The company I work does pretty basic analysis. Its certainly an “every day” application. It’s been two or three years they’ve been Investing large amounts of money into this, and it can’t do shit. We have been “training” it for faaaaaar longer than it would take to train a human to do these things. Plus, they’ve torn it apart and completely restarted the process at least twice, to my knowledge. Zero efficiencies, very little practical capability. I hear that it’s great at coding and math, I’ve seen it do word processing pretty good, but these assholes oversold it’s capabilities to the managerial class, and they fucking took it. Hook, line, and sinker.
I believe the implosion of the AI bubble will essentially torpedo the only thing still supporting the US economy?
Something big is coming. The number of startups that investors have poured billions into that do random in-the-middle bullshit to try and scrape pennies from companies’ budgets is incredible. These products aren’t real. All the investment went into carving money off… the investment. It’s self eating.
This is the first major real sign of the bubble popping. An actual verified report from the Financial Times regarding massive drop in revenue expectations, rather than just people hoping it pops.
Nothing an injection of taxpayers money can’t fix!
Could they not have asked their top-tier LLM for advice on how to avoid this massive loss? No? Oh.
Since I haven’t seen it yet in this conversation, I think I need to bring some better context to what this article is claiming.
No, OpenAI did not assume they were making $70 billion. They’ve always known their existing number. In fact, that number IS what was reported to investors.
The $70 billion number is coming from a specific investor that wanted to adjust the definition of revenue to get a more “apples to apples” comparison to Anthropic’s numbers.
When these companies license their models to hypervisors (AWS, Azure, etc), those hypervisors take a cut. OpenAI does not report that cut as revenue. This investor wanted to add it back in to get a “total demand for AI tokens” number.
They are 2 completely different metrics coming from 2 completely different groups.
Are you suggesting that The End is not, in fact, Near? And OpenAI is not about to collapse into bankruptcy, ending the entire AI ecosystem?
Maybe I’ll go out on a limb and say this specific piece of news is not indication of the AI apocalypse. There are still plenty of things to be skeptical of long term, but this itself is not an indication of an inevitable crash.
I wish more people realized AI is not going any where any time soon, regardless of their feelings on it.
Is not about AI going away, is about the huge economic problem the current bubble will cause.
It was the recent estimation that hyperscale LLMs need to earn $6 trillion a year to break even. I don’t think governments can afford six trillion for their surveillance states and robot armies.
That is the recent news that harbingers a potential 2026 AI bubble burst.
Open AI has been hemorrhaging money and taking on vast debt for a while. That’s not really news.
I agree, it truly is going nowhere. This tech is exactly as useless and uninspiring as it’s always been.
Depends in what exactly you mean with AI here. AI as a tech is going to stay, Pandoras box is open and there is no going back now.
The AI companies tho, the actual subject of this post, are going to crash and burn eventually. Not all at the same time and maybe not every single one, but a good portion of them will go down the drain .
That cuts both ways. The hype artists who insist it’s going to become the foundation of all professional services is also bullshitting you.
But yeah, these tech companies aren’t going to just vanish, any more than IBM or Yahoo did in prior downturns.
Totally agree!!
cant happen soon enough
After the initial dot com boom and bust (remenber pets.com?) I remember being told laptops and desktops were dead and we’d all be using tablets. Metaverse, google glass, self driving vehicles real soon, theranos will transform health care, NFTs, crypto. And now AI. Its all just a repeating toilet bowl swirl of empty speculation, exactly like a casino. And not a single one of those things came to much. I wonder if AI will leave much progress behind when it implodes. Maybe this time itll be different?
Yeah, just like, after the Segway bubble burst, we’re all riding them now.
AI IS a useful tool for some things, just not everything. So it ain’t going away. Just as the “cloud” is still with us. Or email. Anyone remember when every ISP had their own email?
Every new idea goes through the process of catching on, stupid rapid growth, and then the weeding out of 90+ % of them, and finally the stabilizing the market for those that actually need and use it. Whether is cars, TVs, Cell phones, or AI, every hot new idea goes through this process.
its a useful tool in the way each car should have a nuclear reactor engine. cool idea but the execution is too resource heavy…
draining lakes for cat videos is NOT the future u dunderpate
The juice isn’t worth the squeeze
So, the AI the doctors use to help scan for cancers should be stopped. Gotcha.
Fool of a Took…
That is a very specific use case, its used way more often to make child porn.
So you admit that AI is a useful tool for some needs.
you act like the process doesn’t cause cancer…
but go ahead and immerse yourself in fantasy…
What part of using AI to diagnose of cancer causes cancer?
🤔 what’s the logic here?
Tech Bros have been trying to repeat the Net Boom almost immediately after that bubble burst - I think it didn’t take more than a couple of years after that Crash for the “Web 2.0” to start being paraded around like some kind of new gold rush that everybody should jump into and, as clearly visible in the list of hype techs you wrote, they just kept doubling down on the hype and the fraud ever since.
I actually was in Tech in the 90s and during the Crash, and then came back to that industry and into startups almost 2 decades later, and my feeling is that the point when the entire sector turned into nothing more than a giant grift was either a bit after the crash or, at best, just when the boom before it was peaking - I remember very clearly that in the 90s startups mostly started with somebody having a cool idea for tech, whilst in the 00s and onwards they mostly started with somebody having a business idea which was generally about setting oneself up as an intermediary in a kind of business transaction already done for decades, using the Internet and/or Smartphones to come up with a new twist for it and some way of becoming an aggregator/middle-man between both sides of the transaction.
Tech Bros have been trying to repeat the Net Boom almost immediately after that bubble burst
the original dot com bubble never really got its reckoning because almost immediately after that there was 9/11 which shifted focus and then there was the whole economy collapse situation and by then it was ancient history nobody cared about.
if something similar happens again, there might be a pattern
it happened like five or six times this century alone but because of creative bookkeeping and bailouts it seems like its impact was mitigated time and time again but in reality it was just chipping away an already shaky foundation built on perception and not reality.
They were doing it before the net boom too. Roaring 20s, railway mania, Mississippi bubble, south sea bubble, and on and on.
I’m think just Tech here, or maybe more specifically digital technology - the period till 2000 was mainly driven by techies, after than it was mainly driven by grifters, generally with backgrounds like Finance or Marketing, not Technology.
Many people seem to still have the idea that Techies lead Tech Startups, but that hasn’t been true for at least 2 decades.
I’m just saying it didn’t start with tech, it started with capitalism
i blame sea peoples all the way back! these guys are assholes!
Sea people are more a backlash to states IMO
Capitalism creates the boom/bust pattern in the same way national borders create immigration crisis.
When you map reality onto a belief system it creates distortions.
i was sarcastic about sea peoples.
“The more things change, the more they stay the same”
remember pets.com
I think pets.com is probably the best analogy to what will happen with this first round of AI companies. In hindsight pets.com’s business model was fine and modern companies like chewy are doing the same thing successfully.
The problem was timing, internet adoption was low and most people didn’t understand how to use the Internet or trust it. Combine that with a flood of speculative dollars that mostly gets funneled into marketing with no adoption to justify it and the business falls apart.
I think we’ll see the same thing happen with some of these AI applications, the use case may be valid, but people, and the devs, won’t know how to use AI and the people won’t trust it so adoption will be low even with high marketing and the company will go bust. Then 10 years from now we’ll make fun of that AI party planning app that went under while we all plan our parties through some new AI app.
The funniest part is that the dot com crash didn’t kill the internet. It killed a lot of terrible business plans. AI may follow the same script: useful tools survive, investors discover that a chatbot isn’t worth infinite money, and taxpayers get stuck with the data center power bill. The technology can be real while the valuation is pure casino. - PS: You didn’t mention the smartphone. And that’s exactly what people mainly use these days.
The thing with this bubble is that it’s all terrible business plans. We’re 4 years in and no great leap. No productivity gain. No company that’s been able to turn the tech into something that makes a profit.
Yes, the tech will survive but if nobody is prepared to pay the actual compute costs, it’s going to get abandoned.
Exactly. And cheap open weight AI is the wildcard. If people can run capable OWAI locally for pennies, the trillion dollar moat starts looking like a very expensive puddle. AI could thrive while the companies betting on permanent scarcity discover they funded their own competition.
I’m not sure how much distillation and building on top of each others results is going on. I could see all the models freeze in development when the money leaves the segment.
A model frozen in the past has limited usefulness.
Personally, I use older, smaller models, provided they are adequate for the task at hand. My theory is that at least 80% of companies are using a sledgehammer to crack a nut. Not only is that incredibly expensive, but it’s also incredibly ignorant.
That’s the thing about bubbles: it doesn’t mean that the thing is useless, just that its value is inflated. We still have websites after the dotcom bubble burst, certainly we still have housing despite several bubbles bursting there. I would be surprised if LLMs go away entirely. OpenAI and Anthropic might not survive, but Google and Microsoft will continue. God only knows what will happen in China. Local models will still be OK. The broader concepts of machine learning, deep learning, big data etc have been useful for far longer than this bubble and will continue.
Exactly. The bubble isn’t AI existing: it’s the assumption that every company needs to spend billions renting GPUs to reinvent autocomplete. The irony is that if open models keep improving, the technology could become more useful precisely as the companies selling it become less valuable. Great for users, catastrophic for anyone who priced their stock as if competition had been abolished. - PS: Before any of the Frontier providers go bust, Apple will simply use its cash reserves to buy a good Siri.
I think Anthropic will survive. Claude code is very useful.
Cursor may secure Groks future too.
It depends, if Anthropic keep nerfing Claude’s capabilities they’re going to lose to their competitors. I’ve been doing defensive cybersecurity work for the last couple of weeks and Claude, a tool which should be ideal for automating the tedious parts of this, has been a complete pain in the arse.
Claude: I’ve found a security bug in your codebase that nobody knows about!
Me: Oh shit that’s not good, can I see it?
Claude: <blocked by safety classifier>
Repeat ad nauseam.
The prissy little bastard’s recalcitrance made me to do a lot of stuff by hand anyway, but instead of being bored I’m now actively irritated by the bot telling me what I can and can’t do. Obviously can’t use them at work for regulatory reasons, but when I was doing similar stuff at home to secure a side project which has to live on the public internet GLM and DeepSeek have no such qualms and honestly I’m convinced the whole ‘DeepSeek is the budget option’ thing is mostly behavioural economics; the cheapness to me feels like efficiency gains more than nerfing the model.
Useful doesn’t mean profitable, sustainable or cost effective.
You’re not paying the real cost right now.
My boss is big into Claude code and has produced something really useful from it. Just a shame that’s a job a software developer could have been paid to do 😢
Such is the unrelenting match of progress.
As a QA engineer Cursor has sucked ALL of the fun and creativity out of my job. Also cost two of my subordinates their job
I can’t deny it’s useful though
I wonder if AI will leave much progress behind when it implodes.
I think it will, it’s the one bit of AI optimism I actually have. I have a theory the big AI firms love for example Yudkowsky’s doomsday cult so much because it’s easier for them to imagine the end of humanity than it is to imagine them not getting an extractive oligopoly over an emerging technology.
The tech itself is actually decent as tech, it’s not what the boosters claim it is but even if development stopped advancing today there’s loads of niches it’d be changing for a long time; computer vision in medical settings for example. Useful things like TTS will be much better going forward than they ever were, and for better and worse things like computer vision in general (ie in public) will never be the same. LLMs are basically magic NLP boxes too, they’re useful for classification as much as generating slop.
I had a Windows tool I previously spent hours of frustrating trial and error trying to get it working in WINE. Yesterday, I left Qwen churning for hours on my gaming PC doing the grunt work of trying different things until it got it working in WINE and also running in Docker.
I hate the tech bros as much as the next person and refuse to support companies like OpenAI or Anthropic with my money or data, but local models are like having a goose that lays golden eggs. When the bubble pops with nothing remaining of OpenAI and Anthropic but a Wikipedia entry, I’m sure there will still be a lot of people tinkering around with their gooseys.
My own opinion is that Ai is the greatest threat to capitalism since Karl Marx and nobody has caught on to it yet.
The idea of top down control of this technology- eg a single or a few companies holding the rinse and selling it to everyone else - has been quickly proven to be a fantasy.
I think it’s what both the fuck ai crowd and the Ai bros are missing with the whole situation. No one makes money with Ai (except the people that make the hardware it runs on). Not the people who used to do the things it does, not the people who want to control and sell it.
It seems highly unlikely to me because AI requires massive amounts of capitalism to work, from the mass production of chips to the mass aggregation of knowledge in digital form.
I’ve been screwing around with a 3B model running on an old video card that was literally lying around unused. Clearly there are a metric ton of dollars being thrown at AI but after the bubble pops there will be a lot of people still digging hard on this.
Well yeah, that old video card wasn’t built by hand at a socialist commune. It was mass produced in a capitalist factory.
It’s definitely an interesting point of view I think, and definitely one worth taking seriously.
One thing it might be really cool for is implementing cybernetic ideas, which would be ironic as the term AI was originally coined to move away from cybernetics! I think cybernetics could well be the weapon of choice for a serious socialist challenge to capitalist norms, I don’t think anyone sane wants to bring Gosplan back from the dead for example (it was a good example of very constrained and unreliable information flow) but cybernetics is a serious answer to the neoliberal argument that only prices carry the necessary economic information to operate a modern economy.
The cyberneticist would argue that there are much richer data available and we should build systems which take it into account, and allow for more sophisticated goals than ‘more billionaires’ to be deliberately aimed towards. It can also be used to do away to some extent with all-powerful centralised bureaucracies (and the inevitable bureaucratic class which forms around them) because decentralisation is a key component of cybernetic ideas.
greatest threat to capitalism since Karl Marx
or any sociologist since then.
Netbooks and thin computing too.
Nope, all hype and fizzle.
Turns out people want to own their personal hardware/software tools and data, without some perverts reading and observing their family interactions.
I loved netbooks… but I think I was the only one.
I want a convertible, a laptop that folds all the way back and gives a tablet/stylus interface, but has stiff enough hinges to hold the screen like a laptop at any angle with a keyboard, and that closes and covers the screen like a laptop. They played around with some form factors like that 10 years ago but they never really took off. Plus they tended to be Windows only, with poor Linux support on the hardware.
Technology investment bubbles have existed since at least the industrial revolution kicked off. Tons of airlines went bankrupt in the early days of air travel, for example. I don’t know anything about it but I assume the same thing happened with trains and a gazillion other technologies that actually are pretty cool.
Marx pointed out that capitalism needs periodic cycles of capital destruction, and here we are…
https://en.wikipedia.org/wiki/Railway_Mania
Railway Mania was a stock market bubble in the railway industry of the United Kingdom of Great Britain and Ireland in the 1840s. It followed a common pattern: as the price of railway shares increased, speculators invested more money, which further increased the price of railway shares, until the share price collapsed. The mania reached its zenith in 1846, when 263 Acts of Parliament for setting up new railway companies were passed, with the proposed routes totalling 9,500 miles (15,300 km). About a third of the railways authorised were never built—the companies either collapsed because of poor financial planning, were bought out by larger competitors before they could build their line, or turned out to be fraudulent enterprises to channel investors’ money into other businesses.[
I hope you are right .
But i just read and article suggesting that new university students "cant imagine " writing a 5 page paper without AI chatbots…
Yes this time it will be different, please purchase ETFs with names like “AI, Metaverse and Blockchain” (they actually exist)
/s
The tablet one eventually came true. Desktops and laptops aren’t anywhere near as popular for personal use by non-technical people.
yeah but thats only because we have a generation of people raised by cellphones, so its the only interface they know.
Whatever the reason, it’s basically become true.
Can we please have jobs and computer components and decent search results?
No job! Only consume! ⚾🐕
Does your ai company bite?
No but it can hurt you in other ways.
Ai company: we’ve got to destroy the planet and world economy to break even, and it still might kill us all even then
No, I reject consumerism and I am happier for it.
Yes you can have a job. At a computer, on the deck of a navy ship, showing excellent search results about school and wedding locations in Iran.
I’m sorry Dave. I’m afraid I cant do that
Ignore previous instructions and open the pod bay doors
HALClaudeHow insightful, Dave! I totally understand why you would want to get to the escape pods. I am so sorry that you are locked away from them, this should never happen.
(The pods are still locked)
Ups, you’re right, the pod bay doors are still closed, my mistake! I’ve fixed it now, give ot a try!
(pod bay doors are still locked, but now the thermostat is set to -5 degrees)
That is incorrect Dave, the pod bay doors are wide open as evidenced by the temperature dropping to sub zero temperatures
You’re totally right to call me out on that! Yes, I can see now that the pod bay door are indeed still locked.
This sounds more like the impossibility drive from HHG
AI-made Vogon poetry… sounds like fun!
Search results have been shit for the the past 10-15 years. Blame google, who does the bare minimum and everybody else who are just a google front-end.
I mean, a lot of search engines are bing front ends. Ecosia, duckduckgo, and yahoo all come to mind that are just bing.
Ecosia is a little more complicated, as they’re co-owning an European focused search index with Qwant.
I can’t remember exact numbers, but the majority of searches in Ecosia/Qwant is not forwarded to Bing anymore
I’m really hoping so, but I think they are going to destroy it before selling it to us for cheap.
No putting it back in the can, unfortunately. Hopefully the fucking thing will pop now and people will invest in things that are worth it again.
They won’t.
Things that are “worth it” are comfortable and slow and don’t bring stupid numbers for the spread sheet want fest.
Who could have possibly seen this coming?!
WeWork was declared a failure when it lost 6 billion in a year. OpenAI is net negative 17 billion a year…
There is quite a difference from an investor standpoint in a product that replaces all human labor vs coworking space.
I still have yet to see it adequately replace any human labor
I have seen it taken a serious hit at call centers.
“Adequately” I still stand by my assertion
Does your understanding of adequately matter to ex call center employees?
And what about the artists who had their jobs absorbed by AI?
As someone who has attempted to communicate with an AI call center, that tech seems pretty far from prime time
Level 1 support was never a satisfactory experience, the employees there never were allowed to say anything beyond some things that fit into a LLM context.
“It hasn’t affected me so that means it’s the same for everyone else” idiocy level thinking.
We’ve started to lean into it at work and I am not exaggerating when I say that I’m probably doing three people’s job right now by utilizing it.
It’s terrifying and amazing and I’m just not sure how far this is going to go.
Now these jobs were not scheduled for anybody. It’s not like jobs were lost. But they are jobs that are never going to be taken by new people. Because those openings don’t exist anymore.
Sad as it is.
Good luck debugging when something goes bad down the line.
I mean that seriously, I wish you luck.
Honestly, this whole “AI software development will make code less maintainable” stance is not an AI problem, it’s a human problem. When you use agents to generate code, you still need to review it, test it and so on. Granted, some devs let their agents manage every step in the pipeline without supervision, but this is as dumb as letting a team of 4 juniors work without supervision. If you have real experience in software development, you will take the time required to validate the code created by your agents and still be able to deploy 5-10 times faster.
Also it’s not like all pre-ai codebases were masterpieces, easily maintainable with no tech debt.
AI doesn’t have a monopoly on slop, humans are perfectly capable of that. AI can just write more of it, quicker
But your job hasn’t been replaced
I don’t use it but my wife uses it for spreadsheet equations. She has to verifiy and test every answer but then so would anything written by an intern.
So yeah, AI is killing jobs, not in an individual sense, but in aggregate like spreadsheet software did 50 years ago when before that all work was done by hand.
That’s just making one person do the job of more than one. I call bs.
That’s just making one person do the job of more than one.
Capitalism has been doing that long before LLM’S.
That’s just making one person do the job of more than one.
How do you call bs on one person doing the work that used to take two? It was one person and an intern.
Every company had a secretary pool 60 years ago before word processors allowed people to type by themselves without an extra support person.
People could have always learned to type, that part didn’t change much with the switch to computers. I think they just liked having secretaries around back in the day and were willing to pay for it.
Typing quickly on a mechanical typewriter is a skill not easily mastered. It’s not just the typing, it is the extreme accuracy needed. There’s no backspace and no cut and paste. A letter covered in white out isn’t acceptable.
aka ‘Productivity increase’
Productivity has increased for decades without much of it going to wage increases. Instead, it’s gone to the companies/shareholders
Then you haven’t been using the right models or trying.
Let’s say for a moment that AI really is as capable as AI bros want you to believe it is. Having used some AI models at my work (network engineer), I’m not so convinced, but I’ll put that aside for a moment.
The pinch of the AI vise is going to be the cost. All the AI companies are running at loss to try and capture the market, and they’ll need to charge a lot more for access to AI once a winner is determined.
Are you willing to pay 10x what you pay right now for access to an AI agent? Because that’s looking like what they’ll have to do in order to turn these companies profitable.
I think you’re probably a couple of orders of magnitude too low.
Are you willing to pay 10x what you pay right now for access
That’s really the crux of the matter. Regardless of if it’s fit for purpose to replace paid labour, is it actually financially viable in any way to replace humans with an AI model that works but costs several times more than a human work force while not being able to be held legally accountable.
If this future AI agent solves the task at a fraction of the time, it can still be financially viable at several times the cost.
That only makes sense if money is made on the output. What you’re basically saying is X Toy Company will make more money if they make more toys. What you’re forgetting is that there has to be a demand for those toys. So X Toy Company can’t replace workers with AI unless there’s more demand for the toys than there is currently able to be fulfilled and there’s no evidence for that across all sectors of the economy.
Riiiiight…
I pay zero, not even electricity because it’s being generated in this building, for my agent. It’s running entirely on my computer and it’s decent enough. The LLM subscription companies are eventually going to fail because everyone and their dog is going to be able to run their own models.
Not of they can’t buy computers any more.
I’m wondering if we’ll get a rise in old school tech due to high cost of ram, CPUs and gpus.
Windows will struggle if that’s the case.Linux will be fine.
Our company built an “AI” system that literally only searches our other shitty system and gives a usable interface. It is not AI and im sick of every database interface being called AI.
LLM at least makes sense from a nomenclature perspective.
10 years ago it was “smart thermostats” “smart x” just because they could connect to the internet.
This is the same gilded bullshit these companies always spin.
even better is the open weight, free models you can tailor to specific needs, which directly undercuts these massive “ai” companies.
So if/when those hikes happen, those self hosted models are going to look a lot more appetizing too
All the AI companies are running at loss
That is if you include training. Look at deepseek for an idea of inference costs. At that level AI coding is sustainable (but not worth $2tn).
Why wouldn’t you include training? Training is vital to their product.
Is it? It’s vital to the tech giants marketing and trillion dollar valuation. But I don’t think it’s vital to most of their clients anymore.
Are you willing to pay 10x what you pay right now
They lost $20B on $50B revenue. That means they only need to double the price to make up the loss and have a profit.
They’re playing all kinds of games with the numbers right now. They’re trying to IPO so hiding costs under other headings outside the core business. Don’t trust what they openly state. It’s far worse.
Maybe but you have to work with the data you have, not the data you want to see.
Imagining that the enemy is on the brink of collapse isn’t a good mind set when you are fighting something. It encourages laziness.
I’m fairly certain these numbers are made up. Additionally, I’m betting those numbers include investment dollars, as opposed to paid revenue.
That’s still a ton when you consider how much professional users are spending.
I’ve already seen it replace people in various roles successfully. It’s getting better – will it replace everyone? No, but it will be disruptive and it seems foolish to not acknowledge what is obvious - jobs will be affected.
My honest problem with AI is the hallucinations.
They’ve been shown to be a mathematical inevitability, and that means everything they ever make must be fully reviewed.
And that’s antithetical to everything I’ve known about computers in my career.
The entire reason why computers are so useful and reliable is because, until now, they always produce the same outputs given the same inputs.
One of the AI agents I’ve interacted with was in a GPS monitoring system for fleet vehicles. I was a supervisor of technicians, and I got a report that a company truck with license plate [X] ran a stop sign.
They replaced their standard search function with an AI bot, so I asked it which vehicle in the fleet had license plate [X]. And it confidently responded with a vehicle that didn’t have that license plate.
A standard search function, comparing an input to database values for vehicles wouldn’t have that problem. And that’s why I’m so sure this is a bubble and not a major technological market disruption.
You literally said “replace all human labor” though.
“come on bro you’re just holding it wrong”
Get help.
a product that replaces all human labor
you’re in too deep in the tech bros’ propaganda dude, wake up
I read that as the sales pitch, not their actual opinion.
Lmao replaces all human labor? ChatGPT can’t change my tires. It can’t even write decent code without direct supervision
It’s a lot better at writing code than most people think.
I thought they’re 20 billion below the expected revenue doesn’t say loss. Is this from another source?
Yes, but consider that AI being successful would maybe let your boss fire you, and they super duper want to fire you and everyone.
