecasting, the first two of which consist of exactly three bullet points]
...
What About a Short-Term Correction?
Could there still be a short-term crash? Absolutely.
Scenarios that could trigger a correction:
1. Agent adoption hits a wall ...
[continues to list exactly three "scenarios"]
The Key Difference From S:
Even if there's a correction, the underlying dynamics are different. E did F, then watched G. The result: H.
If we do I and only get J, that's not K - that's just L.
A correction might mean M, N, and O as P. But that's fundamentally different from Q while R. ...
The key insight people miss ...
If it's not AI slop, it's a human who doesn't know what they're talking about: "enormous strides were made on the optical transceivers, allowing the same fibre to carry 100,000x more traffic over the following decade. Just one example is WDM multiplexing..." when in fact wavelength division multiplexing multiplexing is the entirety of those enormous strides.
Although it constantly uses the "rule of three" and the "negative parallelisms" I've quoted above, it completely avoids most of the overused AI words (other than "key", which occurs six times in only 2257 words, all six times as adjectival puffery), and it substitutes single hyphens for em dashes even when em dashes were obviously meant (in 20 separate places—more often than even I use em dashes), so I think it's been run through a simple filter to conceal its origin.
Remember we have about 20 years of poorly written articles along with a few well written ones for the LLM to be trained on. I'm confident that attempting to tell LLM from human writing is a waste of time now that the year is almost over.
Other than that I'd rather choose a comprehensive article than a summary.
I agree, and it feels like an allergy by now to that style specifically. This is doubly annoying because it ruins the reading experience and just makes me question myself constantly because you often can't be quite certain especially for shorter posts/comments.
On topic: It is always quite easy to be the cynical skeptic, but a better question in my view: Is the current AI boom closer to telecoms in 2000 or to video hosting in 2005? Because parallels are strong to both, and the outcomes vastly different (Cisco barely recovered by now compared to 1999 while youtube is printing money).
> This is the opposite of what happened in telecoms. We're not seeing exponential efficiency gains that make existing infrastructure obsolete. Instead, we're seeing semiconductor physics hitting fundamental limits.
What about the possibility of improvements in training and inference algorithms? Or do we know we won't get any better than grad descent/hessians/etc ?
No, because the datacenters will get used. The demand side exists, whether it’s LLM AIs or something completely different that isn’t AI related. That’s very different from a crash where there is absolutely nothing valuable/useable/demanded underneath the bubble.
‘LLM tuned GPUs’ are just GPUs. The tuning refers to the models and how they use something like CUDA or whatever. There was a GPU shortage even before LLMs properly burst onto the scene, back with crypto mining. Now, it’s possible TPUs might add a wrinkle to later demand side issues when there is a crash but that will depend on how useful TPUs actually end up being to those outside Google. But GPUs will remain useful, whether it’s for gaming, machine learning (not the AI slop variety of this, but more categorising like for self-driving cars or medical imaging etc), or for the next crypto scam. Surely you agree that powerful computing capacity, independent of AI scams, is here to stay, right?
My main point in arguing that now isn’t like 2000 is that unlike in 2000 we have actual hardware and physical assets underpinning this bubble. In 2000 the assets were literally just imaginary. Yes there is speculation now but it is underpinned by silicon that will still be worth decent money even after LLMs are exposed as a hallucinatory mirage.
Hardware growth is slow and predictable, but one breakthrough algorithm completely undercuts any finance hypothesis premised on compute not flowing out of the cloud and back to the edges and into the phones.
This is a kind of risk that finance people are completely blind to. Open AI won't tell them because it keeps capital cheap. Startups that must take a chance on hardware capability remaining centralized won't even bother analyzing the possibility. With so many actors incentivized to not know or not bother asking the question, there's the biggest systematic risk.
The real whiplash will come from extrapolation. If an algorithm advance shows up promising to halve hardware requirements, finance heads will reason that we haven't hit the floor yet. A lot of capital will eventually re-deploy, but in the meantime, a great deal of it will slow down, stop, or reverse gears and get un-deployed.
AI had a kind of Jevons paradox approach to efficiency improvements, unfortunately - if you halve the compute requirements with an algorithmic advance, you can run a model twice as big.
The large SOTA models have hit very diminishing returns on further scaling, I think. So you’d rather double the number of models you can run in parallel.
The article claims that AI services are currently over-utilised. Well isn't that because customers are being undercharged for services? A car when in neutral will rev up easily if the accelerator pedal is pushed even very slightly, because there's no load on the engine. But in gear the same engine will rev up much less when the accelerator is pushed the same amount. Will there be the same overutilisation occurring if users have to financially support the infrastructure, either through subscriptions or intrusive advertising?
I doubt it.
And what if the technology to locally run these systems without reliance on the cloud becomes commonplace, as it now is with open source models? The expensive part is in the training of these models more than the inference.
> Will there be the same overutilisation occurring if users have to financially support the infrastructure, either through subscriptions or intrusive advertising? > I doubt it.
I agree. Right now a lot of AI tools are underpriced to get customers hooked, then they'll jack up the prices later. The flaw is that AI does not have the ubiquitous utility internet access has, and a lot of people are not happy with the performance per dollar TODAY, much less when prices rise 80%. We already see companies like Google raising prices stating it's for "AI" and we customers can't opt out of AI and not pay the fee.
At my company we've already decided to leave Google Workspace in the spring. GW is a terrible product with no advanced features, garbage admin tools, uncompetitive pricing, and now AI shoved in everywhere and no way to granularly opt out of a lot of it. Want spell check? Guess what, you need to leave Gemini enabled! Shove off, Google.
I am totally on the other end of the spectrum. For $20 a month, the amount of value I get from ChatGPT is incredible. I can talk to it in voice mode to help brainstorm ideas, it teaches me about different subjects, it (+ claude code) helps me write boilerplate code so I can spend more time doing things I enjoy.
I'm going through the process of buying a home, and the amount of help its given is incredible. Analyzing disclosures, loan estimates, etc. Our accountant charges $200 an hour to basically confirm all the same facts that ChatGPT already gave us. We can go into those meetings prepped with 3 different scenarios that ChatGPT already outlined, and all they have to do is confirm.
Its true that its not always correct, but, I've also had paid specialists like real estate agents and accountants give me incorrect information, at the cost of days of scheduling, and hundreds of dollars. They also aren't willing to answer questions at 2am in the morning.
> For $20 a month, the amount of value I get from ChatGPT is incredible.
I agree. Wait until it's $249 a month. You'll feel differently.
At $249/month the market adoption will crash resulting in somewhere in the middle pricing that the market can bear
Competition and local SLMs will also provide a counter to massive price increase
> At $249/month the market adoption will crash resulting in somewhere in the middle pricing that the market can bear
Or much like what is going to happen with Alexa, it just dies because the cost of the service is never going to align with “what the market can bear”. Even at $75/mo, the average person will probably stop being lazy and just go back to doing 10 minutes worth of searching to find answers to basic questions.
This is what I believe will happen. Once pricing is raised to sustainable levels (regardless of profit) I think the customer base will drop off.
That is also a possibility
For what its worth, I already pay 20 for Chatgpt + 100 / month for Claude Code.
Well in some areas, the professionals are on the hook if they are incorrect, ChatGPT is not.
I feel like I can get all of that for free already. Not sure why I would pay a monthly subscription when I'm already getting Gemini across the Google ecosystem.
Yeah the premise of the comment is that at some point prices will have to raised, and when that happens, if people will churn
I learned today there are people who upload personal information to chatbots
Lol what. Analyzing disclosures? What information of use could it possibly synthesize from a three column table with checkmarks in [no disclosure]? For your sake I sincerely hope you aren't using ChatGPT when you should be getting inspections.
Unfortunately in SF at the moment, it seems basically all competitive offers need to waive inspection. Which is also what we did
The sellers already had an inspection done. The full report is over 100 pages
Who told you to waive inspection?
Inspections are fairly worthless if you are remotely handy and competent at basic stuff. I don’t need someone to go through and catalog the make and models of all the appliances, and I can visually inspect my own water heater and furnace pretty trivially.
You can get “real” inspections done but they cost thousands of dollars and take a full or more day to do with multiple subject matter experts. Almost no one does this.
Waiving inspection other than for major material defect is what I’ve done for all my real estate transactions. I’m not putting in an offer to nickel and dime someone over trivial bullshit like a busted GFCI circuit. My offer simply accounts for the trivial odds and ends I’ll have to take care of. Plus I’d much rather get the work done myself since I don’t trust a seller to do anything but the bare minimum.
Every one of my friends who have had five figures or more of surprise repair work on homes they purchased all had an inspection done. None of those could have found the various hidden damages for those buildings short of destructive stuff like pulling drywall out or lifting up shingles from a roof. Don’t worry though, the inspectors found stuff like a bathroom faucet with a crack in the knob.
Waiving inspection is waiving seller liability, you know that right?
Liability for what? The only real liability in my state is for outright misrepresentation or fraud via failing to disclose. The disclosure form covers anything material I'd care about. Even then - good luck actually proving anything short of exceptional circumstances.
If you look at the standard offer document for waiving inspection it's pretty easy to walk it back. You're simply waiving a contingency - you can typically still inspect the property itself. I'm sure if you get way off the beaten path you are correct, but almost no one is engaging in totally non-standard contracts where I'm at.
I'm curious what liability you think would apply for an inspection that misses whatever it may be that ends up in dispute after the transaction closes - since the whole point in the inspection is finding that beforehand? If I find a material defect in the foundation after I close - it won't matter if I had an inspection or not. Unless I can prove the seller knew about it and failed to disclose.
And if I ever sell any properties - I will be pretty loath to sell to anyone demanding an inspection contingency. They are almost always used for nickel and dime BS that I really don't have time for. If you walk the place, get your inspector to do so too, and come up with a punch list and still want to make an offer, discount it appropriately and fix it yourself after you close. It's nearly always either pointless or used as a negotiation tool after the fact due to the fact buyers can expect a seller to not want to walk away from the middle of a transaction (sunk cost/time). I'd much rather take an offer at 5% less up-front than deal with someone wasting 30-45 days on the market and my time dealing with trivial items.
Seller inspectors are notorious for missing important things.
Waiving inspections means even if they do, they’re not on the hook anymore.
>The disclosure form covers anything material I'd care about. Even then - good luck actually proving anything short of exceptional circumstances.
Which is why you get an inspection.
The market.
That's a bubble.
Indeed, but it can remain grossly overinflated longer than most buyers are willing to wait for it to pop.
> Will there be the same overutilisation occurring if users have to financially support the infrastructure, either through subscriptions or intrusive advertising? > I doubt it.
Yea, I think this is wrong. The analogy is more like the App Store, in that there is very little to do currently other than a better Google Search with the product. The bet is that over time (short time) there are much more financially valuable use cases with a more mature ecosystem and tech.
That's it.
We're in the "dial up era" of AI.
Unlike the smartphone adoption era where everything happened rather rapidly, we're in this weird place where labs have invented a bunch of model categories, but they aren't applicable to a wide variety of problems yet.
The dial up -> broadband curve took almost a decade to reach penetration and to create the SaaS market. It's kind of a fluke that Google and Amazon came out of the dial up era - that's probably what investors were hoping for by writing such large checks.
They found chat as one type of product. Image gen as another. But there's really not much "native AI" stuff going about. Everyone is bolting AI onto products and calling it a done day (or being tasked with clueless leadership to do it with even worse results).
This is not AI. This is early cycle WebVan-type exploration. The idea to use AI in a given domain or vertical might be right, but the tools just don't exist yet.
We don't need AI models with crude APIs. We need AI models we can pull off the shelf, fine tune, and adapt to novel UI/UX.
Adobe is showing everyone how they're thinking about AI in photoshop - their latest conference showed off AI-native UX. And it was really slick. Dozens of image tools (relighting, compositing, angle adjustment) that all felt fast, magical, and approachable as a beginner. Nobody else is doing that. They're just shoving a chat interface in your hands and asking you to deal with it.
We're too early. AI for every domain isn't here yet.
We're not even in the dialup era, honestly.
I'd expect the best categories of AI to invest in with actually sound financials will be tool vendors (OpenRouter, FAL, etc.) and AI-native PLG-type companies.
Enterprise is not ready. Enterprise does not know what the hell to do with these APIs.
I've got 4 different chat windows open on 4 free plans. That's not counting the free IDE-autocomplete I use.
In any given day I never have no access to free LLM help.
Since all the models are converging onto the same level of performance, I mostly can't even tell responses from ChatGPT and Claude apart.
> Right now a lot of AI tools are underpriced to get customers hooked, then they'll jack up the prices later.
Good luck with that. I mean it.
The ChatAI TAM is now so saturated with free offerings that the first supplier to blink will go out of business before they are done blinking.
I see people (like sibling reply to parent) boasting about the amount of value they get from the $20/m subscription, but I don't see how that is $20 better than just using the free ChatAIs.
The only way out of the red for ChatAI products is to add in advertising slowly; they have to boil the frog. A subscription may have made sense when ChatGPT was the only decent game in town. Subscriptions don't make sense now - I can get 90% of the value of a ChatAI for 0% of the cost.
Where are you moving to?
Best Comment
Yes, where are you moving to from GW? We're looking for options...
> The article claims that AI services are currently over-utilised. Well isn't that because customers are being undercharged for services?
Absolutely, not only are most AI services free but even the paid portion is coming from executives mandating that their employees use AI services. It's a heavily distorted market.
On the flip side, there are many indications that a huge part of corporate AI usage is "Shadow AI", i.e. where workers use GenAI without their employer's blessing or knowledge. E.g. https://cacm.acm.org/news/the-real-significant-threat-of-sha...
And a majority of those workers do not reveal their AI usage, so they either take credit for the faster work or use the extra time for other activities, which further confounds the impact of AI.
This is also distorting the market, but in other ways.
We're talking miraculous level of improvement for a SOA LLM to run on a phone without crushing battery life this decade.
People are missing the forest for the trees here. Being the go to consumer Gen AI is a trillion+ dollar business. How many 10s of billions you waste on building unnecessary data centers is a rounding error. The important number is your odds of becoming that default provider in the minds of consumers.
It’s extremely easy to switch.
I used ChatGPT for every day stuff, but in my experience their responses got worse and I had to wait much longer to get them. I switched to Gemini and their answers were better and were much faster.
I don’t have any loyalty to Gemini though. If it gets slow or another provider gives better answers, I’ll change. They all have the same UI and they all work the same (from a user’s perspective).
There is no moat for consumer genAI. And did I mention I’m not paying for any of it?
It’s like quick commerce, sure it’s easy to get users by offering them something expensive off of VC money. The second they raise prices or offer degraded experience to make the service profitable, the users will leave for another alternative.
> The important number is your odds of becoming that default provider in the minds of consumers.
I haven't seen any evidence that any Gen AI provider will be able to build a moat that allows for this.
Some are better than others at certain things over certain time periods, but they are all relatively interchangeable for most practical uses and the small differences are becoming less pronounced, not more.
I use LLMs fairly frequently now and I just bounce around between them to stay within their free tiers. Short of some actual large breakthrough I never need to commit to one, and I can take advantage of their own massive spends and wait it out a couple of years until I'm running a local model self-hosted with a cloudflare tunnel if I need to access it on my phone.
And yes, most people won't do that, but there will be a lot of opportunity for cheap providers to offer that as a service with some data center spend, but nowhere near the massive amounts OpenAI, Google, Meta, et al are burning now.
The moat will be memory.
As a regular user, it becomes increasingly frustrating to have to remind each new chat “I’m working on this problem and here’s the relevant context”.
GenAI providers will solve this, and it will make the UX much, much smoother. Then they will make it very hard to export that memory/context.
If you’re using a free tier I assume you’re not using reasoning models extensively, so you wouldn’t necessarily see how big of a benefit this could be.
The moat is the chat history and the flywheel of user feedback improving the product.
Given how often smaller LLM companies train on the output of bigger LLM companies, it's not much of a moat.
LLMs complete text. Every query they answer is giving away the secret ingredient in the shape of tokens.
> The important number is your odds of becoming that default provider in the minds of consumers.
Markets that have a default provider are basically outliers (desktop OS, mobile OS, search, social networks, etc).
All other industries don't have a single dominant supplier who is the default provider.
I am skeptical that this market is going to be one of the outliers.
All the other markets with a default provider basically rely on network effects to become and remain the default provider.
There is nothing here (in this market) that relies on network effects.
Just build more tulip fields/railways/websites no matter the cost - the golden age is inevitable! ;-)
In fact it's apparently $5.2 trillion by 2030 [0] (out of $6.7T total data center spend; meaning all of "traditional IT needs" are less than a quarter of the total). That's the total if you add up all of the firms chasing this opportunity.
I do wonder, if you (and the commenter you replied to) think this is a good thing, will you be OK with a data center springing up in your neighbourhood, driving up water or power prices, emitting CO2? Then if SOTA LLMs become efficient enough to run on a smartphone will you be OK with a data center bailout coming from your tax dollars?
[0]: https://www.mckinsey.com/industries/technology-media-and-tel...
How big does an LLM need to be to support natural language queries with RAG?
My hot (maybe just warm these days) take is, the problem with voice assistants on phones is they have to be able to have reasonable responses to a long tail or users will learn not to use them, since the use cases aren’t discoverable and the primarily value is talking to it like a person.
So voice assistants backed by very large LLMs over the network are going to win even if we solve the (substantial) battery usage issue.
Why even bother with the text generation then? You could just make a phone call to an LLM with a TTS frontend. Like with directory enquiries back in the day. Which can be set up as easily as a BBS if you have a home server rack like Jeff Geerling makes youtube videos about.
> Being the go to consumer Gen AI is a trillion+ dollar business.
I use LLM’s all day and a highly doubt this. I’d love to hear your argument for how this plays out.
Yes, over-utilization is a natural response to being undercharged. And being undercharged is a natural result when investors are throwing money at you. During bubbles, Silicon Valley often goes to "lose money, make it up with scale". With the vague idea that after you get to scale, THEN you can figure out how to make money. And fairly consistently, their idea for how to make money is "sell ads".
Past successes like Google encourage hope in this strategy. Sure, it mostly doesn't work. Most of of everything that VCs do doesn't work. Returns follow a power law, and a handful of successes in the tail drive the whole portfolio.
The key problem here doesn't lie in the fact that this strategy is being pursued. The key problem is that it is rare for first mover advantages to last with new technologies. That's why Netscape and Yahoo! aren't among the FAANGs today. The long-term wins go to whoever successfully create a sufficient moat for themselves to protect lasting excess returns. And the capabilities of each generation of AI leapfrogs the last so well that nobody has figured out how to create such a moat.
Today, 3 years after launching the first LLM chatbot, OpenAI is nowhere near as dominant as Netscape was in late 1997, 3 years after launching Netscape Navigator. I see no reason to expect that 30 years from now OpenAI will be any more dominant than Netscape is today.
Right now companies are pouring money into their candidates to win the AI race. But if the history of browsers repeats itself, the company that wins in the long-term would launch in about a year from now, focused on applications on top of AI. And its entrant into the AI wars wouldn't get launched until a decade after that! (Yes, that is the right timeline for the launch of Google, and Google's launch of Chrome.)
Investing in silicon valley is like buying a positive EV lottery ticket. An awful lot of people are going to be reminded the hard way that it is wiser to buy a lot of lottery tickets, than it is to sink a fortune into a single big one.
> Today, 3 years after launching the first LLM chatbot, OpenAI is nowhere near as dominant as Netscape was in late 1997.
Incorrect. There were about 150 millions Internet users in 1998, or 3.5% of total population. The number grew 10 times by 2008 [0]. Netwcape had about 50% of the browser market at the time [1]. In other words, Netscape dominated a small base and couldn’t keep it up.
ChatGPT has about 800 millions monthly users, or already 10% of total current population. Granted, not exclusively. ChatGPT is already a household name. Outside of early internet adopters, very few people knew who Netscape or what Navigator was.
[0] https://archive.globalpolicy.org/component/content/article/1...
[1] https://www.wired.com/1999/06/microsoft-leading-browser-war/...
I was not addressing the size of the market. But the share.
According to https://en.wikipedia.org/wiki/Usage_share_of_web_browsers, Netscape had 60-70% market share. According to https://firstpagesage.com/reports/top-generative-ai-chatbots..., ChatGPT currently has a 60% market share.
But I consider the enterprise market a better indicator of where things are going. As https://menlovc.com/perspective/2025-mid-year-llm-market-upd... shows, OpenAI is one of a pack of significant competitors - and Anthropic is leading the pack.
Furthermore my point that the early market leaders are seldom the lasting winners is something that you can see across a large number of past financial bubbles through history. You'll find the same thing in, for example, trains, automobiles, planes, and semiconductors. The planes example is particularly interesting. Airline companies not only don't have a good competitive moat, but the mechanics of chapter 11 mean that they keep driving each other bankrupt. It is a successful industry, and yet it has destroyed tons of investment capital!
Despite your quibbles over the early browser market, my broader point stands. It's early days. AI companies do not have a competitive moat. And it is way to premature to reliably pick a winner.
Why look at percentage of population?
Netscape in 1997/1998 had about 90% of the target market.
OpenAI today has about 30% of the target market, maybe? (seeing as how every single Windows installation comes with copilot chat already, it's probably less. Every non-tech user I know has already used copilot because it was bundled and Windows prompted them into using it with a popup. Only one of those non-tech users even heard of OpenAI, maybe 50% of them have heard that there are alternatives to Copilot, but they still aren't using those alternatives)
[dead]
Of all the players, I'd argue Google certainly knows how to give away a product for free and still make money.
The local open source argument doesn't hold water for me -- why does anyone buy Windows, Dropbox, etc when there's free alternatives?
> why does anyone buy Windows, Dropbox, etc when there's free alternatives?
No one buys Windows - it comes with the PC.
If people were shipped blank computers and told to order the OS separately, they wouldn't be buying Windows at the current price point.
Not many people buy Windows, they buy laptops that happen to have Windows installed. IMO this is a worthwhile distinction because most people don’t really care about operating systems anyway, and would happily (I suspect, at least) use an Open Source one if it came installed and configured on a device that they got in a store.
Installing an OS is seen as a hard/technical task still. Installing a local program, not so much. I suspect people install LLM programs from app stores without knowing if they are calling out to the internet or running locally.
Yes, it seems to me their strategy is to watch the OpenAI, Anthropics, etc of the world bleed themselves to death.
Besides the fact that this article is obviously AI generated (and not even well, why is there mismatches in british/american english? I can only assume that the few parts in british english are the human author's writing or edits), yes "overutilization" is not a real thing. There is a level of utilization at every price point. If something is "overutilizated" that actually means it's just being offered at a low price, which is good for consumers. It's a nice scare word though and there's endless appetite at the moment for ai-doomer articles.
Author here, I mix up American and British English all the time. It's pretty common for us Brits to do that imo.
See also how all (?) Brits pronounce Gen Z in the American way (ie zee, not zed).
Brit here… I say Gen Zed!
Sorry but it's highly suspect to be spelling the same word multiple different ways across paragraphs. You switch between using centre/center or utilization/utilisation? It is a very weird mistake to make for a human.
I mix British and American English all the time. Subconsciously I type in British English but since I work in American English, my spell checkers are usually configured for en-US and that usually means a weird mix of half and half by the time I've fixed the red squiggles I notice.
I dunno, I switch between grey and gray all the time; comes with having worked in so many different countries.
> why is there mismatches in british/american english
You sometimes see this with real live humans who have lived in multiple counties.
> You sometimes see this with real live humans who have lived in multiple counties.
Also very common with... most Canadians. We officially use an English closer to British English (Zed not zee, honour not honor) however geographically and culturally we're very close to the US.
At school you learn "X, Y, Zed". The toy you buy your toddler is made for the US and Canadian market and sings "X, Y, Zee" as does practically every show on TV. The dictionary says it's spelled "colour" but most of the books you read will spell it "color". Most people we communicate with are either from Canada or the US, so much of our personal communication is with US English.
But also there are a number of British shows that air here, so some particularly British phrases do sneak in to a lot of people's lexicon.
See a similar thing in the way we measure things.
We use celsius for temperature but most of our thermostats default to Fahrenheit and most cookbooks are primarily in imperial measures and units because they're from the US. The store sells everything in grams and kilograms, but most recipes are still in tablespoons/cups/etc.
Most things are sold in metric, but when you buy lumber it's sold in feet, and any construction site is going to be working primarily in feet and inches.
If anything I expect any AI-written content would be more consistent about this than I usually am.
> multiple counties
Pay no attention to those fopheads from Kent. We speak proper British English here in Essex
I do this because I'm a non-native english speaker. My preference varies from word to word. I write color, but i also write aliminium.
One of my least favorite things to come from AI is labelling any writing someone doesn't like as "obviously AI generated". I've read 3 of these kinds of comments on HN just today.
As non native English speaker I mix British and American English all the time, and you should hear me speaking. I mix in some novel accent too. Anyway, the author answered in a sibling reply.
By this logic loss leaders to drive out competition are good gor the consumer, no?
> why is there mismatches in british/american english
Some people are not from usa or England.
To be honest it doesn't feel manually edited.
Bullet points hell, a table that feels it came straight out of grok.
Will the OpenRouter marketplace of M clouds X N models die if the investor money stops? I believe its a free and profitable service, offered completely pay as you go.
> I doubt it.
I don't. This is simply the "drug dealer" model where the first hit is free. They know that once people are addicted, they will keep coming back.
The question of course is, will they keep coming back? I think they very much will. There are indications that GenAI adoption is already increasing labor producitivity labor improvements at a national scale, which is quite astounding for a technology just 3 years old: https://news.ycombinator.com/item?id=46061369
Imagine a magic box where you put in some money and get more productivity back. There is no chance Capitalism (with a capital "C") is going to let such a powerful growth machine wither on the vine. This mad AI rush is all about that.
OpenAI has 800,000,000 weekly users but only 20,000,000 are paying while 780,000,000 are free riding. Should they by accident under provision then they could simply remove the freebee and raise the prices for the paying clients. But that is not what they want.
IMHO the investors are betting on a winner-takes-it-all market and that some magic AGI will be coming out of OpenAI or Anthropic.
The questions are:
How much money can they make by integrating advertising and/or selling user profiles?
What is the model competition going to be?
What is the future AI hardware going to be - TPUs, ASICs?
Will more people have powerful laptops/desktops to run a mid-sized models locally and be happy with it?
The internet didn't stop after the dotcom crash and the AI wont stop either should there be a market correction.
>OpenAI has 800,000,000 weekly users but only 20,000,000 are paying while 780,000,000 are free riding.
By itself, this doesn't tell us much.
The more interesting metric would be token use comparison across free users, paid users, API use, and Azure/Bedrock.
I'm not sure if these numbers are available anywhere. It's very possible B2B use could be a much bigger market than direct B2C (and the free users are currently providing value in terms of training data).
If OpenAi starts charging money for it, I will stop using it..
> How much money can they make by integrating advertising and/or selling user profiles?
I would say if executed well the revenue per user could be at least an order of magnitude more than Google search ads as the ads could be much more convincing and the information density is higher in chat.
800M people are not using chatGPT every week. Be a little less credulous.
I just thought about it, and honestly, from my surroundings of people aged 12-70s, across multiple continents, I can’t think of anyone who isn’t using some sort of LLM once a week.
American software engineer here, and I have only used LLM tools ~3 dozen times: tried at work a few times, and been unimpressed/actively frustrated; a few times to ask questions I would be normally survey blog posts about (social/lifestyle things); and the rest has been generating images with my kids.
But currently, aside from generating "creative media", I'd say I'm pretty much opposed to LLM tools. They have yet to demonstrate any value to me at work or with respect to the areas of research I am interested in, and given the kind of statistical mechanism that they are, I do not believe they are capable of doing so.
> aside from generating "creative media"
Interesting take, because I'm the opposite of it. My biggest use case is getting into a completely new topic, as it's the most frictionless starting point for most of the queries. Then I look around based on the rudimentary knowledge that I can gather from LLMs. However, I'm completely opposed to any sort of creative media created by LLMs and try to avoid it as much as I can (music, images, and etc.).
Also, it has become the natural workflow for me to throw bunch secondary priority work stuff to Claude and let it do its things, while I focus on the important stuff.
My point is, everyone finds a way to use it. Some are opposed to specific things, others are using other parts.
These takes feel like a failure compared to my daily usage, which is literally non-stop for ten hours a day. Want to construct a niche jq, curl, or find linux command? Can't remember the parameters for a function? Don't want to leave your terminal to search for something? ctrl+I and type in readable english.
I wonder if this is counting the "AI Summary" that is gratuitously included in a standard web search now?
That's the only AI I use anywhere near weekly. I have tried claude a few times, it was useless at helping me with my questions. I haven't really been back.
The thing that makes AI investment hard to reason about for individuals is that our expectations are mostly driven by a single person’s usage, just like many of the numbers reported in the article.
But the AI providers are betting, correctly in my opinion, that many companies will find uses for LLM’s which are in the trillions of tokens per day.
Think less of “a bunch of people want to get recipe ideas.”
Think more of “a pharma lab wants to explore all possible interactions for a particular drug” or “an airline wants its front-line customer service fully managed by LLM.”
It’s unusual that individuals and industry get access to basically similar tools at the same time, but we should think of tools like ChatGPT and similar as “foot in the door” products which create appetite and room to explore exponentially larger token use in industry.
> Think more of “a pharma lab wants to explore all possible interactions for a particular drug”
Pharma does not trust OpenAI with their data, and they don't work on tokens for any of the protein or chemical modeling.
There will undoubtedly be tons of deep nets used by pharma, with many $1-10k buys replacing more expensive physical assays, but it won't be through OpenAI, and it won't be as big as a consumer business.
Of course there may be other new markets opened up but current pharma is not big enough to move the needle in a major way for a company with an OpenAI valuation.
My claim is that there will exist some company which pharma is willing to trust for AI research…they presumably trust Microsoft with their email today.
But my bigger claim is that ~half the Fortune 500 will be able to profitably deploy AI with spends in the tens or hundreds of millions per year quite soon. Not that pharma itself is a major contributor to that effect.
But for 'AI' to be a winner-take-all market, it seems that the winner would have to be using customer data to improve the 'AI'. Not only do you have to believe that one of these (relatively) under-capitalized upstarts can corral the money, but also that they can convince (enterprise) customers to 'fork over' their proprietary data to only one provider, and also that the provider can then charge a monopoly rent.
Those all seem possible, but I wouldn't assign greater than a 50% probability to any of them, and the valuations seem to imply near-certainty.
When I'm building out a new feature, I can churn through millions of tokens in Claude code. And that's just me... Now think about Claude code but integrated with Excel or datadog, or whatever app could be improved through LLM integration. Think about the millions of office workers, beyond just software engineers, who will be running hundreds of thousands or millions of tokens per day through these tools.
Let's estimate 200 million office workers globally as TAM running an average of 250k tokens. That's 50 trillion tokens DAILY. Not sure what model provider profit per token is, but let's say it's .001 cents.
Thats $500M per day in profit.
I’m with you on the Claude Code example —- it matches my experience.
But I do think the important thing to look forward to is AI work which is totally detached from human intervention.
I find it absurd to pay for tokens I cant control, predict or even check in any reasonable way. It is literally amounts to "pay whatever random money the company asks you to pay" kind of contract.
I pay $100/mo for CC and have functionally unlimited tokens.
I find it irreplaceable.
Currently there is no profit per token, quite a bit of loss per token, that's the problem. Your not going to make it up in volume.
Do you have a source for that? I'm especially interested in a source for Anthropic.
https://www.wsj.com/tech/ai/openai-anthropic-profitability-e...
Anthropic expects to break even in 2028. They’re all unprofitable now.
paywalled.
Are they unprofitable because they don't profit on inference, or because they reinvest all of the profit into scaling up?
Remember how long Amazon was unprofitable, by choice.
> Are they unprofitable because they don't profit on inference, or because they reinvest all of the profit into scaling up?
They are scaling up using VC money, not revenue. As far as profit on inference goes, it's hard to separate it out from training: they cannot, at any given time, simply stop training because that would kill any advantage they have 6 months down the line.
For all practical purposes, you can't look at their inference costs independent of the training cost; they need to keep spending on both if they want to continue doing inference.
> Remember how long Amazon was unprofitable, by choice.
That was a very different scenario - AMZ was not spending their revenue on land-grabbing, they were spending their revenue on long-lived infra, while AI companies now are spending VC investment, not revenue, on land-grabbing.
The difference between spending your revenue on short-lived infra (training a new model, acquiring GPUs) and long-lived infra is that with long-lived infra, at any time, even after 10+ years, you can stop expanding your infra and keep the resulting revenue as profit.
With short-lived infra (models, GPUs), you can't simply stop infra spending and collect profit from the revenue, because the infra reached end-of-life and needs to be replaced anyway.
> “a pharma lab wants to explore all possible interactions for a particular drug”
How would an LLM be any good at this?
> “an airline wants its front-line customer service fully managed by LLM.”
This has been experimented on before by many companies over the recent years, most notably Klarna which was among the earliest guinea pigs for it and had to later on backtrack on this "novel" idea when the results came out.
Does neither mean it can't work nor that it can't work with LLMs. Just that hacking together some RAG chatbot is probably not it.
But if I'm a pharma lab, I don't want to rely on a statistical engine that makes mistakes to answer those questions, I want to query a database that is deterministic.
Sam Altman says that he thinks scientific research is a huge opportunity for AI to contribute as it more fully develops and I think he’s right.
Since I’m not a scientific researcher, I have no idea if he’s just blowing smoke but I think it’s reasonable to think of a purpose-built system which has an LLM component being used by a team to do something useful.
This. LLM is NOT the tool for a pharma lab - properly trained ML is the right tool. Heck, English is probably not even the right LANGUAGE to use for discussing chemical interactions.
Would the proper language be math, or another human language?
I could see things like "nitrate" and "nitrite" possibly being a stumbling block for an LLM.
OpenAI has first mover advantage with respect to raising money but not wrt tech dominance.
Consensus theory: If AGI then superintelligence.
AI CapEx plans are not ROI based. Rather, they are the cost of "how do I remain competitive in the race to attain AGI" coupled with conveniently deep pockets. The money is being spent because the spenders can afford it and they see it as an existential risk as much as a profit opportunity.
Maybe OP's conclusion about the headline question blunts some political opposition to data centers, but that's not the salient issue.
The issue is this: America is betting a meaningful chunk of GDP that AGI is possible. This is The Manhattan Project 2.0.
After reading the article and the comments, here are a few points people are missing from their analysis:
- OverUtilized/UnderCharged: doesn't matter because...
- Lead Time vs. TCO vs. IRS Asset Deprecation: The moment you get it fully built, it's already obsolete. Thus from a CapEx point of view, if you can lease your compute (including GPU) and optimize the rest of the inputs for similar then your CapEx overall is much lower and tied to the real estate - not the technology. The rest is cost of doing business and deductible in and of itself.
- The "X" factor: Someone mentioned TPU/ASIC but then there is the DeepSeek factor - what if we figure out a better way of doing the work that can shortcut the workflow?
- AGI partnerships: Right now, you see a lot of Mega X giving billions to Mega Y because all of them are trying to get their version of Linux or Apache or whatever at parity with the rest. Once AGI is settled and confirmed, then most all of these partnerships will be severed because it then becomes which company is going to get their AI model into that high prestige Montessori school and into the right ivy league schools - like any other rich parent would for their "bot" offspring.
So what will it look like when it crashes? A bunch of bland empty "warehouses" with mobile PDU's once filling all their parking lot space gone. Whatever "paradise" that was there may come back... once you bulldoze all that concrete and steel. The money will do something else like a Don McLean song.
>once you bulldoze all that concrete and steel
You're not quite thinking things through there man. Once the elites who built these follies have gone, the mob will go shopping for building materials. I wouldn't be surprised even if people end up living in these datacentres once they become derelict. They have AC after all.
Yes or no conclusions aside (and despite its title, the article deserves better than that), the key point is I think this one: “But unlike telecoms, that overcapacity would likely get absorbed.”
Telecom (dark fiber) capacity got absorbed too. Eventually. After a ton of bankruptcies.
Stylistically, this smells like it was copy and pasted from straight out Deep Research. Substantively, I could use additional emphasis on the mismatch between expectations and reality with regards to telco debt-repayment schedule.
Some of the utilization comparisons are interesting, but the article says 2 trillion was spent on laying fiber but that seems suspicious.
There's an enormous amount of unused, abandoned fiber. All sorts of fiber was run to last mile locations, across most cities in the US, and a shocking amount effectively got abandoned in the frenzy of mergers and acquisitions. 2 trillion seems like a reasonable estimate.
Giant telecoms bought big regional telecoms which came about from local telecoms merging and acquiring other local telecoms. A whole bunch of them were construction companies that rode the wave, put in resources to run dark fiber all over the place. Local energy companies and the like sometimes participated.
There were no standard ways of documenting runs, and it was beneficial to keep things relatively secret, since if you could provide fiber capabilities in a key region, but your competition was rolling out DSL and investing lots of money, you could pounce and make them waste resources, and so on. This led to enormous waste and fraud, and we're now on the outer edge of usability for most of the fiber that was laid - 29-30 years after it was run, most of it will never be used, or ever have been used.
The 90s and early 2000's were nuts.
I so desperately wish it weren't abandoned. I hate that it's almost 2026 and I still can't get a fiber connection to my apartment in a dense part of San Diego. I've moved several times throughout the years and it has never been an option despite the fact that it always seems to be "in the neighborhood".
That has nothing to do with fiber, it’s all about politics and a regulatory environment where nobody is incented to act. Basically, the states can’t fully regulate internet and the Federal government only wants to fund buildouts on a pork barrel basis. Most recently rural.
At the local level, there is generally a cable provider with existing rights of way. To get a fiber provider, there’s 4 possible outcomes: universal service with subsidy (funded by direct subsidy), cherry-picked service (they install where convenient), universal service (capitalized by the telco) and “fuck you”, where they refuse to operate. (ie. Verizon in urban areas)
The private capitalized card was played out by cable operators in the 80s (they were innovators then, and AT&T was just broken up and in chaos). They have franchise agreements whose exclusivity was used as loan collateral.
Forget about San Diego, there are neighborhoods in Manhattan with the highest population density in the country where Verizon claims it’s unprofitable to operate.
I served on a city commission where the mayor and county were very interested in getting our city wired, especially as legacy telco services are on the way out and cable costs are escalating and will accelerate as the merger agreement that formed Spectrum expires. The idea was to capitalize last mile with public funds and create an authority that operated both the urban network and the rural broadband in the county funded by the Federal legislation. With the capital raised with grants and low cost bonding (public authority bonds are cheap and backed by revenue and other assets), it would raise a moderate amount of income in <10 years.
We had the ability to get the financing in place, but we would have needed legislation passed to get access to rights of way. Utilities have lots of ancient rights and laws that make disruption difficult. The politicians behind it turned over before that could be changed.
The worst part is it'd probably cost less than $100 of fiber and labor to splice something into your building, maybe $200-400 of gear to light it up, and you'd have a 10 gbps pipe back to some colo. It's more economical to run new fiber in most places these days, even if the local ISP knows exactly where all the old abandoned legacy lines are run, because of subsidization and basically scamming. The big companies like Lumen keep their knowledge regionally compartmented, legally shielded, and deliberately obfuscated, because if it became known that existing fiber was already run to a place they claim they can't serve, they can't get access to yet more funding for their eternal "service for the underserved" government money grift.
I stumbled on old maps that showed a complete coverage of fiber in my municipality, paperwork from a company that was acquired, and which in turn merged, then was bought out by one of the big 5 ISPs. When local officials requested information regarding existing fiber, this ISP refused and said any such information was proprietary. They later bid on and won contracts to run new fiber (parallel to existing lines which they owned, which still had more than a decade of service life left in them at that point).
I estimated that only around 10-15% of the funding went toward actual labor and materials, the remainder was pure profit. The local government considered it a major victory, money well spent.
For infrastructure, central planning and state-run systems make a lot of sense - this after all is how the USA's interstate highway system was built. The important caveat is that system components and necessary tools should be provided by the competitive private sector through transparent bidding processes - eg, you don't have state-run factories for making switches, fiber cable, road graders, steel rebar, etc. There are all kinds of debatable issues, eg should system maintenance be contracted out to specialized providers, or kept in-house, etc.
The GDP 1995-2000 (inclusive) was about $52T. So that assertion would mean that about %3.8 of the US' economic activity was laying fiber. That seems like a lot, but in my ignorance doesn't sound totally impossible.
Don’t think looking at power consumption of b200s is a good measure of anything. Could well be an indication of higher density rather than hitting limits and cranking voltage to compensate
Yes, one of NVidia's selling points for the b200 is that performance per watt is better than before. High power consumption without controlling for performance means nothing.
In the telecoms case, just trowing in extra dark fiber just in case was an almost negligeable cost in light of digging new connections. So who cares if in year 1 you have just 4 fibers lit of the 244 you trew into the trench?
This is not the case for AI data centers at all. The compute is a major cost in the whole build budget. Having that installed and not used is financial ruin.
> Having that installed and not used is financial ruin.
And having it used consumes so much energy that it's everyone else's ruin.
I am still a little skeptical about utilisation rates. If demand is so extreme, wouldn't we see rental prices for H100/A100 prices go up or maintain? Wouldn't the cost for such a gpu still be high (you can get em 3k used).
On "runpod community cloud" renting a 5090 costs $0.69/hour [1] and it consumes about $0.10/hour electricity, if running at full power and paying $0.20/kWh.
On Amazon, buying a 5090 costs $3000 [2]
That's a payback time of 212 days. And Runpod is one of the cheaper cloud providers; for the GPUs I compared, EC2 was twice the price for an on-demand instance.
Rental prices for GPUs are pretty darn high.
[1] https://www.runpod.io/pricing [2] https://www.amazon.com/GIGABYTE-Graphics-WINDFORCE-GV-N5090G...
A 5090 gaming card is a different beast to the 80gb ai cards. That one was 40k usd so for renting that to hit 1.50 dollar per hour is interesting.
Is there a way in which this is good for a segment of consumers? When the current gen of GPUs are too old, will the market be flooded with cheap GPUs that benefit researchers and hobbyists who therwis would not afford them?
GPUs age surprisingly gracefully. If a GPU isn't cutting edge, you just tie two or more of them together for a bit more power consumption to get more or less the same result as the next generation GPU.
if there's ever a glut in GPUs that formula might change but it sure hasn't happened yet. Also, people deeply underestimate how long it would take a competing technology to displace them. It took GPUs nearly a decade and the fortunate occurrence of the AI boom to displace CPUs in the first place despite bountiful evidence in HPC that they were already a big deal.
>GPUs age surprisingly gracefully. source: it was revealed to me in a dream
source: Nvidia Shields still shipping with Maxwell GPUs from 2014 and many university labs still running on Pascal GPUs from 2016. But stupid facts don't matter, this one hit you wrong in the feels, amIRight?
Unlikely, for a few reasons:
* The GPUs in use in data centers typically aren’t built for consumer workloads, power systems, or enclosures.
* Data Centers often shred their hardware for security purposes, to ensure any residual data is definitively destroyed
* Tax incentives and corporate structures make it cheaper/more profitable to write-off the kit entirely via disposal than attempt to sell it after the fact or run it at a discount to recoup some costs
* The Hyperscalers will have use for the kit inside even if AI goes bust, especially the CPUs, memory, and storage for added capacity
That’s my read, anyway. They learned a lot from the telecoms crash and adjusted business models accordingly to protect themselves in the event of a bubble crash.
We will not benefit from this failure, but they will benefit regardless of its success.
Some of them will probably be starving, homeless, or bedridden by the time that happens but yes they can get cheap GPUs
Many researchers and hobbyists cannot even plug in a 10 KW 8 GPU DGX server.
Why not? It’s 40A at 240V, or 25% of the continuous load rating of a 200A 240V single-phase service.
If someone can afford an 8 GPU server, they should be able to afford some #6 wire, a 50A 2P breaker, and a 50A receptacle. It has the same exact power requirements as an L2 EV charger.
You probably run a bigger (perhaps double) neutral and care about a stronger ground. But yeah, the $12 is rounding error at this scale
In the US, you don’t need a neutral for a 240V 50A circuit on a residential single-phase service, there are two line conductors (120V to ground) both connected to a 2-pole single-phase breaker, line-to-line between the two is 240V.
You would need a neutral if it was a 208/120V three-phase service.
Neutrals and grounds are sized per the NEC, neutrals are the same size as the line conductors and equipment grounds are sized off of a table.
#6 conductors and #10 ground is what the NEC calls for.
I just haven't seen server rooms that don't demand a doubly sized neutral when one is required.
I live the NYC 208V life doing mostly resi, though.
Quick search of the spec for that is 6 power supplies, 2 of which are redundant. Looks to use a neutral to me. Says it uses C19/C20 connectors
edit: wait most ranges use 14-50R outlets and need a neutral ran. I am calling your statement into question. Surely harmonics and 120V internal draws cause non-zero neutral current. And I'm sure GPUs have harmonics being semiconductor flavored.
Turns out you are correct on both points!
My bad, NEMA 14-50R is a 4-wire receptacle with a neutral.
I learned something new today, 200% neutrals are not required by the NEC but can help with non-linear loads, and certain transformers that mitigate harmonics need 200% neutrals.
I'm thinking you'd have a hard time fitting it in a smaller residential unit purely because of the load calculations. You'd probably have to add 3000W worth of AC units or heat pumps. Considering it's 4+ space heaters worth of output, I think you could fit it on a 200A service. 150A if you're bold and use optional calcs (these assuming you have electric cooking). To your point, basically the same as if you have an EV but with some cooling in there.
That doesn't exactly bode well for the EV revolution, then, does it?
The average commute in the United States is about 24 miles a day round trip. That's about 10 kWH. That's enough to charge overnight on a 15A circuit.
You can drive around with 24 miles of charge, but ... I'll pass, thanks.
In reality, if you have a dryer outlet, you have a good fraction of 10 kW available.
I just put in a dedicated 50 amp circuit myself and my car charges from ~25% to full in about 6-7 hours. But I wanted to present the lazy worst case scenario. The warrior's FUD here is that there isn't enough easily available lithium for everyone in just California alone to have one.
Almost all home EV charging is <=10kW.
wut?
The 2001 telecoms crash drove benefits for companies that came later in the availability of inexpensive dark fiber after the bubble popped. WorldCom, ICG, Williams sold off to Verizon, Level 3, Teleglobe, and others. That in turn helped future Internet companies gain access to plentiful and inexpensive bandwidth. Cable telephony companies such as Cablevision Systems, Comcast, Cox Communications, and Time Warner, used the existing coaxial connections into the home to launch voice services.
Rail and fiber deprecates on multiple decade timescales. AI data centers close to tulips. Even assuming we manage to make data center stretch to 10 years, these assets won't be around long enough to support ecosystem of new companies if the economics stops making sense. Ultimately the only durable thing is any power infra that gets built, vs rail and fiber where inheritance isn't just rail networks or fiber but like 1000s of kilometers of earthwork projects to build out massive physical networks.
Data centers last decades. Many or the current AI hosting vendors such as Coreweave have crypto origin. Their data centers were built out in 2010s, early 2020s.
Many of legacy systems still running today are IBM or Solaris servers at 20, 30 year old. No reason to believe GPU won’t still be in use in some capacity (e.g. interference) a decade from now.
Skeletons of data centers and some components (i.e. cooling) have long shelf life, but they're also ~10% of investment. Plurality of fiber and rail went towards building out linear infrastructure where improvements can be milked at nodes to improve network efficiency (better switches etc).
VS plurality of AI investment, i.e. trillions are going towards fast deprecating components where we can say with relative confidence will likely be net negative stranded assets in terms of amoritization costs if current semi manufacturing trends continues.
Keeping some mission critical legacy systems around is different than having trillions that makes no financial sense to keep on the books, i.e. post bubble new gen hardware will likely not have scarcity pricing or better compute efficiency (better apex and opex), there is no reason to believe companies will legacy GPUs around at scale if every rack loses them money relative to new hardware. And depending on actual commercialization compute demand, it can simply make more economic sense to retire them than keep them going.
Used to last decades, the world didn't move at this speed before.
Is this a question of the GPU chips dying due to be being warm semiconductors, or of them becoming outdated in relation to new chips?
Both. Semi vs concrete, depreciating vs durable assets. Durable linear assets you upgrade switches to improve fiber/rail which is where most of the investment was in. GPUs you replace the racks which is where most of the investment is in. Either way it cannot be stretched the same way materially and most important economically, i.e. new chips with better power efficiency = running old chips is literally losing money squatting on an data center slot. There is very little reason to believe new chips will cost more than legacy (current chips) for the simple reason much of currnet chips are acquired on scarcity pricing, i.e. Nvidia margins went from 50% to 70%. That 20% is massive capex premium that is not going to be competitive if bubble pops and Nvidia has to sell new hardware on commodity pricing, new hardware that is in all likelihood also going to be more compute efficient in terms of power (opex). Even if you stretch existing compute past 3-5 to 10 years it is still closer to tuplids than rail or fiber in terms of economically productive timescale.
TLDR old durable infra tends to retains positive residual value because they're not easy to replace economically/frequently, old compute has negative residual value because they are easy to replace economically/frequently.
This is indeed true, but doesn't fiber have a far longer lifetime than GPU heavy data centers? The major cost center is the hardware, which has a fairly short shelf life.
Well you still get the establishment of 1) large industrial buildings 2) water/electricity distribution 3) trained employees who know how to manage a data center
Even if all of the GPUs inside burn out and you want to put something else entirely inside of the building, that's all still ready to go.
Although there is the possibility they all become dilapidated buildings, like abandoned factories
The building and electrical infrastructure are far cheaper than the hardware. So much so that the electricity is a small cost of the data center build out, but a major cost for the grid.
Of the most valuable part is quickly depreciating and goes unused within the first few years, it won't have a chance for long term value like fiber. If data centers become, I don't know, battery grid storage, it will be very very expensive grid storage.
Which is to say that while there was an early salivation for fiber that was eventually useful, overallocation of capital to GPUs goes to pure waste.
I'm sure there are other "emerging" markets that could make use of the GPUs, I heard game streaming is relatively popular so you can play PC games on your phone for example. I'd guess things similar to that would benefit from a ton of spare GPUs and become significantly more viable.
>The building and electrical infrastructure are far cheaper than the hardware.
Maybe it's cheaper if we measure by dollars or something, but at the same time we lack the political will to actually do it without something like AI on the horizon.
For example, many data center operators are pushing for nuclear power: https://www.ehn.org/why-microsoft-s-move-to-reopen-three-mil...
That's one example among many.
So I'm hesitant to believe that "electricity is a small cost" of the whole thing, when they are pushing for something as controversial as nuclear.
Also the 2 are not mutually exclusive. Chip fabs are energy intensive. https://www.tomshardware.com/tech-industry/semiconductors/ts...
Nuclear is not very controversial, there are tons of places that would be very happy to have additional reactors, namely those with successful reactors right now. It's just super expensive to build and usually a financial boondoggle.
AI companies are saying they are trying to build nuclear because it makes them sound serious. But they are not going to build nuclear, solar and storage is cheaper more flexible and faster to build. The only real nuclear commitment is Microsoft reopening an old nuclear reactor that had become uneconomic to operate. Building anything new would be a five+ year endeavor, if we were in a place with high construction productivity like China. In the US, new nuclear is 10 years away.
But as soon as Microsoft restarted an old reactor, all their competitors felt like they had to sound as serious, so they did showy things that won't result in solving their immediate needs. Everybody's renewable commitments dwarf their nuclear commitments.
AI companies can flaunt expensive electricity at high cost for high investor impact precisely because electricity is a small cost component of their inputs. It's a hugely necessary input, and the limiting factor for most of their plans, but the dollar amount for the electricity is small. The current valuations of AI assume that a kWh put towards AI will generate far far more value than the average kWh on the grid.
Welp Gemini got me. Using G3 to improve what I write, generate specific images, and use NotebookLM to dive into some research materials. Tried to do a bit each day with my free credits, but hit the limit too often. G2.5 was not nearly as useful. So I upgraded my baselevel Google workspace plan. Recently spoke to someone who is also using G3 a lot with good results. YMMV re: G3, but Google hooked me, and now I pay more. However I think it is worth it for what I do. G3 is my helpful, nerdy work mate. I never plan to use agenic AI. Not using ChatGPT much if any at all anymore. Sorry Sam.
For those who can’t/won’t read the article - the author argues that the answer is “no - it’s not a repeat”.
> Model efficiency could improve faster than expected.
There are compounding incentives for this. I see this as the most likely outcome, though it will likely be stepwise rather than gradual.
I'm curious about the telecom collapse. How was it not in the pipeline that much better use of the fibres was around the corner? Surely they would have known that people were looking into it with some promise?
It's human nature not to admit the obvious until it is way way way too late.
Simultaneous claims that 'agentic' models are dramatically less efficient, but also forecasts efficiency improvements? We're in full-on tea-leaves-reading mode.
Yes
No, because at least dark fiber is useful. AI GPUs will be shipped off to developing nations to be dissolved for rare earth metals once the third act of this clown show is over.
Amazing article, I found it fascinating.
> You can already use Claude Code for non engineering tasks in professional services and get very impressive results without any industry specific modifications
After clicking on the link, and finding that Claude Code failed to accurately answer the single example tax question given, very impressive results! After all, why pay a professional to get something right when you can use Claude Code to get it wrong?
This seems to be either LLM AI slop or a person working very hard to imitate LLM writing style:
The key dynamic: X were Y while A was merely B. While C needed to be built, there was enormous overbuilding that D ...
Why Forecasting Is Nearly Impossible
Here's where I think the comparison to telecoms becomes both interesting and concerning.
[lists exactly three difficulties with forecasting, the first two of which consist of exactly three bullet points]
...
What About a Short-Term Correction?
Could there still be a short-term crash? Absolutely.
Scenarios that could trigger a correction:
1. Agent adoption hits a wall ...
[continues to list exactly three "scenarios"]
The Key Difference From S:
Even if there's a correction, the underlying dynamics are different. E did F, then watched G. The result: H.
If we do I and only get J, that's not K - that's just L.
A correction might mean M, N, and O as P. But that's fundamentally different from Q while R. ...
The key insight people miss ...
If it's not AI slop, it's a human who doesn't know what they're talking about: "enormous strides were made on the optical transceivers, allowing the same fibre to carry 100,000x more traffic over the following decade. Just one example is WDM multiplexing..." when in fact wavelength division multiplexing multiplexing is the entirety of those enormous strides.
Although it constantly uses the "rule of three" and the "negative parallelisms" I've quoted above, it completely avoids most of the overused AI words (other than "key", which occurs six times in only 2257 words, all six times as adjectival puffery), and it substitutes single hyphens for em dashes even when em dashes were obviously meant (in 20 separate places—more often than even I use em dashes), so I think it's been run through a simple filter to conceal its origin.
Remember we have about 20 years of poorly written articles along with a few well written ones for the LLM to be trained on. I'm confident that attempting to tell LLM from human writing is a waste of time now that the year is almost over.
Other than that I'd rather choose a comprehensive article than a summary.
Attempting to tell LLM writing from human writing is usually a waste of time, but apparently not in this case.
Did you see the reply from the author to similar statements?
You mean https://news.ycombinator.com/item?id=46138663? He said that he mixed up British and American English all the time, but didn't try to deny that the article was AI slop.
I agree, and it feels like an allergy by now to that style specifically. This is doubly annoying because it ruins the reading experience and just makes me question myself constantly because you often can't be quite certain especially for shorter posts/comments.
On topic: It is always quite easy to be the cynical skeptic, but a better question in my view: Is the current AI boom closer to telecoms in 2000 or to video hosting in 2005? Because parallels are strong to both, and the outcomes vastly different (Cisco barely recovered by now compared to 1999 while youtube is printing money).
You guys make me laugh.
> This is the opposite of what happened in telecoms. We're not seeing exponential efficiency gains that make existing infrastructure obsolete. Instead, we're seeing semiconductor physics hitting fundamental limits.
What about the possibility of improvements in training and inference algorithms? Or do we know we won't get any better than grad descent/hessians/etc ?
Nice article; far from bullet-proof, but it brings up some interesting points. HN comments are vicious on the topic of AI non-bubbles.
No, because the datacenters will get used. The demand side exists, whether it’s LLM AIs or something completely different that isn’t AI related. That’s very different from a crash where there is absolutely nothing valuable/useable/demanded underneath the bubble.
> The demand side exists, whether it’s LLM AIs or something completely different that isn’t AI related.
What do you think LLM tuned GPUs or TPUs are going to be used for that is completely different and not AI related?
‘LLM tuned GPUs’ are just GPUs. The tuning refers to the models and how they use something like CUDA or whatever. There was a GPU shortage even before LLMs properly burst onto the scene, back with crypto mining. Now, it’s possible TPUs might add a wrinkle to later demand side issues when there is a crash but that will depend on how useful TPUs actually end up being to those outside Google. But GPUs will remain useful, whether it’s for gaming, machine learning (not the AI slop variety of this, but more categorising like for self-driving cars or medical imaging etc), or for the next crypto scam. Surely you agree that powerful computing capacity, independent of AI scams, is here to stay, right?
My main point in arguing that now isn’t like 2000 is that unlike in 2000 we have actual hardware and physical assets underpinning this bubble. In 2000 the assets were literally just imaginary. Yes there is speculation now but it is underpinned by silicon that will still be worth decent money even after LLMs are exposed as a hallucinatory mirage.
I'm not convinced building nuclear plants to generate fake crap faster is here to stay, it better not or we're toast.
But surely you agree those plants would get used either way?
Telco capex was $100 billion at the peak of the IT bubble, give or take. There's going to be $400 billion investments in AI in 2025. Just saying.
No. AI data center, or any data center is designed with incorrect data structure resulting in over utilisation of computing resource.
Hardware growth is slow and predictable, but one breakthrough algorithm completely undercuts any finance hypothesis premised on compute not flowing out of the cloud and back to the edges and into the phones.
This is a kind of risk that finance people are completely blind to. Open AI won't tell them because it keeps capital cheap. Startups that must take a chance on hardware capability remaining centralized won't even bother analyzing the possibility. With so many actors incentivized to not know or not bother asking the question, there's the biggest systematic risk.
The real whiplash will come from extrapolation. If an algorithm advance shows up promising to halve hardware requirements, finance heads will reason that we haven't hit the floor yet. A lot of capital will eventually re-deploy, but in the meantime, a great deal of it will slow down, stop, or reverse gears and get un-deployed.
AI had a kind of Jevons paradox approach to efficiency improvements, unfortunately - if you halve the compute requirements with an algorithmic advance, you can run a model twice as big.
The large SOTA models have hit very diminishing returns on further scaling, I think. So you’d rather double the number of models you can run in parallel.
Confusing telephones with AGI is some cope from humans.
[dead]
Holly cow, we've found an exception to Betteridge's Law of Headlines! Talk about burying the lede!
If you read the article, then this is not an exception to the law
If you read the article _and_ agree with the author's conclusions. I did the former, but not the later.