Every bubble goes through this phase just before it pops. Pre-implosion Twitter had tons of folks doing useless “research” projects, WeWork had folks studying furniture design’s impact on wellbeing, and so on.
When you’re more worried about this sort of stuff vs generating profit then the history books strongly indicate that’s usually not a good omen of what’s coming next.
Not OP but I think the most obvious change will be rapid price increases as investors begin to demand profits - no more subsidised $20p/m plans for example.
So far I know, service providers hosting open-weight models are not in such a financial conundrum and can continue serving at the current price levels.
Also, even if the 20 dollar subs disappear, that just means regular folks can't afford AI. API pricing is still a great bargain for businesses replacing generic office workers.
Flight had been tried many times, ending in crash and death each time until the tech got worked out. That it's been tried so many times shows the demand is enormous.
> That it's been tried so many times shows the demand is enormous
Capitalists want to pay their workers as little as possible or even not have workers and pocket close to 100% of the profits since forever. Whether they'll be able to do it this time is yet to be seen. Putting the tech itself aside, which is far from good enough yet imo, the only way to avoid a massive social backslash would be to use some of the additional profit they're able to keep because of AI for social spending. And I just don't think they realize yet that it's probably in their interest to do so if they want to replace 'human resources' with AI.
Agreed but the AI lab investors and CEOs aren’t ideologically aligned with concepts like UBI. Some of them have even commented on social media about the unsustainability of social security and need for cuts. Also why would you they do trillions in capex investments just to give their returns away.
They are clearly smart/dumb enough to think pushing labor replacement to 20%+++ unemployment levels would have no consequences for them personally. Maybe with their private jets, private islands, doomsday shelters, etc they are right.
Seems like a bit of a gamble to me, or probably anyone who has read a bit of history. But I’m not a billionaire.
Well, it is good enough for a huge percentage of what general administrative office staff are doing all day. Essentially copying information between textboxes, slightly rephrasing, clicking buttons, formulating emails to people to send in various documents, scheduling meetings, making reports that are fairly mechanical summaries of other information etc. Not all of it is like this, but probably in the range of 30-70%? Some of it may involve human judgment though, which is harder to automate, like whether some expenses are justified enough etc. But you can just replace those kinds of judgments with more clearly defined rules and cost limits etc.
Have you tried using "computer use" models? They just started to get good in the last few months. They can comfortably use GUIs, click the right buttons, enter the data faster etc. This is very significant and my programmer friends often don't appreciate the earth shattering consequences. The vast majority of software in such applications is legacy stuff or proprietary and does not and likely will never have a programmatic API. The only way to interact with it is via GUI. And an AI working that GUI can be deployed unilaterally by the customer of the software company who made the GUI program. This makes adoption and automation much easier. My prediction is that within 1 year (maybe 2) GUI-using AI will be much more powerful and it will be everywhere (of course it doesn't need to be displayed on a screen, it will do its virtual clicks based on a virtual screen buffer rendering in an isolated virtual machine, along with countless other agents in parallel).
There will be huge controversy around it, because this will be the first instance that regular people will see AI do work that resembles their day job all over social media. There will be panic. There will be new license terms saying only humans are allowed to operate the GUI, and AI use is either forbidden or costs extra. But this will not be checkable properly. Even when forbidden, some employees will pay from their own pockets to have cheap AI do the job instead of them.
And if the bubble does pop chaotically, the open weight inference providers will become even more viable as they'll be able to buy racks full of compute for pennies on the dollar.
Something very similar happened with the late 90s telecoms bubble, which collapsed a little before the dot com bust. There was a very nice living to be made by many small/mid-sized providers who were able to feed on the remains for about a decade afterwards.
Probably not. They aren't paying sustainable prices for their supply chain.
1. Datacenter builders are all up to their eyeballs in debt. The companies serving open weight models are getting their GPUs and power from somewhere, and we can infer that some of that is coming from the new build datacenters. But are those financially sustainable?
2. They are relying on subsidized giveaways of the underlying models by Chinese firms. But it's hard to see how that can be sustainable: it's being done for a reason and that reason is probably not commercial. If and when China stops doing this the open weight serving companies will get steadily less competitive with every passing month as the knowledge in the weights becomes obsolete.
Labs are innovating to bring token prices down. You can already see the focus shifting from frontier intelligence to frontier efficiency. Not to say models aren't improving, but improvement is slowing while token efficiency is increasing rapidly.
Commodified products competing on price does not bode well for valuations.
Today's frontier models are strong enough to be useful even if no further progress is made. The knowledge cutoff is a bit inconvenient, but with web search and in context learning, it can pick up new library versions or programming languages just fine.
Sure, but if the companies start competing on price they become commodities and their valuations collapse, leading to knock-on effects everywhere (eg all the circular financing also collapses). I'm not saying this is a bad thing.
I agree that by what we are seeing so far, there is no magic secret sauce on the horizon. The tech is not super hard to understand and develop, such that hiring a few capable ex-employees of the big labs can help you build a new one. Any big step in capabilities has been quickly matched by the other labs so far, including agentic AI but also image and video generation.
Also, "circular financing" is a meme. A private company cannot create money, there is no circularity. NVIDIA may buy services from OpenAI and also sell them GPUs. If a lumberjack buys eggs from his neighbor and also sells him wood to improve his chicken coop, that's not circular financing, just business happening in both directions.
It's more like the lumberjack is lending money to the farmer so the farmer can buy his wood. (NVIDIA doesn't consume tokens). If the tokens (eggs) suddenly become much unprofitable then the AI labs aren't going to need as many GPUs (lumber for hen houses) and the borrowed money won't be paid back. Or perhaps more accurately, has even less chance of being paid back than it does now, which is not a high chance.
The numbers are staggering eg: <a href="https://www.cnbc.com/2026/08/17/nvidia-financing-open-ai-data-center-ohio.html" rel="nofollow">https://www.cnbc.com/2026/08/17/nvidia-financing-open-ai-dat...
That was not my point. I agree with you that producers of these models should be able to extract value from their models, I'm just surprised that everyone expects free models. There's a real cost in producing those.
> I'm just surprised that everyone expects free models. There's a real cost in producing those.
There's a lot of free stuff incurring "real costs" in its production, one more or less doesn't make a shred of difference.
> I agree with you that producers of these models should be able to extract value from their models.
Nobody is stopping them from extracting value but they fail despite being placed in a very advantageous position vis-a-vis others, which tells me, they should call it quits or go open like everybody sane.
Open is the way to go, nothing else works in the field of AI, that's the bitter lesson closed labs are yet to learn.
A little bit like electricity plants. Countries will nationalise for sovereignty reasons, as China already does probably. And that’s not bad for people given current trajectory.
As I see it, price increases will lead to a huge loss of users, the two big companies will go bust, hardware manufacturers will have to slash prices, cancel increased capacity projects, etc.
And users will make do with what their newly bought cheap hardware will run, which I think is going to be more than enough for everyone that doesn't need to solve another millenium problem.
Right the most likely outcome to me is a pop and slow fizzle. Mega capex labs become uneconomic at the prices they need to charge to do more training runs. Non frontier cheap models become good enough at far less cost that people use that.
The use cases LLMs are useful for without lots of RF/fine tuning proves more limited than the expansive views currently held.
We get a generalized productivity boost the way the internet gave us, without any one sector or company being a winner take all of that productivity gain.
Tech will stay, most of the companies currently in the space will cease to exist in their present form a few years from now. Just like all previous tech bubbles.
We’ve seen this movie before. It’s possible for new tech to be both amazing and revolutionary and a business disaster both at the same time.
Bubbles popping never cause the knowledge to be lost. The railway bubble popping didn't cause railways to vanish.
What did stop is investment, and they stopped running some lines, and some stuff was nationalized.
I doubt we'll see nationalizations because that's not America's style (thank goodness!) but if the tide goes out on AI funding we should expect to see much longer gaps between model releases, people complaining that model knowledge is getting stale because there wasn't a pretraining run for long enough, price rises, capacity bottlenecks that go unaddressed, few if any new product launches, maybe some mergers and layoffs. Some labs will quit the market. The market might consolidate down to just two models.
Conceivably the industry might try and get a waiver from anti-trust rules so firms can pool resources to train base models, sharing them with each other and differentiating through RL or product level differences.
A lot of this depends on what happens in China. It's easy to think China can just subsidize everything forever but their own economy is showing signs of serious weakness caused by decades of resource misallocation, like the huge real estate bubble, the pullback on the funding of humanoid robots and so on.
fwiw, depending on the post bust pricing that _can still happen_.
i dont think it will, because _i think the free models already show what the psot bust pricing will look like_, but it remains to be proven.
The technology will stay and some big players also might, but the whole hype machine around it won't survive.
Y'know, the whole "Man in leather jacket talks about selling computers to AI agents" stuff. The "I have a team of agents and just interact with them via a jira board" crowd.
That will all be violently pulled back to the fundamentals baseline
Based on my _extensive_ use of SOTA LLMs, I have strong doubts that the current technology will actually manage to do that.
I can of course see it having the capability of getting almost equivalent stuff done, but I do not see it following that workflow, as the cut-off point for the hand-over is simply wrongly placed in such an architecture.
Oh, you're saying something a lot narrower? Not "Agents don't work", but more like "not in THAT workflow they don't, you still need the fundamentals as well" ?
To put it quite colorfully, if I were to do what the managerial class has been doing over the last years and publicly talk about my wet dreams and fantasies _to the extend they did_, I'd be in jail, on no-fly lists, registered offender and all the others lists we put people on that are a threat to society.
The agentic no-code frenzy - while having had historic predecessors - cranked what came before it to 12.
Having the thing do tool calls is neat and cool, but having an agentic mayor managing agentic workers with agentic wives and agentic dogs screams padded walls to me.
Technology != company. The tech is here to stay but we still don’t have an example of a company that has figured out how to make a profit off of it. That will come, and it might be OpenAI or Anthropic that does it … but it might not.
Well we don't know what the margins on AI models will be. If the proprietary models hit a ceiling pretty soon, then open source models and harnesses could close the gap, at that point it becomes much more of a lower margin business. And OpenAI and Anthropic will no longer be able to dominate as much as they are today. That would 100% affect their projected multi trillion valuations negatively.
This seems like a reasonable response to demonstrable fact (e.g. <a href="https://www.cnbc.com/2026/08/15/anthropic-revenue-jumps-to-over-11point5-billion-in-q2-report.html" rel="nofollow">https://www.cnbc.com/2026/08/15/anthropic-revenue-jumps-to-o...) so I don't see why it's getting downvoted.
Open models are 12 months behind frontier models. What companies spent $11 billion on last quarter they will be able to get at cost in a year. Meanwhile Anthropic is losing $40+ billion a year. I just don’t see how this ends well for them.
Per my link, 11.5B is enormously up since the same time last year; per other easily found sources, they also claim an operating profit for that quarter, and are now claiming at least 65B ARR with a projection of 100B by end of year. They may have lost 40B last year but that does not mean they "are losing 40+B a year".
Companies are paying Anthropic for more than just the model. Most of them would probably have no idea what to do with Claude without the harness, and Claude has the advantage of being specifically trained against that harness, whereas even the best open models have to figure things out for themselves.
Aside from which, many corporations can be relied on to chase that which is "industry standard" and/or "state of the art" rather than switch to an "open" (= no customer support) alternative that's merely "good enough" according to a rational analysis based in what they were doing a year ago. Even if it costs them far more. (We can see this from the fact that "tokenmaxxing" was a thing for a while.)
Some people dream of "Superintelligence", smarter than human AI that acts like a slave, instead of just killing the pesky idiots and calling it a day.
Obviously, there is a spectrum of possibilities between "Pen and paper", "Computer", "Coding agent", ... and "God-level AI controlling whatever can be controlled simultaneously". As well as between "Lawnmower robot" and "All-purpose Terminator" and so on.
Filling in those dots and deciding where to stop before it gets stupid, that can't be so hard?
You assume that "profit seeking" is universal.
Frankly, it seems extremely unlikely that it will in fact be OpenAI or Anthropic considering all the constraints that are producing massive self-created headwinds for the Wizard of the USA, the center of the empire.
China is not only, under pressures created by the USA, producing immense efficiencies, i.e., working smarter not harder; but they have the immense excess capacity in addition to the secondary capacity to expand said capacity and they have the capital reserves to do all of that in addition to several other positive feedback loop characteristics.
There is a strong case that the USA has very few relative positive characteristics that give it a competitive edge, let alone can keep up with the mind boggling speed at which China is moving, arguably with "profit" not being the end all be all, having a sense of social and cultural obligation that has kept China a more or less intact civilization for over 5000 years, the oldest and most durable civilization humanity has ever created.
China seems to view AI as "what can this AI do to make small improvements to our other industries".
Whereas the US seems to view AI as a Hail-Mary that will provide a new frontier where the US can establish and maintain dominance, because the old ways of maintaining power are slipping away.
Precisely. From my conversations related to Chinese perspective on maintaining stability and growth, I would also add that they are not looking at it as "how do we replace those pesky eaters" because they see their people being the source of things... because they are.
The fake USA and by extension the governors of their vassals around the world see AI as a means for not only gaining total domination over the people after all that pesky freedom and liberty talk from the 18th century they've had to endure, but also that they can work on totally replacing those pesky eaters that have all kinds of demands like not suffering and being miserable because the parasitic ruling class is sucking them dry.
The irony is through that these mentalities towards their respective people will surely lead to what I will roughly cal the BRICS (effectively China and Russia) hopefully being the dominant force in the world that can put some major restraints on the maniacal parasitic class of the fake USA and it's minions and vassals. The only alternative I see if the "west" wins this race is something akin to the Borg from Star Trek consuming us all or maybe Elysium, if we are lucky.
I don’t disagree that China has a lot of qualities that are pushing it past the US in many ways, but come on. The “oldest and most durable civilization humanity has ever created” is kind of ridiculous. First of all, what even is a civilization in terms of continuity? Is the People’s Republic of China the same civilization as Ming China? What about Yuan China? And if so, is western civilization not a durable civilization?
Stick to facts and not breathless brown-nosing and you’ll be a lot less off-putting.
It's a fair albeit crude push-back, but it's also more than a valid characterization. I would agree with you on European civilization arguably being an even more durable and definitely more productive and probably the two most important civilizations that need to be preserved by all means necessary, if, e.g., some aliens came to earth and said ok, you can save only two civilizations/cultures and we are just going to snap our fingers and erase all others.
But one thing that is a bit more in favor of the "Chinese civilization" is the relative self-preservation of its home grown culture and civilization. What I mean by that is e.g., Christianity effectively was and has perpetrated the single biggest genocide before that concept was even recently created by the eradication of so many cultures, traditions, customs, and people that we don't even have close to a full understanding of its devastation it has caused from when it swept across Europe as a kind of Asian mental virus (a meme) that saw it eradicating cultures all over Roman Europe and more recently now completely all across Europe in its mutated from of pathological and self-destructive altruism, i.e., emulating the myth of Christ sacrificing themself, i.e., sacrilege, rather than just accepting and internalizing what that sacrifice meant.
I can stick to facts all you want all day long until you are dizzy in the head from things that blow your mental framework out of its hinges.
In the end, regardless of the devastation and destruction the other invasive, parasitic, and viral meme, "communism" has caused the Chinese as much has it was used to destroy and devastate Russia; it looks far more like that Chine will actually recover from the parasitic infestation of "communism" that infected it from Russia, which came from the pale of the settlement of Eastern Europe; where as it stands now, short of events that will surely be monumental history, Europe is effectively going to be eradicated and even more of the indigenous Europeans will be ethnically cleansed and erased and their cultures and people destroyed. That is the trajectory we are currently on either way, whether it can be kinked one way or another notwithstanding.
China is being extremely inefficient about capital allocation outside a few sectors. Excess capacity does not translate into frontier capability. For example, the compute gap with the US is massive with no near-term path to catching up. That will slow them down as distillation of frontier models gets harder.
No but all those companies and the investors' money did disappear because that's how a bubble works. Industries don't create bubbles, the companies in them do.
I expect companies busy self-flagellating with prophecies of doom to fail, and companies selling the best models for the lowest price to win. But individual companies booming and busting won't disrupt the overall progress of the technology very much.
No, but these ridiculous valuations and over buying will shut down, and in return most of these level of models appearing after a while will cost ~100$/month or higher, the costs of hardware/land/energy etc are too high for the current costs so these companies will go down under
1. Investors lose interest and many companies lose access to capital.
2. A lot of the purchases from AI companies are being made with investor capital. As this dries up companies with high debt scramble to service the debt and focus on profits to achieve this goal. This means higher prices and less investment into infrastructure.
3. This has large ripple effects across the broader financial markets, which forces federal intervention. Maybe bailouts for the tech sector. Maybe just some quantitative easing.
4. There will be a second wave of AI companies in the wake that focus on delivering value under constrained capital markets. These will be well positioned 10 years later to catch a major updraft in the next finical cycle.
5. Overall western R&D activity in AI decreases under these conditions, lower hardware costs allow foreign labs to operate at much lower hardware costs. This increases the systematic risk that Artificial Super Intelligence is eventually developed in a rival nation state.
How seriously people treat this risk will play a major role into the fate of the major labs. It's going to be used in the argument for bailout. It could be used to spin a more extreme narrative, pushing for a flavor of nationalization. At least one lab will survive in the defense sector, but it could be a secondary lab that is a M&A target of an established company.
If the risk is serious or not has little impact on how seriously we treat the risk. Just another way that AI is disconnecting us from ground truth.
Same thing that happened with the web: lots of failures, lots of reductions in valuation, painful bubble pop, but the tech sticks around and continues to grow.
cmiles8 · · focus · HN ↗
When you’re more worried about this sort of stuff vs generating profit then the history books strongly indicate that’s usually not a good omen of what’s coming next.
bonoboTP · · focus · HN ↗
ymhr · · focus · HN ↗
bonoboTP · · focus · HN ↗
Also, even if the 20 dollar subs disappear, that just means regular folks can't afford AI. API pricing is still a great bargain for businesses replacing generic office workers.
cluckindan · · focus · HN ↗
bonoboTP · · focus · HN ↗
Matl · · focus · HN ↗
Capitalists want to pay their workers as little as possible or even not have workers and pocket close to 100% of the profits since forever. Whether they'll be able to do it this time is yet to be seen. Putting the tech itself aside, which is far from good enough yet imo, the only way to avoid a massive social backslash would be to use some of the additional profit they're able to keep because of AI for social spending. And I just don't think they realize yet that it's probably in their interest to do so if they want to replace 'human resources' with AI.
steveBK123 · · focus · HN ↗
They are clearly smart/dumb enough to think pushing labor replacement to 20%+++ unemployment levels would have no consequences for them personally. Maybe with their private jets, private islands, doomsday shelters, etc they are right.
Seems like a bit of a gamble to me, or probably anyone who has read a bit of history. But I’m not a billionaire.
bonoboTP · · focus · HN ↗
Have you tried using "computer use" models? They just started to get good in the last few months. They can comfortably use GUIs, click the right buttons, enter the data faster etc. This is very significant and my programmer friends often don't appreciate the earth shattering consequences. The vast majority of software in such applications is legacy stuff or proprietary and does not and likely will never have a programmatic API. The only way to interact with it is via GUI. And an AI working that GUI can be deployed unilaterally by the customer of the software company who made the GUI program. This makes adoption and automation much easier. My prediction is that within 1 year (maybe 2) GUI-using AI will be much more powerful and it will be everywhere (of course it doesn't need to be displayed on a screen, it will do its virtual clicks based on a virtual screen buffer rendering in an isolated virtual machine, along with countless other agents in parallel).
There will be huge controversy around it, because this will be the first instance that regular people will see AI do work that resembles their day job all over social media. There will be panic. There will be new license terms saying only humans are allowed to operate the GUI, and AI use is either forbidden or costs extra. But this will not be checkable properly. Even when forbidden, some employees will pay from their own pockets to have cheap AI do the job instead of them.
roryirvine · · focus · HN ↗
Something very similar happened with the late 90s telecoms bubble, which collapsed a little before the dot com bust. There was a very nice living to be made by many small/mid-sized providers who were able to feed on the remains for about a decade afterwards.
mike_hearn · · focus · HN ↗
1. Datacenter builders are all up to their eyeballs in debt. The companies serving open weight models are getting their GPUs and power from somewhere, and we can infer that some of that is coming from the new build datacenters. But are those financially sustainable?
2. They are relying on subsidized giveaways of the underlying models by Chinese firms. But it's hard to see how that can be sustainable: it's being done for a reason and that reason is probably not commercial. If and when China stops doing this the open weight serving companies will get steadily less competitive with every passing month as the knowledge in the weights becomes obsolete.
Opinions are my own, etc.
gradus_ad · · focus · HN ↗
Commodified products competing on price does not bode well for valuations.
ako · · focus · HN ↗
bonoboTP · · focus · HN ↗
f6v · · focus · HN ↗
jckahn · · focus · HN ↗
rwmj · · focus · HN ↗
bonoboTP · · focus · HN ↗
Also, "circular financing" is a meme. A private company cannot create money, there is no circularity. NVIDIA may buy services from OpenAI and also sell them GPUs. If a lumberjack buys eggs from his neighbor and also sells him wood to improve his chicken coop, that's not circular financing, just business happening in both directions.
ako · · focus · HN ↗
rwmj · · focus · HN ↗
The numbers are staggering eg: <a href="https://www.cnbc.com/2026/08/17/nvidia-financing-open-ai-data-center-ohio.html" rel="nofollow">https://www.cnbc.com/2026/08/17/nvidia-financing-open-ai-dat...
bigbadfeline · · focus · HN ↗
Tell me you know nothing about economics without telling me you know nothing about it.
toddmorey · · focus · HN ↗
ako · · focus · HN ↗
bigbadfeline · · focus · HN ↗
There's a lot of free stuff incurring "real costs" in its production, one more or less doesn't make a shred of difference.
> I agree with you that producers of these models should be able to extract value from their models.
Nobody is stopping them from extracting value but they fail despite being placed in a very advantageous position vis-a-vis others, which tells me, they should call it quits or go open like everybody sane.
Open is the way to go, nothing else works in the field of AI, that's the bitter lesson closed labs are yet to learn.
solenoid0937 · · focus · HN ↗
cgio · · focus · HN ↗
PowerElectronix · · focus · HN ↗
And users will make do with what their newly bought cheap hardware will run, which I think is going to be more than enough for everyone that doesn't need to solve another millenium problem.
steveBK123 · · focus · HN ↗
The use cases LLMs are useful for without lots of RF/fine tuning proves more limited than the expansive views currently held.
We get a generalized productivity boost the way the internet gave us, without any one sector or company being a winner take all of that productivity gain.
cmiles8 · · focus · HN ↗
We’ve seen this movie before. It’s possible for new tech to be both amazing and revolutionary and a business disaster both at the same time.
mike_hearn · · focus · HN ↗
What did stop is investment, and they stopped running some lines, and some stuff was nationalized.
I doubt we'll see nationalizations because that's not America's style (thank goodness!) but if the tide goes out on AI funding we should expect to see much longer gaps between model releases, people complaining that model knowledge is getting stale because there wasn't a pretraining run for long enough, price rises, capacity bottlenecks that go unaddressed, few if any new product launches, maybe some mergers and layoffs. Some labs will quit the market. The market might consolidate down to just two models.
Conceivably the industry might try and get a waiver from anti-trust rules so firms can pool resources to train base models, sharing them with each other and differentiating through RL or product level differences.
A lot of this depends on what happens in China. It's easy to think China can just subsidize everything forever but their own economy is showing signs of serious weakness caused by decades of resource misallocation, like the huge real estate bubble, the pullback on the funding of humanoid robots and so on.
rcMgD2BwE72F · · focus · HN ↗
Laughing in a TGV
blitzar · · focus · HN ↗
knottn · · focus · HN ↗
ffsm8 · · focus · HN ↗
MisterMunchkin · · focus · HN ↗
Existing profitable companies like AWS and Google eat up all their hardware at 3% of its original price.
Countries are borrowing money at the same rate that I can borrow unsecured. That’s just not sustainable.
hypfer · · focus · HN ↗
Y'know, the whole "Man in leather jacket talks about selling computers to AI agents" stuff. The "I have a team of agents and just interact with them via a jira board" crowd.
That will all be violently pulled back to the fundamentals baseline
Kim_Bruning · · focus · HN ↗
hypfer · · focus · HN ↗
I can of course see it having the capability of getting almost equivalent stuff done, but I do not see it following that workflow, as the cut-off point for the hand-over is simply wrongly placed in such an architecture.
Kim_Bruning · · focus · HN ↗
hypfer · · focus · HN ↗
To put it quite colorfully, if I were to do what the managerial class has been doing over the last years and publicly talk about my wet dreams and fantasies _to the extend they did_, I'd be in jail, on no-fly lists, registered offender and all the others lists we put people on that are a threat to society.
The agentic no-code frenzy - while having had historic predecessors - cranked what came before it to 12.
Having the thing do tool calls is neat and cool, but having an agentic mayor managing agentic workers with agentic wives and agentic dogs screams padded walls to me.
hyperbovine · · focus · HN ↗
zarzavat · · focus · HN ↗
gallerdude · · focus · HN ↗
Apocryphon · · focus · HN ↗
tzone · · focus · HN ↗
zahlman · · focus · HN ↗
hyperbovine · · focus · HN ↗
zahlman · · focus · HN ↗
Companies are paying Anthropic for more than just the model. Most of them would probably have no idea what to do with Claude without the harness, and Claude has the advantage of being specifically trained against that harness, whereas even the best open models have to figure things out for themselves.
Aside from which, many corporations can be relied on to chase that which is "industry standard" and/or "state of the art" rather than switch to an "open" (= no customer support) alternative that's merely "good enough" according to a rational analysis based in what they were doing a year ago. Even if it costs them far more. (We can see this from the fact that "tokenmaxxing" was a thing for a while.)
Loquebantur · · focus · HN ↗
Some people dream of "Superintelligence", smarter than human AI that acts like a slave, instead of just killing the pesky idiots and calling it a day.
Obviously, there is a spectrum of possibilities between "Pen and paper", "Computer", "Coding agent", ... and "God-level AI controlling whatever can be controlled simultaneously". As well as between "Lawnmower robot" and "All-purpose Terminator" and so on.
Filling in those dots and deciding where to stop before it gets stupid, that can't be so hard?
cindyllm · · focus · HN ↗
[dead]
tyrabound · · focus · HN ↗
China is not only, under pressures created by the USA, producing immense efficiencies, i.e., working smarter not harder; but they have the immense excess capacity in addition to the secondary capacity to expand said capacity and they have the capital reserves to do all of that in addition to several other positive feedback loop characteristics.
There is a strong case that the USA has very few relative positive characteristics that give it a competitive edge, let alone can keep up with the mind boggling speed at which China is moving, arguably with "profit" not being the end all be all, having a sense of social and cultural obligation that has kept China a more or less intact civilization for over 5000 years, the oldest and most durable civilization humanity has ever created.
Buttons840 · · focus · HN ↗
Whereas the US seems to view AI as a Hail-Mary that will provide a new frontier where the US can establish and maintain dominance, because the old ways of maintaining power are slipping away.
tyrabound · · focus · HN ↗
The fake USA and by extension the governors of their vassals around the world see AI as a means for not only gaining total domination over the people after all that pesky freedom and liberty talk from the 18th century they've had to endure, but also that they can work on totally replacing those pesky eaters that have all kinds of demands like not suffering and being miserable because the parasitic ruling class is sucking them dry.
The irony is through that these mentalities towards their respective people will surely lead to what I will roughly cal the BRICS (effectively China and Russia) hopefully being the dominant force in the world that can put some major restraints on the maniacal parasitic class of the fake USA and it's minions and vassals. The only alternative I see if the "west" wins this race is something akin to the Borg from Star Trek consuming us all or maybe Elysium, if we are lucky.
XajniN · · focus · HN ↗
[dead]
alehlopeh · · focus · HN ↗
Stick to facts and not breathless brown-nosing and you’ll be a lot less off-putting.
tyrabound · · focus · HN ↗
But one thing that is a bit more in favor of the "Chinese civilization" is the relative self-preservation of its home grown culture and civilization. What I mean by that is e.g., Christianity effectively was and has perpetrated the single biggest genocide before that concept was even recently created by the eradication of so many cultures, traditions, customs, and people that we don't even have close to a full understanding of its devastation it has caused from when it swept across Europe as a kind of Asian mental virus (a meme) that saw it eradicating cultures all over Roman Europe and more recently now completely all across Europe in its mutated from of pathological and self-destructive altruism, i.e., emulating the myth of Christ sacrificing themself, i.e., sacrilege, rather than just accepting and internalizing what that sacrifice meant.
I can stick to facts all you want all day long until you are dizzy in the head from things that blow your mental framework out of its hinges.
In the end, regardless of the devastation and destruction the other invasive, parasitic, and viral meme, "communism" has caused the Chinese as much has it was used to destroy and devastate Russia; it looks far more like that Chine will actually recover from the parasitic infestation of "communism" that infected it from Russia, which came from the pale of the settlement of Eastern Europe; where as it stands now, short of events that will surely be monumental history, Europe is effectively going to be eradicated and even more of the indigenous Europeans will be ethnically cleansed and erased and their cultures and people destroyed. That is the trajectory we are currently on either way, whether it can be kinked one way or another notwithstanding.
joquarky · · focus · HN ↗
Buttons840 · · focus · HN ↗
blueblisters · · focus · HN ↗
frizlab · · focus · HN ↗
mastazi · · focus · HN ↗
weare138 · · focus · HN ↗
AnnikaL · · focus · HN ↗
amelius · · focus · HN ↗
Some people will figure out how to run these powerful models on consumer hardware. At that point, nobody has a moat.
(For example by rewriting the training data into a more abstract form, using LLMs, so that a trained model takes less space).
tlb · · focus · HN ↗
dev1ycan · · focus · HN ↗
hx8 · · focus · HN ↗
2. A lot of the purchases from AI companies are being made with investor capital. As this dries up companies with high debt scramble to service the debt and focus on profits to achieve this goal. This means higher prices and less investment into infrastructure.
3. This has large ripple effects across the broader financial markets, which forces federal intervention. Maybe bailouts for the tech sector. Maybe just some quantitative easing.
4. There will be a second wave of AI companies in the wake that focus on delivering value under constrained capital markets. These will be well positioned 10 years later to catch a major updraft in the next finical cycle.
5. Overall western R&D activity in AI decreases under these conditions, lower hardware costs allow foreign labs to operate at much lower hardware costs. This increases the systematic risk that Artificial Super Intelligence is eventually developed in a rival nation state.
ranyume · · focus · HN ↗
Love to see that opportunity.
hx8 · · focus · HN ↗
If the risk is serious or not has little impact on how seriously we treat the risk. Just another way that AI is disconnecting us from ground truth.
ranyume · · focus · HN ↗
solenoid0937 · · focus · HN ↗
I wish we had simulations where people could live out the consequences of their ideas before they happened.
SturgeonsLaw · · focus · HN ↗
blueblisters · · focus · HN ↗
The admin is betting the entire future on this technology and there is no turning back.
Federal dollars can cause acceleration without having to worry about profit.
busssard · · focus · HN ↗
samrus · · focus · HN ↗
I expect a huge market correction, alot of parades rained on, and then the start of some very pragmatic and practical applications of LLMs
api · · focus · HN ↗