11 ms·

I think two things can be true simultaneously: 1. LLMs are a new technology and it's hard to put the genie back in the bottle with that. It's difficult to imag

by lsy 1y ago

I think two things can be true simultaneously: 1. LLMs are a new technology and it's hard to put the genie back in the bottle with that. It's difficult to imagine a future where they don't continue to exist in some form, with all the timesaving benefits and social issues that come with them. 2. Almost three years in, companies investing in LLMs have not yet discovered a business model that justifies the massive expenditure of training and hosting them, the majority of consumer usage is at the free tier, the industry is seeing the first signs of pulling back investments, and model capabilities are plateauing at a level where most people agree that the output is trite and unpleasant to consume. There are many technologies that have seemed inevitable and seen retreats under the lack of commensurate business return (the supersonic jetliner), and several that seemed poised to displace both old tech and labor but have settled into specific use cases (the microwave oven). Given the lack of a sufficiently profitable business model, it feels as likely as not that LLMs settle somewhere a little less remarkable, and hopefully less annoying, than today's almost universally disliked attempts to cram it everywhere.

xnx 1y ago
> (the supersonic jetliner) ... (the microwave oven) But have we ever had a general purpose technology (steam engine, electricity) that failed to change society?
- blueflow 1y ago
  It wouldn't be general purpose if it fails to bring change. I'd take every previous iteration of "AI" as example, IBM Watson, that stuff
fendy3002 1y ago
LLMs need significant optimization or we get significant improvement on computing power while keeping the energy cost the same. It's similar with smartphone, when at the start it's not feasible because of computing power, and now we have one that can rival 2000s notebooks. LLMs is too trivial to be expensive EDIT: I presented the statement wrongly. What I mean is the use case for LLM are trivial things, it shouldn't be expensive to operate
- trashchomper 1y ago
  Calling LLMs trivial is a new one. Yea just consume all of the information on the internet and encode it into a statistical model, trivial, child could do it /s
  - hammyhavoc 1y ago
    > all of the information on the internet Total exaggeration—especially given Cloudflare providing free tools to block AI and now tools to charge bots for access to information.
  - fendy3002 1y ago
    well I presented the statement wrongly. What I mean is the use case for LLM are trivial things, it shouldn't be expensive to operate
- lblume 1y ago
  Imagine telling a person from five years ago that the programs that would basically solve NLP, perform better than experts at many tasks and are hard not to anthropomorphize accidentally are actually "trivial". Good luck with that.
  - jrflowers 1y ago
    >programs that would basically solve NLP There is a load-bearing “basically” in this statement about the chat bots that just told me that the number of dogs granted forklift certification in 2023 is 8,472.
    - lblume 1y ago
      Sure, maybe solving NLP is too great a claim to make. It is still not at all ordinary that beforehand we could not solve referential questions algorithmically, that we could not extract information from plain text into custom schemas of structured data, and context-aware mechanical translation was really unheard of. Nowadays LLMs can do most of these tasks better than most humans in most scenarios. Many NLP questions at least I find interesting reduce to questions of the explanability of LLMs.
eric-burel 1y ago
Developers haven't even started extracting the value of LLMs with agent architectures yet. Using an LLM UI like open ai is like we just figured fire and you use it to warm you hands (still impressive when you think about it, but not worth the burns), while LLM development is about building car engines (here is you return on investment).
- clarinificator 1y ago
  Every booster argument is like this one. $trite_analogy triumphant smile
  - Karrot_Kream 1y ago
    [flagged]
    - rrr_oh_man 1y ago
      Throwing genetic fallacies around instead of engaging with the comment at hand... :)
- Jensson 1y ago
  > Developers haven't even started extracting the value of LLMs with agent architectures yet There are thousands of startups doing exactly that right now, why do you think this will work when all evidence points towards it not working? Or why else would it not already have revolutionized everything a year or two ago when everyone started doing this?
  - eric-burel 1y ago
    Most of them are a bunch of prompts and don't even have actual developers. For the good reason that there is no training system yet and the wording of how you call the people that build these system isn't even there or clearly defined. Local companies haven't even setup a proper internal LLM or at least a contract with a provider. I am in France so probably lagging behind USA a bit especially NY/SF but the word "LLM developer" is just arriving now and mostly under the pressure of isolated developers and companies like me. This feel really really early stage.
Msurrow 1y ago
> first signs of pulling back investments I agree with you, but I’m curious; do you have link to one or two concrete examples of companies pulling back investments, or rolling back an AI push? (Yes it’s just to fuel my confirmation bias, but it’s still feels nice:-) )
- 0xAFFFF 1y ago
  Most prominent example was this one: https://www.reuters.com/technology/microsoft-pulls-back-more-data-center-leases-us-europe-analysts-say-2025-03-26/ https://www.reuters.com/technology/microsoft-pulls-back-more...
  - durumu 1y ago
    I think that's more reflective of the deteriorating relationship between OpenAI and Microsoft than an true lack of demand for datacenters. If a major model provider (OpenAI, Anthropic, Google, xAI) were to see a dip in available funding or stop focusing on training more powerful models, that would convince me we may be in a bubble about to pop, but there are no signs of that as far as I can see.
moffkalast 1y ago
ML models have the good property of only requiring investment once and can then be used till the end of history or until something better replaces them. Granted the initial investment is immense, and the results are not guaranteed which makes it risky, but it's like building a dam or a bridge. Being in the age where bridge technology evolves massively on a weekly basis is a recipe for being wasteful if you keep starting a new megaproject every other month though. The R&D phase for just about anything always results in a lot of waste. The Apollo programme wasn't profitable either, but without it we wouldn't have the knowledge for modern launch vehicles to be either. Or to even exist. I'm pretty sure one day we'll have an LLM/LMM/VLA/etc. that's so good that pretraining a new one will seem pointless, and that'll finally be the time we get to (as a society) reap the benefits of our collective investment in the tech. The profitability of a single technology demonstrator model (which is what all current models are) is immaterial from that standpoint.
- wincy 1y ago
  Nah, if TSMC got exploded and there was a world war, in 20 years all the LLMs would bit rot.
  - moffkalast 1y ago
    Eh, I doubt it, tech only got massively better in each world war so far, through unlimited reckless strategic spending. We'd probably get a TSMC-like fab on every continent by the end of it. Maybe even optical computers. Quadrotor UAV are the future of warfare after all, and they require lots of compute. Adjusted for inflation it took over 120 billion to build the fleet of liberty ships during WW2, that's like at least 10 TSMC fabs.
    - aydyn 1y ago
      Technology is an exponential process, and the thing about exponentials is that they are chaotic. You cant use inductive reasoning vis a vis war and technology. The next big one could truly reset us to zero or worse.
erlend_sh 1y ago
Exactly. This is basically the argument of “AI as Normal Technology”. https://knightcolumbia.org/content/ai-as-normal-technology https://knightcolumbia.org/content/ai-as-normal-technology https://news.ycombinator.com/item?id=43697717 https://news.ycombinator.com/item?id=43697717
- highfrequency 1y ago
  Thanks for the link. The comparison to electricity is a good one, and this is a nice reflection on why it took time for electricity’s usefulness to show up in productivity stats: > What eventually allowed gains to be realized was redesigning the entire layout of factories around the logic of production lines. In addition to changes to factory architecture, diffusion also required changes to workplace organization and process control, which could only be developed through experimentation across industries.
- SirHumphrey 1y ago
  This seems like one the only sane arguments in this whole sea of articles.
ludicrousdispla 1y ago
>> There are many technologies that have seemed inevitable and seen retreats under the lack of commensurate business return 120+ Cable TV channels must have seemed like a good idea at the time, but like LLMs the vast majority of the content was not something people were interested in.
strangescript 1y ago
I think the difference between all previous technologies is scope. If you make a super sonic jet that gets people from place A to place B faster for more money, but the target consumer is like "yeah, I don't care that much about that at that price point", then your tech sort is of dead. You are also fully innovated on that product, like maybe you can make it more fuel efficient, sure, but your scope is narrow. AI is the opposite. There are numerous things it can do and numerous ways to improve it (currently). There is lower upfront investment than say a supersonic jet and many more ways it can pivot if something doesn't work out.
- digianarchist 1y ago
  It's not a great analogy. The only parallel with Concorde is energy consumption. I think a better analogy would have been VR.
  - strangescript 1y ago
    I mean, thats the point, they aren't the same. Concorde was one dimensional, AI is not.
- sumeno 1y ago
  The number of things it can actually do is significantly lower than the number of things the hype men are claiming it can do.
- peder 1y ago
  Most of the comments here feel like cope about AI TBH. There's never been an innovation like this ever, and it makes sense to get on board rather than be left behind.
  - Gormo 1y ago
    > There's never been an innovation like this ever There have been plenty of innovations like this. In fact, much of the hype around LLMs is a rehash of the hype around "expert systems" back in the '80s. LLMs are marginally more effective than those systems, but only marginally.
alonsonic 1y ago
I'm confused with your second point. LLM companies are not making any money from current models? Openai generates 10b USD ARR and has 100M MAUs. Yes they are running at a loss right now but that's because they are racing to improve models. If they stopped today to focus on optimization of their current models to minimize operating cost and monetizing their massive user base you think they don't have a successful business model? People use this tools daily, this is inevitable.
- airstrike 1y ago
  No, because if they stop to focus on optimizing and minimizing operating costs, the next competitor over will leapfrog them with a better model in 6-12 months, making all those margin improvements an NPV negative endeavor.
- bbor 1y ago
  It’s just the natural counterpart to dogmatic inevitabilism — dogmatic denialism. One denies the present, the other the (recent) past. It’s honestly an understandable PoV though when you consider A) most people understand “AI” and “chatbot” to be synonyms, and B) the blockchain hype cycle(s) bred some deep cynicism about software innovation. Funny seeing that comment on this post in particular, tho. When OP says “I’m not sure it’s a world I want”, I really don’t think they’re thinking about corporate revenue opportunities… More like Rehoboam, if not Skynet.
  - dvfjsdhgfv 1y ago
    > most people understand “AI” and “chatbot” to be synonyms This might be true (or not), but for sure not on this site.
    - bbor 1y ago
      I mean... LLMs have not yet discovered a business model that justifies the massive expenditure of training and hosting them, The only way one could say such a thing is if they think chatbots are the only real application.
- 1y ago
nyarlathotep_ 1y ago
I do wonder where in the cycle this all is given that we've now seen yet another LLM/"Agentic" VSCode fork. I'm genuinely surprised that Code forks and LLM cli things are seemingly the only use case that's approached viability. Even a year ago, I figured there'd be something else that's emerged by now.
- alonsonic 1y ago
  But there are a ton of LLM powered products in the market. I have a friend in finance that uses LLM powered products for financial analysis, he works in a big bank. Just now anthropic released a product to compete in this space. Another friend in real estate uses LLM powered lead qualifications products, he runs marketing campaigns and the AI handles the initial interaction via email or phone and then ranks the lead in their crm. I have a few friends that run small businesses and use LLM powered assistants to manage all their email comms and agendas. I've also talked with startups in legal and marketing doing very well. Coding is the theme that's talked about the most in HN but there are a ton of startups and big companies creating value with LLMs
  - Jach 1y ago
    Yup. Lots of products in the education space. Even doctors are using LLMs, while talking with patients. All sorts of teams are using the adjacent products for image and (increasingly) video generation. Translation freelancers have been hit somewhat hard because LLMs do "good enough" quite a bit better than old google translate. Coding is relevant to the HN bubble, and as tech is the biggest driver of the economy it's no surprise that tech-related AI usages will also be the biggest causes of investment, but it really is used in quite a lot of places out there already that aren't coding related at all.
  - materiallie 1y ago
    It feels like there's a lot of shifting goalposts. A year ago, the hype was that knowledge work would cease to exist by 2027. Now we are trying to hype up enhanced email autocomplete and data analysis as revolutionary? I agree that those things are useful. But it's not really addressing the criticism. I would have zero criticisms of AI marketing if it was "hey, look at this new technology that can assist your employees and make them 20% more productive". I think there's also a healthy dose of skepticism after the internet and social media age. Those were also society altering technologies that purported to democratize the political and economic system. I don't think those goals were accomplished, although without a doubt many workers and industries were made more productive. That effect is definitely real and I'm not denying that. But in other areas, the last 3 decades of technological advancement have been a resounding failure. We haven't made a dent in educational outcomes or intergenerational poverty, for instance.
dcow 1y ago
The difference is that the future is now with LLMs. There is a microwave (some multiple) in almost every kitchen in the world. The Concord served a few hundred people a day. LLMs are already ingrained into hundreds of millions if not billions of people’s lives, directly and indirectly. My dad directly uses LLMs multiple times a week if not daily in an industry that still makes you rotate your password every 3 months. It’s not a question of whether the future will have them, it’s a question of whether the future will get tired of them.
- jayd16 1y ago
  The huge leap that is getting pushback is the sentiment that LLMs will consume every use case and replace human labor. I don't think many are arguing LLMs will die off entirely.
smrtinsert 1y ago
They didn't really need the cloud either and yet...
JimmaDaRustla 1y ago
Investments are mostly in model training. We have trained models now, we'll see a pullback in that regard as businesses will need to optimize to get the best model without spending billions in order to compete on price, but LLMs are here to stay.
dmix 1y ago
> model capabilities are plateauing at a level where most people agree that the output is trite and unpleasant to consume. What are you basing this on? Personal feelings?
api 1y ago
My take since day one: (1) Model capabilities will plateau as training data is exhausted. Some additional gains will be possible by better training, better architectures, more compute, longer context windows or "infinite" context architectures, etc., but there are limits here. (2) Training on synthetic data beyond a very limited amount will result in overfitting because there is no new information. To some extent you could train models on each other, but that's just an indirect way to consolidate models. Beyond consolidation you'll plateau. (3) There will be no "takeoff" scenario -- this is sci-fi (in the pejorative sense) because you can't exceed available information. There is no magic way that a brain in a vat can innovate beyond available training data. This includes for humans -- a brain in a vat would quickly go mad and then spiral into a coma-like state. The idea of AI running away is the information-theoretic equivalent of a perpetual motion machine and is impossible. Yudkowski and the rest of the people afraid of this are crackpots, and so are the hype-mongers betting on it. So I agree that LLMs are real and useful, but the hype and bubble are starting to plateau. The bubble is predicated on the idea that you can just keep going forever.
- ogogmad 1y ago
  The next step is clearly improved vision and everyday-physics models. These can also solve hallucinations.
  - chrz 1y ago
    clearly
giancarlostoro 1y ago
> 2. Almost three years in, companies investing in LLMs have not yet discovered a business model that justifies the massive expenditure of training and hosting them, the majority of consumer usage is at the free tier, the industry is seeing the first signs of pulling back investments, and model capabilities are plateauing at a level where most people agree that the output is trite and unpleasant to consume. You hit the nail on why I say to much hatred from "AI Bros" as I call them, when I say it will not take off truly until it runs on your phone effortlessly, because nobody wants to foot a trillion dollar cloud bill. Give me a fully offline LLM that fits in 2GB of VRAM and lets refine that so it can plug into external APIs and see how much farther we can take things without resorting to burning billions of dollars' worth of GPU compute. I don't care that my answer arrives instantly, if I'm doing the research myself, I want to take my time to get the correct answer anyway.
- DSingularity 1y ago
  You aren’t extrapolating enough. Nearly the entire history of computing has been one that isolates between shared computing and personal computing. Give it time. These massive cloud bills are building the case for accelerators in phones. It’s going to happen just needs time.
  - giancarlostoro 1y ago
    That's fine, that's what I want ;) I just grow tired of people hating on me for thinking that we really need to localize the models for them to take off.
    - DSingularity 1y ago
      I’m not sure why people are hating on you. If you love being free then you should love the idea of being independent when it comes to common computing. If LLM is to become common we should all be rooting for open weights and efficient local execution. It’s gonna take some time but it’s inevitable I think.
- saratogacx 1y ago
Jach 1y ago
I don't really buy your point 2. Just the other day Meta announced hundreds of billions of dollars investment into more AI datacenters. Companies are bringing back nuclear power plants to support this stuff. Earlier this year OpenAI and Oracle announced their $500bn AI datacenter project, but admittedly in favor of your point have run into funding snags, though that's supposedly from tariff fears with foreign investors, not lack of confidence in AI. Meta can just finance everything from their own capital and Zuck's decree, like they did with VR (and it may very well turn out similarly). Since you brought up supersonic jetliners you're probably aware of the startup Boom in Colorado trying to bring it back. We'll see if they succeed. But yes, it would be a strange path, but a possible one, that LLMs kind of go away for a while and try to come back later. You're going to have to cite some surveys for the "most people agree that the output is trite and unpleasant" and "almost universally disliked attempts to cram it everywhere" claims. There are some very vocal people against LLM flavors of AI, but I don't think they even represent the biggest minority, let alone a majority or near universal opinions. (I personally was bugged by earlier attempts at cramming non-LLM AI into a lot of places, e.g. Salesforce Einstein appeared I think in 2016, and that was mostly just being put off by the cutesy Einstein characterization. I generally don't have the same feelings with LLMs in particular, in some cases they're small improvements to an already annoying process, e.g. non-human customer support that was previously done by a crude chatbot front-end to an expert system or knowledge base, the LLM version of that tends to be slightly less annoying.)
- Jach 1y ago
  Sort of a followup to myself if I come back searching this comment or someone sees this thread later... here's a study that just came out on AI attitudes: https://report2025.seismic.org/ https://report2025.seismic.org/ I don't think it supports the bits I quoted, but it does include more negativity than I would have predicted before seeing it.
brokencode 1y ago
> “most people agree that the output is trite and unpleasant to consume” That is a such a wild claim. People like the output of LLMs so much that ChatGPT is the fastest growing app ever. It and other AI apps like Perplexity are now beginning to challenge Google’s search dominance. Sure, probably not a lot of people would go out and buy a novel or collection of poetry written by ChatGPT. But that doesn’t mean the output is unpleasant to consume. It pretty undeniably produces clear and readable summaries and explanations.
- sejje 1y ago
  Maybe he's referencing how people don't like when other humans post LLM responses in the comments. "Here's what chatGPT said about..." I don't like that, either. I love the LLM for answering my own questions, though.
  - jack_pp 1y ago
    "Here's what chatGPT said about..." Is the new lmgtfy
    - zdragnar 1y ago
      lmgtfy was (from what I saw) always used as a snarky way to tell someone to do a little work on their own before asking someone else to do it for them. I have seen people use "here's what chatGPT" said almost exclusively unironically, as if anyone else wants humans behaving like agents for chatbots in the middle of other people's discussion threads. That is to say, they offer no opinion or critical thought of their own, they just jump into a conversation with a wall of text.
      - SoftTalker 1y ago
        Yeah I don't even read those. If someone can't be bothered to communicate their own thoughts in their own words, I have little belief that they are adding anything worth reading to the conversation.
UncleOxidant 1y ago
Let's not ignore the technical aspects as well: LLMs are probably a local minima that we've gotten stuck in because of their rapid rise. Other areas in AI are being starved of investment because all of the capital is pouring into LLMs. We might have been better off in the long run if LLMs hadn't been so successful so fast.
philomath_mn 1y ago
> most people agree that the output is trite and unpleasant to consume This is likely a selection bias: you only notice the obviously bad outputs. I have created plenty of outputs myself that are good/passable -- you are likely surrounded by these types of outputs without noticing. Not a panacea, but can be useful.
magic_hamster 1y ago
There are pretty hidden assumption in this comment. First of all, not every business in the AI space is _training_ models, and the difference between training and inference is massive - i.e. most businesses can easily afford inference, perhaps depending on model, but they definitely can. Another several unfounded claims were made here, but I just wanted to say LLMs with MCP are definitely good enough for almost every use case you can come up with as long as you can provide them with high quality context. LLMs are absolutely the future and they will take over massive parts of our workflow in many industries. Try MCP for yourself and see. There's just no going back.
- dontlikeyoueith 1y ago
  > I just wanted to say LLMs with MCP are definitely good enough for almost every use case you can come up with as long as you can provide them with high quality context. This just shows you lack imagination. I have a lot of use cases that they are not good enough for.
  - thunspa 1y ago
    I mean, it all depends on how one defines "high quality context".
- ramoz 1y ago
  LLMs with tools* MCP isn’t inherently special. A Claude Code with Bash() tool can do nearly anything a MCP server will give you - much more efficiently. Computer Use agents are here and are only going to get better. The conversation shouldn’t be about LLMs any longer. Providers will be providing agents.
  - anthonypasq 1y ago
    correct and companies will be exposing their data via mcp instead of standard rest apis.
    - ramoz 1y ago
strange_quark 1y ago
> There are many technologies that have seemed inevitable and seen retreats under the lack of commensurate business return (the supersonic jetliner) I think this is a great analogy, not just to the current state of AI, but maybe even computers and the internet in general. Supersonic transports must've seemed amazing, inevitable, and maybe even obvious to anyone alive at the time of their debut. But hiding under that amazing tech was a whole host of problems that were just not solvable with the technology of the era, let alone a profitable business model. I wonder if computers and the internet are following a similar trajectory to aerospace. Maybe we've basically peaked, and all that's left are optimizations around cost, efficiency, distribution, or convenience. If you time traveled back to the 1970s and talked to most adults, they would have witnessed aerospace go from loud, smelly, and dangerous prop planes to the 707, 747 and Concorde. They would've witnessed the moon landings and were seeing the development of the Space Shuttle. I bet they would call you crazy if you told this person that 50 years later, in 2025, there would be no more supersonic commercial airliners, commercial aviation would basically look the same except more annoying, and also that we haven't been back to the moon. In the previous 50 years we went from the Wright Brothers to the 707! So maybe in 2075 we'll all be watching documentaries about LLMs (maybe even on our phones or laptops that look basically the same), and reminiscing about the mid-2020s and wondering why what seemed to be such a promising technology disappeared almost entirely.
- kenjackson 1y ago
  I think this is both right and wrong. There was a good book that came out probably 15 years ago about how technology never stops in aggregate, but individual technologies tend to grow quickly and then stall. Airplane jets were one example in the book. The reason why I partially note this as wrong is that even in the 70s people recognized that supersonic travel had real concrete issues with no solution in sight. I don't think LLMs share that characteristic today. A better example, also in the book, are skyscrapers. Each year they grew and new ones were taller than the ones last year. The ability to build them and traverse them increased each year with new technologies to support it. There wasn't a general consensus around issues that would stop growth (except at more extremes like air pressure). But the growth did stop. No one even has expectations of taller skyscrapers any more. LLMs may fail to advance, but not because of any consensus reason that exists today. And it maybe that they serve their purpose to build something on top of them which ends up being far more revolutionary than LLMs. This is more like the path of electricity -- electricity in itself isn't that exciting nowadays, but almost every piece of technology built uses it. I fundamentally find it odd that people seem so against AI. I get the potential dystopian future, which I also don't want. But the more mundane annoyance seems odd to me.
MonkeyIsNull 1y ago
> 2. Almost three years in, companies investing in LLMs have not yet discovered a business model that justifies the massive expenditure of training and hosting them, I always think back to how Bezos and Amazon were railed against for losing money for years. People thought that would never work. And then when he started selling stuff other than books? People I know were like: please, he's desperate. Someone, somewhere will figure out how to make money off it - just not most people.
gonzobonzo 1y ago
My guess is that LLM's are bridge technology, the equivalent of cassette tapes. A big step forward, allowing things that we couldn't before. But before long they'll be surpassed by much better technology, and future generations will look back on them as primitive. You have top scientists like LeCun arguing this position. I'd imagine all of these companies are desperately searching for the next big paradigm shift, but no one knows when that will be, and until then they need to squeeze everything they can out of LLMs.
jittery41 1y ago
Oh wow I forgot that the microwave oven was once marketed as the total replacement of cooking chores and in futuristic life people can just press a button and have a delicious good meal ( well you can now but microwave meals are often seen as worse than fastfood ).
alexpotato 1y ago
To use the Internet as a comparison: Phase 1 - mid to late 1990s: - "The Internet is going to change EVERYTHING!!!" Phase 2 - late 1990s to early 2000s: - "It's amazing and we are all making SO much money!" - "Oh no! The bubble burst" - "Of course everyone could see this coming: who is going to buy 40 lb bags of dogfood or their groceries over the Internet?!?!?" Phase 3 - mid 2000s to 2020: - "It is astounding the amount of money being by tech companies" - "Who could have predicted that social media would change the ENTIRE landscape??"

by lsy 1y ago

xnx 1y ago

> (the supersonic jetliner) ... (the microwave oven) But have we ever had a general purpose technology (steam engine, electricity) that failed to change society?

blueflow 1y ago
It wouldn't be general purpose if it fails to bring change. I'd take every previous iteration of "AI" as example, IBM Watson, that stuff

fendy3002 1y ago

LLMs need significant optimization or we get significant improvement on computing power while keeping the energy cost the same. It's similar with smartphone, when at the start it's not feasible because of computing power, and now we have one that can rival 2000s notebooks. LLMs is too trivial to be expensive EDIT: I presented the statement wrongly. What I mean is the use case for LLM are trivial things, it shouldn't be expensive to operate

trashchomper 1y ago
Calling LLMs trivial is a new one. Yea just consume all of the information on the internet and encode it into a statistical model, trivial, child could do it /s
- hammyhavoc 1y ago
  > all of the information on the internet Total exaggeration—especially given Cloudflare providing free tools to block AI and now tools to charge bots for access to information.
- fendy3002 1y ago
  well I presented the statement wrongly. What I mean is the use case for LLM are trivial things, it shouldn't be expensive to operate
lblume 1y ago
Imagine telling a person from five years ago that the programs that would basically solve NLP, perform better than experts at many tasks and are hard not to anthropomorphize accidentally are actually "trivial". Good luck with that.
- jrflowers 1y ago
  >programs that would basically solve NLP There is a load-bearing “basically” in this statement about the chat bots that just told me that the number of dogs granted forklift certification in 2023 is 8,472.
  - lblume 1y ago
    Sure, maybe solving NLP is too great a claim to make. It is still not at all ordinary that beforehand we could not solve referential questions algorithmically, that we could not extract information from plain text into custom schemas of structured data, and context-aware mechanical translation was really unheard of. Nowadays LLMs can do most of these tasks better than most humans in most scenarios. Many NLP questions at least I find interesting reduce to questions of the explanability of LLMs.

eric-burel 1y ago

Developers haven't even started extracting the value of LLMs with agent architectures yet. Using an LLM UI like open ai is like we just figured fire and you use it to warm you hands (still impressive when you think about it, but not worth the burns), while LLM development is about building car engines (here is you return on investment).

clarinificator 1y ago
Every booster argument is like this one. $trite_analogy triumphant smile
- Karrot_Kream 1y ago
  [flagged]
  - rrr_oh_man 1y ago
    Throwing genetic fallacies around instead of engaging with the comment at hand... :)
Jensson 1y ago
> Developers haven't even started extracting the value of LLMs with agent architectures yet There are thousands of startups doing exactly that right now, why do you think this will work when all evidence points towards it not working? Or why else would it not already have revolutionized everything a year or two ago when everyone started doing this?
- eric-burel 1y ago
  Most of them are a bunch of prompts and don't even have actual developers. For the good reason that there is no training system yet and the wording of how you call the people that build these system isn't even there or clearly defined. Local companies haven't even setup a proper internal LLM or at least a contract with a provider. I am in France so probably lagging behind USA a bit especially NY/SF but the word "LLM developer" is just arriving now and mostly under the pressure of isolated developers and companies like me. This feel really really early stage.

Msurrow 1y ago

> first signs of pulling back investments I agree with you, but I’m curious; do you have link to one or two concrete examples of companies pulling back investments, or rolling back an AI push? (Yes it’s just to fuel my confirmation bias, but it’s still feels nice:-) )

0xAFFFF 1y ago
Most prominent example was this one: https://www.reuters.com/technology/microsoft-pulls-back-more-data-center-leases-us-europe-analysts-say-2025-03-26/ https://www.reuters.com/technology/microsoft-pulls-back-more...
- durumu 1y ago
  I think that's more reflective of the deteriorating relationship between OpenAI and Microsoft than an true lack of demand for datacenters. If a major model provider (OpenAI, Anthropic, Google, xAI) were to see a dip in available funding or stop focusing on training more powerful models, that would convince me we may be in a bubble about to pop, but there are no signs of that as far as I can see.

moffkalast 1y ago

ML models have the good property of only requiring investment once and can then be used till the end of history or until something better replaces them. Granted the initial investment is immense, and the results are not guaranteed which makes it risky, but it's like building a dam or a bridge. Being in the age where bridge technology evolves massively on a weekly basis is a recipe for being wasteful if you keep starting a new megaproject every other month though. The R&D phase for just about anything always results in a lot of waste. The Apollo programme wasn't profitable either, but without it we wouldn't have the knowledge for modern launch vehicles to be either. Or to even exist. I'm pretty sure one day we'll have an LLM/LMM/VLA/etc. that's so good that pretraining a new one will seem pointless, and that'll finally be the time we get to (as a society) reap the benefits of our collective investment in the tech. The profitability of a single technology demonstrator model (which is what all current models are) is immaterial from that standpoint.

wincy 1y ago
Nah, if TSMC got exploded and there was a world war, in 20 years all the LLMs would bit rot.
- moffkalast 1y ago
  Eh, I doubt it, tech only got massively better in each world war so far, through unlimited reckless strategic spending. We'd probably get a TSMC-like fab on every continent by the end of it. Maybe even optical computers. Quadrotor UAV are the future of warfare after all, and they require lots of compute. Adjusted for inflation it took over 120 billion to build the fleet of liberty ships during WW2, that's like at least 10 TSMC fabs.
  - aydyn 1y ago
    Technology is an exponential process, and the thing about exponentials is that they are chaotic. You cant use inductive reasoning vis a vis war and technology. The next big one could truly reset us to zero or worse.

erlend_sh 1y ago

Exactly. This is basically the argument of “AI as Normal Technology”. https://knightcolumbia.org/content/ai-as-normal-technology https://knightcolumbia.org/content/ai-as-normal-technology https://news.ycombinator.com/item?id=43697717 https://news.ycombinator.com/item?id=43697717

highfrequency 1y ago
Thanks for the link. The comparison to electricity is a good one, and this is a nice reflection on why it took time for electricity’s usefulness to show up in productivity stats: > What eventually allowed gains to be realized was redesigning the entire layout of factories around the logic of production lines. In addition to changes to factory architecture, diffusion also required changes to workplace organization and process control, which could only be developed through experimentation across industries.
SirHumphrey 1y ago
This seems like one the only sane arguments in this whole sea of articles.

ludicrousdispla 1y ago

>> There are many technologies that have seemed inevitable and seen retreats under the lack of commensurate business return 120+ Cable TV channels must have seemed like a good idea at the time, but like LLMs the vast majority of the content was not something people were interested in.

strangescript 1y ago

I think the difference between all previous technologies is scope. If you make a super sonic jet that gets people from place A to place B faster for more money, but the target consumer is like "yeah, I don't care that much about that at that price point", then your tech sort is of dead. You are also fully innovated on that product, like maybe you can make it more fuel efficient, sure, but your scope is narrow. AI is the opposite. There are numerous things it can do and numerous ways to improve it (currently). There is lower upfront investment than say a supersonic jet and many more ways it can pivot if something doesn't work out.

digianarchist 1y ago
It's not a great analogy. The only parallel with Concorde is energy consumption. I think a better analogy would have been VR.
- strangescript 1y ago
  I mean, thats the point, they aren't the same. Concorde was one dimensional, AI is not.
sumeno 1y ago
The number of things it can actually do is significantly lower than the number of things the hype men are claiming it can do.
peder 1y ago
Most of the comments here feel like cope about AI TBH. There's never been an innovation like this ever, and it makes sense to get on board rather than be left behind.
- Gormo 1y ago
  > There's never been an innovation like this ever There have been plenty of innovations like this. In fact, much of the hype around LLMs is a rehash of the hype around "expert systems" back in the '80s. LLMs are marginally more effective than those systems, but only marginally.

alonsonic 1y ago

I'm confused with your second point. LLM companies are not making any money from current models? Openai generates 10b USD ARR and has 100M MAUs. Yes they are running at a loss right now but that's because they are racing to improve models. If they stopped today to focus on optimization of their current models to minimize operating cost and monetizing their massive user base you think they don't have a successful business model? People use this tools daily, this is inevitable.

airstrike 1y ago
No, because if they stop to focus on optimizing and minimizing operating costs, the next competitor over will leapfrog them with a better model in 6-12 months, making all those margin improvements an NPV negative endeavor.
bbor 1y ago
It’s just the natural counterpart to dogmatic inevitabilism — dogmatic denialism. One denies the present, the other the (recent) past. It’s honestly an understandable PoV though when you consider A) most people understand “AI” and “chatbot” to be synonyms, and B) the blockchain hype cycle(s) bred some deep cynicism about software innovation. Funny seeing that comment on this post in particular, tho. When OP says “I’m not sure it’s a world I want”, I really don’t think they’re thinking about corporate revenue opportunities… More like Rehoboam, if not Skynet.
- dvfjsdhgfv 1y ago
  > most people understand “AI” and “chatbot” to be synonyms This might be true (or not), but for sure not on this site.
  - bbor 1y ago
    I mean... LLMs have not yet discovered a business model that justifies the massive expenditure of training and hosting them, The only way one could say such a thing is if they think chatbots are the only real application.
1y ago

nyarlathotep_ 1y ago

I do wonder where in the cycle this all is given that we've now seen yet another LLM/"Agentic" VSCode fork. I'm genuinely surprised that Code forks and LLM cli things are seemingly the only use case that's approached viability. Even a year ago, I figured there'd be something else that's emerged by now.

alonsonic 1y ago
But there are a ton of LLM powered products in the market. I have a friend in finance that uses LLM powered products for financial analysis, he works in a big bank. Just now anthropic released a product to compete in this space. Another friend in real estate uses LLM powered lead qualifications products, he runs marketing campaigns and the AI handles the initial interaction via email or phone and then ranks the lead in their crm. I have a few friends that run small businesses and use LLM powered assistants to manage all their email comms and agendas. I've also talked with startups in legal and marketing doing very well. Coding is the theme that's talked about the most in HN but there are a ton of startups and big companies creating value with LLMs
- Jach 1y ago
  Yup. Lots of products in the education space. Even doctors are using LLMs, while talking with patients. All sorts of teams are using the adjacent products for image and (increasingly) video generation. Translation freelancers have been hit somewhat hard because LLMs do "good enough" quite a bit better than old google translate. Coding is relevant to the HN bubble, and as tech is the biggest driver of the economy it's no surprise that tech-related AI usages will also be the biggest causes of investment, but it really is used in quite a lot of places out there already that aren't coding related at all.
- materiallie 1y ago
  It feels like there's a lot of shifting goalposts. A year ago, the hype was that knowledge work would cease to exist by 2027. Now we are trying to hype up enhanced email autocomplete and data analysis as revolutionary? I agree that those things are useful. But it's not really addressing the criticism. I would have zero criticisms of AI marketing if it was "hey, look at this new technology that can assist your employees and make them 20% more productive". I think there's also a healthy dose of skepticism after the internet and social media age. Those were also society altering technologies that purported to democratize the political and economic system. I don't think those goals were accomplished, although without a doubt many workers and industries were made more productive. That effect is definitely real and I'm not denying that. But in other areas, the last 3 decades of technological advancement have been a resounding failure. We haven't made a dent in educational outcomes or intergenerational poverty, for instance.

dcow 1y ago

The difference is that the future is now with LLMs. There is a microwave (some multiple) in almost every kitchen in the world. The Concord served a few hundred people a day. LLMs are already ingrained into hundreds of millions if not billions of people’s lives, directly and indirectly. My dad directly uses LLMs multiple times a week if not daily in an industry that still makes you rotate your password every 3 months. It’s not a question of whether the future will have them, it’s a question of whether the future will get tired of them.

jayd16 1y ago
The huge leap that is getting pushback is the sentiment that LLMs will consume every use case and replace human labor. I don't think many are arguing LLMs will die off entirely.

smrtinsert 1y ago

They didn't really need the cloud either and yet...

JimmaDaRustla 1y ago

Investments are mostly in model training. We have trained models now, we'll see a pullback in that regard as businesses will need to optimize to get the best model without spending billions in order to compete on price, but LLMs are here to stay.

dmix 1y ago

> model capabilities are plateauing at a level where most people agree that the output is trite and unpleasant to consume. What are you basing this on? Personal feelings?

api 1y ago

My take since day one: (1) Model capabilities will plateau as training data is exhausted. Some additional gains will be possible by better training, better architectures, more compute, longer context windows or "infinite" context architectures, etc., but there are limits here. (2) Training on synthetic data beyond a very limited amount will result in overfitting because there is no new information. To some extent you could train models on each other, but that's just an indirect way to consolidate models. Beyond consolidation you'll plateau. (3) There will be no "takeoff" scenario -- this is sci-fi (in the pejorative sense) because you can't exceed available information. There is no magic way that a brain in a vat can innovate beyond available training data. This includes for humans -- a brain in a vat would quickly go mad and then spiral into a coma-like state. The idea of AI running away is the information-theoretic equivalent of a perpetual motion machine and is impossible. Yudkowski and the rest of the people afraid of this are crackpots, and so are the hype-mongers betting on it. So I agree that LLMs are real and useful, but the hype and bubble are starting to plateau. The bubble is predicated on the idea that you can just keep going forever.

ogogmad 1y ago
The next step is clearly improved vision and everyday-physics models. These can also solve hallucinations.
- chrz 1y ago
  clearly

giancarlostoro 1y ago

> 2. Almost three years in, companies investing in LLMs have not yet discovered a business model that justifies the massive expenditure of training and hosting them, the majority of consumer usage is at the free tier, the industry is seeing the first signs of pulling back investments, and model capabilities are plateauing at a level where most people agree that the output is trite and unpleasant to consume. You hit the nail on why I say to much hatred from "AI Bros" as I call them, when I say it will not take off truly until it runs on your phone effortlessly, because nobody wants to foot a trillion dollar cloud bill. Give me a fully offline LLM that fits in 2GB of VRAM and lets refine that so it can plug into external APIs and see how much farther we can take things without resorting to burning billions of dollars' worth of GPU compute. I don't care that my answer arrives instantly, if I'm doing the research myself, I want to take my time to get the correct answer anyway.

DSingularity 1y ago
You aren’t extrapolating enough. Nearly the entire history of computing has been one that isolates between shared computing and personal computing. Give it time. These massive cloud bills are building the case for accelerators in phones. It’s going to happen just needs time.
- giancarlostoro 1y ago
  That's fine, that's what I want ;) I just grow tired of people hating on me for thinking that we really need to localize the models for them to take off.
  - DSingularity 1y ago
    I’m not sure why people are hating on you. If you love being free then you should love the idea of being independent when it comes to common computing. If LLM is to become common we should all be rooting for open weights and efficient local execution. It’s gonna take some time but it’s inevitable I think.
saratogacx 1y ago

Jach 1y ago

I don't really buy your point 2. Just the other day Meta announced hundreds of billions of dollars investment into more AI datacenters. Companies are bringing back nuclear power plants to support this stuff. Earlier this year OpenAI and Oracle announced their $500bn AI datacenter project, but admittedly in favor of your point have run into funding snags, though that's supposedly from tariff fears with foreign investors, not lack of confidence in AI. Meta can just finance everything from their own capital and Zuck's decree, like they did with VR (and it may very well turn out similarly). Since you brought up supersonic jetliners you're probably aware of the startup Boom in Colorado trying to bring it back. We'll see if they succeed. But yes, it would be a strange path, but a possible one, that LLMs kind of go away for a while and try to come back later. You're going to have to cite some surveys for the "most people agree that the output is trite and unpleasant" and "almost universally disliked attempts to cram it everywhere" claims. There are some very vocal people against LLM flavors of AI, but I don't think they even represent the biggest minority, let alone a majority or near universal opinions. (I personally was bugged by earlier attempts at cramming non-LLM AI into a lot of places, e.g. Salesforce Einstein appeared I think in 2016, and that was mostly just being put off by the cutesy Einstein characterization. I generally don't have the same feelings with LLMs in particular, in some cases they're small improvements to an already annoying process, e.g. non-human customer support that was previously done by a crude chatbot front-end to an expert system or knowledge base, the LLM version of that tends to be slightly less annoying.)

Jach 1y ago
Sort of a followup to myself if I come back searching this comment or someone sees this thread later... here's a study that just came out on AI attitudes: https://report2025.seismic.org/ https://report2025.seismic.org/ I don't think it supports the bits I quoted, but it does include more negativity than I would have predicted before seeing it.

brokencode 1y ago

> “most people agree that the output is trite and unpleasant to consume” That is a such a wild claim. People like the output of LLMs so much that ChatGPT is the fastest growing app ever. It and other AI apps like Perplexity are now beginning to challenge Google’s search dominance. Sure, probably not a lot of people would go out and buy a novel or collection of poetry written by ChatGPT. But that doesn’t mean the output is unpleasant to consume. It pretty undeniably produces clear and readable summaries and explanations.

sejje 1y ago
Maybe he's referencing how people don't like when other humans post LLM responses in the comments. "Here's what chatGPT said about..." I don't like that, either. I love the LLM for answering my own questions, though.
- jack_pp 1y ago
  "Here's what chatGPT said about..." Is the new lmgtfy
  - zdragnar 1y ago
    lmgtfy was (from what I saw) always used as a snarky way to tell someone to do a little work on their own before asking someone else to do it for them. I have seen people use "here's what chatGPT" said almost exclusively unironically, as if anyone else wants humans behaving like agents for chatbots in the middle of other people's discussion threads. That is to say, they offer no opinion or critical thought of their own, they just jump into a conversation with a wall of text.
    - SoftTalker 1y ago
      Yeah I don't even read those. If someone can't be bothered to communicate their own thoughts in their own words, I have little belief that they are adding anything worth reading to the conversation.

UncleOxidant 1y ago

Let's not ignore the technical aspects as well: LLMs are probably a local minima that we've gotten stuck in because of their rapid rise. Other areas in AI are being starved of investment because all of the capital is pouring into LLMs. We might have been better off in the long run if LLMs hadn't been so successful so fast.

philomath_mn 1y ago

> most people agree that the output is trite and unpleasant to consume This is likely a selection bias: you only notice the obviously bad outputs. I have created plenty of outputs myself that are good/passable -- you are likely surrounded by these types of outputs without noticing. Not a panacea, but can be useful.

magic_hamster 1y ago

There are pretty hidden assumption in this comment. First of all, not every business in the AI space is _training_ models, and the difference between training and inference is massive - i.e. most businesses can easily afford inference, perhaps depending on model, but they definitely can. Another several unfounded claims were made here, but I just wanted to say LLMs with MCP are definitely good enough for almost every use case you can come up with as long as you can provide them with high quality context. LLMs are absolutely the future and they will take over massive parts of our workflow in many industries. Try MCP for yourself and see. There's just no going back.

dontlikeyoueith 1y ago
> I just wanted to say LLMs with MCP are definitely good enough for almost every use case you can come up with as long as you can provide them with high quality context. This just shows you lack imagination. I have a lot of use cases that they are not good enough for.
- thunspa 1y ago
  I mean, it all depends on how one defines "high quality context".
ramoz 1y ago
LLMs with tools* MCP isn’t inherently special. A Claude Code with Bash() tool can do nearly anything a MCP server will give you - much more efficiently. Computer Use agents are here and are only going to get better. The conversation shouldn’t be about LLMs any longer. Providers will be providing agents.
- anthonypasq 1y ago
  correct and companies will be exposing their data via mcp instead of standard rest apis.
  - ramoz 1y ago

strange_quark 1y ago

> There are many technologies that have seemed inevitable and seen retreats under the lack of commensurate business return (the supersonic jetliner) I think this is a great analogy, not just to the current state of AI, but maybe even computers and the internet in general. Supersonic transports must've seemed amazing, inevitable, and maybe even obvious to anyone alive at the time of their debut. But hiding under that amazing tech was a whole host of problems that were just not solvable with the technology of the era, let alone a profitable business model. I wonder if computers and the internet are following a similar trajectory to aerospace. Maybe we've basically peaked, and all that's left are optimizations around cost, efficiency, distribution, or convenience. If you time traveled back to the 1970s and talked to most adults, they would have witnessed aerospace go from loud, smelly, and dangerous prop planes to the 707, 747 and Concorde. They would've witnessed the moon landings and were seeing the development of the Space Shuttle. I bet they would call you crazy if you told this person that 50 years later, in 2025, there would be no more supersonic commercial airliners, commercial aviation would basically look the same except more annoying, and also that we haven't been back to the moon. In the previous 50 years we went from the Wright Brothers to the 707! So maybe in 2075 we'll all be watching documentaries about LLMs (maybe even on our phones or laptops that look basically the same), and reminiscing about the mid-2020s and wondering why what seemed to be such a promising technology disappeared almost entirely.

kenjackson 1y ago
I think this is both right and wrong. There was a good book that came out probably 15 years ago about how technology never stops in aggregate, but individual technologies tend to grow quickly and then stall. Airplane jets were one example in the book. The reason why I partially note this as wrong is that even in the 70s people recognized that supersonic travel had real concrete issues with no solution in sight. I don't think LLMs share that characteristic today. A better example, also in the book, are skyscrapers. Each year they grew and new ones were taller than the ones last year. The ability to build them and traverse them increased each year with new technologies to support it. There wasn't a general consensus around issues that would stop growth (except at more extremes like air pressure). But the growth did stop. No one even has expectations of taller skyscrapers any more. LLMs may fail to advance, but not because of any consensus reason that exists today. And it maybe that they serve their purpose to build something on top of them which ends up being far more revolutionary than LLMs. This is more like the path of electricity -- electricity in itself isn't that exciting nowadays, but almost every piece of technology built uses it. I fundamentally find it odd that people seem so against AI. I get the potential dystopian future, which I also don't want. But the more mundane annoyance seems odd to me.

MonkeyIsNull 1y ago

> 2. Almost three years in, companies investing in LLMs have not yet discovered a business model that justifies the massive expenditure of training and hosting them, I always think back to how Bezos and Amazon were railed against for losing money for years. People thought that would never work. And then when he started selling stuff other than books? People I know were like: please, he's desperate. Someone, somewhere will figure out how to make money off it - just not most people.

gonzobonzo 1y ago

My guess is that LLM's are bridge technology, the equivalent of cassette tapes. A big step forward, allowing things that we couldn't before. But before long they'll be surpassed by much better technology, and future generations will look back on them as primitive. You have top scientists like LeCun arguing this position. I'd imagine all of these companies are desperately searching for the next big paradigm shift, but no one knows when that will be, and until then they need to squeeze everything they can out of LLMs.

jittery41 1y ago

Oh wow I forgot that the microwave oven was once marketed as the total replacement of cooking chores and in futuristic life people can just press a button and have a delicious good meal ( well you can now but microwave meals are often seen as worse than fastfood ).

alexpotato 1y ago

To use the Internet as a comparison: Phase 1 - mid to late 1990s: - "The Internet is going to change EVERYTHING!!!" Phase 2 - late 1990s to early 2000s: - "It's amazing and we are all making SO much money!" - "Oh no! The bubble burst" - "Of course everyone could see this coming: who is going to buy 40 lb bags of dogfood or their groceries over the Internet?!?!?" Phase 3 - mid 2000s to 2020: - "It is astounding the amount of money being by tech companies" - "Who could have predicted that social media would change the ENTIRE landscape??"