You must log in or # to comment.
Everyone saying this is different than the dot com bubble: Yes, I know LLMs aren’t 100 percent hot air and there are legitimate use cases for them, but the internet also wasn’t useless and it didn’t go away. It was just overhyped. It’s still an apt comparison
It is, but I think we are a little bit past the point of the froth of that time. The mania is simpler, but the amount of capital going at it just has to be a lot bigger.
Most of the capital is going to the big 7 though, and they spend a lot of their own money on it as well. To me they are just chasing this pipe dream of AGI, which they are nowhere near to because their approach is fundamentally flawed. I genuinely believe AGI is possible, but not in the way they are going about it. That is actually a very good thing, bubbly as it is at the moment, because in their race towards AGI, which would be a true game changer, they are creating a black box with all the safety measures removed and even those companies don’t know what the hell it does or how the hell it does it exactly.
The only person who I think might manage to build AGI in our lifetime is Song-Chun Zhu, because he is the only person who has an approach that makes sense.
What is Song-Chun Zhus approach? I couldn’t really find an answer to that with a quick search.
This article doesn’t go into great detail about his work (it’s more about his life) but does touch on the notions of it: https://www.theguardian.com/news/ng-interactive/2025/sep/16/song-chun-zhu-why-one-of-the-worlds-most-brilliant-ai-scientists-left-the-us-for-china
Basically he states the big data approach is fundamentally flawed because the LLM’s cannot reason at all. Everything it generates is probabilistic. So the systems and robots he builds are built to be “aware” of their surroundings and context, not relying on a dataset, but on ad-hoc input. An example cited in the article is the Tongtong 2.0 robot which uses a pillow to give it some extra height so that it can retrieve a book from a bookshelf otherwise out of reach. It sees the book, realizes it can’t reach it and then finds a solution so that it can.
Also need to add that there is so much in the field of AI and machine learning that AREN’T LLMs
Lol wut? LLMs are a joke. The Internet is useful.
No bailouts this time.
Narrator: They were bailed out
Hoo boy, it’s going to be the bailout of bailouts.
Well at least it isn’t going to be our taxes doing it…
Ah, so, about that…
There is always bailouts, the rich will be saved while poor are left to suffer
https://projects.propublica.org/bailout/
The bailouts are, by the numbers, workable. There is an argument that a bailout would be fiscally responsible.
There’s also an argument that the whole thing should have been given up as a bad job decades ago.
Good arguments aren’t going to decide how it’s handled though, so let’s not bicker. Life’s too short.
LOL. Trump loves bailouts.
The rich have failed to use AI to pivot their wealth into global automated fascism.
Doesn’t mean they will stop trying.
It only stops when we redistribute their wealth or they die.
Redistribution of wealth is a temporary solution if we don’t destroy the underlying systems that allow for such individuals to exist.
I guess we’ll all have to die, then. Can’t let billionaires suffer any consequences!
They have, however, made great progress creating tools that destroy consensual reality. Which was the point all along.
Mmmm, less the outright point, more a desired side effect. It helps keep the masses in line if they don’t know what’s real, but doesn’t directly concentrate wealth on the scale they want.
That’s a better way of putting the causal chain. That said, ultimately the oligarchs’ goal is to “concentrate wealth,” so having a reality destroying machine is a really accelerant for that process.
It only stops when we redistribute their wealth
orand they die.Fixed that for you.
Ed Zitron has a pretty good analysis of this, pointing out that Nvidia is “investing” in AI cloud providers who then go and buy Nvidia chips. So they’re essentially inventing their own biggest customers.
And then these providers take out big loans using the chips as collateral. So there’s really nothing to go back to Nvidia if they fail.
And none of these are even close to profitable. So they will probably fail.
And Nvidia makes up 7% of the S&P 500 at this point. So if they lose their biggest customers, and their investments, and the marketing hype they’re capitalizing on by propping up these zero-sum chip sales… there’s a potentially large fallout.
That’s not even counting all the other big tech companies that have placed huge bets on AI, and therefore Nvidia. The magnificent 7 all together are more than 30% of the S&P 500. And if Nvidia falls, they probably all fall too.
I wouldn’t bet too hard against NVIDIA. Sure their margins/extortion pricing can go down with fewer customers, but LLMs are here to stay. Datacenter and mediocre models (Open AI) and getting good ROI from buying from NVIDIA is what is nearly impossible. US models tend to all concentrate on megalithic US military Skynet ambitions, and every release is a step towards Skynet. Open models, mostly from China, tend to be smaller (in GPU/memory requirements) but have better quality/cost ratios including use on accessible (non datacenter) hardware.
It’s the datacenter gpu customers, and mediocre software/llms renting/owning them, that are the huge risk. At the same time, US empire bankster allies will invest for Skynet.
None of the AI providers actually make a profit on their searches and the marginal cost per user and search isn’t dropping.
LLMs don’t benefit from economies of scale. Usually, each successive generation of a technology is cheaper to produce, or stays the same but with much greater efficiency/power/efficacy/etc. For LLMs, each successive generation costs much more to produce for lesser and lesser benefits.
LLMs don’t benefit from economies of scale.
For training, compute and memory scale does matter, including networked large scale clusters (of GPUs). No money is made in training. Inference (where money is made/charged or benefits obtained), memory more important, but compute still extremely important. At Skynet level, models over 512gb are used. But consumer level, and every level smaller models are much faster. 16gb, 24gb, 32gb, 96gb, 128gb, and 512gb are each somewhat approachable. But each of these thresholds are some version of scale.
each successive generation of a technology is cheaper to produce, or stays the same but with much greater efficiency/power/efficacy/etc.
The roadmaps for GPU makers are, well for nvidia only for simplicity, Rubin will have 5 times the bandwidth, double the memory and at least double the compute. For what is likely 2x the cost, less than 2x the power. A big issue with bubble status is a fairly sharp depreciation in existing leading edge devices. Bigger memory alone is always a faster overall solution than networking/connections.
For LLMs, each successive generation costs much more to produce for lesser and lesser benefits.
Bigger parameter models are slower for same training data sets than smaller parameter models. Skynet ambitions do involve ever larger parameters, and sure more training data is added rather than any removed. There is innovation in generations on the smaller/efficiency side too, though Skynet funding is for the former.
To play the devils advocate. Someone some where every day has been calling for a massive crash since the last massive crash. Their thesis is based whatever current events are at the time.
That aside. I think the so called AI bubble is distracting from something worse. The west has not been inventing or producing anything in a long time. The prevailing zeitgeist had shifted to a greater-fool pyramid-scheme type mindset. It happened a long time ago. Long before AI.
Everyone believes they’re smart for thinking they can park their money in the right investments and live passively off others doing the hard work. Nobody wants to work anymore.
Sorry (not sorry) if that last sentence makes Lemmy users irate like it does on other social media. I don’t know if it does or doesn’t. But I think it’s true.
It’s not just billionaires scooping up all the wealth. People have been fighting each other for the scraps too.
If the so called AI bubble bursts. I think a bigger domino falls. The one where everyone has to face a sobering reality. They all thought they were smarter than each other by trying to get rich off the other person doing all the work. Which has led us to a point where nobody has been doing anything. Not inventing new things. Not producing good product. Not manufacturing anything. Doing nothing but watching imaginary lines go up. This is not just AI but the whole of society has come to be built on this mindset.
Everything has been hyper-capitalized. The kid with a lemonade stand has been trying to figuring out how he can minmax his investment by giving you a tiniest pinch of lemonade powder per cup of water. It’s crazy times. Not just the AI industry.
That aside. I think the so called AI bubble is distracting from something worse. The west has not been inventing or producing anything in a long time. The prevailing zeitgeist had shifted to a greater-fool pyramid-scheme type mindset. It happened a long time ago. Long before AI.
It’s like the NFT bubble, and the crypto bubble before it, and the Metaverse, and VR, and WeWork, gigifcation of everything, etc etc. These bubbly tech products keep booming and busting because the West isn’t actually creating real value anymore, it’s just chasing every latest tech fad and then moving on to the next shiny thing when the previous fad fizzles out.
The everything bubble, basically.
It’s sobering working in engineering and seeing the MBA bros take over.
No long term planning, no pride in the work, and no care about quality, safety or utility.
Just corporate slogans and bullshit.
I feel like MBAs are one of the worst things to happen to industry in general. I know one that is a chore to work with, even though I’d definitely consider him one of the better ones in terms of aiming for quality and long-term stability – he’s just super difficult when it comes to technical understanding and thinks he can tell me what I need for my job.
Me: “I need to document exactly what my users want this thing to do, so here’s a form we can fill out to make my life easier.”
MBA: “Make a spreadsheet with a table, one column per thing it can possibly do, and have people mark what they need. Trust me, that way they’ll do your work for you! You’ll make things so much easier for yourself!”
User: proceeds to fill out the table with garbage that would be useless if I didn’t already have the prototype we’re using as exampleHe also sent me a wonderful visual metaphor for my job, where the components I build from are different lego bricks and the end result is a lego house. No shit, sherlock, but I don’t need people to tell me what bricks they want the house to be built from. I need to know what the house is supposed to look like.
He’s also trying to have AI do part of my job. Not that I’m concerned – it could do half my job and I’d still have too much to do, and I’m pretty sure it can’t even do a tenth as it is. I’m more worried about the AI spitting unexplained garbo and making more work for me when I need to explain why its output is different.
The problems related to over-indebtedness and the high risk levels in Banking after the Glass-Steagal was overturned and which led to the 2008 Crash were never solved.
Instead what was done was the “hack” of lowering interest rates to levels far below the Historical Trend, supposedly temporarily, but in fact they never went back to trend (just like growth itself, fell well below that trend and never recovered back to it). The technique used reduced the costs of servicing the debt (if interest rates are lower, then the interest on that debt that needs to regularly be paid is less, which all things together means the Economy doesn’t need to create quite as much new wealth to service debt) whilst not actually solving the problems of overindebtness.
Further, Finance Industry regulations never went back to what they were under Glass-Steagal and similar - some things were tightened but at the same time the concept of Too Big To Fail banks rose, which means large banking institutions know they will always be saved by the State no matter what, hence take more risks.
Basically, the whole system has taken a turn towards the road to Stagnation under the hope that somehow a path back to growth would be found, and that hasn’t happened and instead what we have is all that cheap money feeding a string of speculative bubbles in things that don’t in any way add to our productive ability - or in other words: to our ability to create wealth - and just blow up after a while having done nothing else than move money around and misallocated resources which were thus wasted, thus pushing us deeper in the road to stagnation.
Pop it.
I reject with the premise.
For good reason, it feels that the only major discussion in markets is whether AI is in a bubble or whether it’s actually the early innings of a revolutionary phrase.
I think everyone knows that AI is over-hyped, and you can call that a bubble if you will.
The question which is unanswerable is how over hyped is it and when will the hype train stop or when will the bubble burst.
Everyone’s pension fund is balls-deep in shares of the magnificent 7. It’s the only game in town and you kinda can’t afford not to be.
Suppose you decide the bubble is going to burst tomorrow, so you dispose of all your mag-7 shares and buy into renewables. Then maybe the bubble doesn’t burst tomorrow, but in 2 years time, by which time everyone else who still has their mag-7 shares has doubled their value.
How many of those “mag-7” are also profiting from the genocide in Gaza?
I don’t feel sorry for war profiteers loosing their pensions.
Not everything is about gaza.
If you are Dutch as your instance might suggest, Pillar II funds are also invested in the Mag-7 whether you like it or not, so welcome to the club of war profiteers.
Don’t worry guys. Sure the economy is kinda sucking ass but we got wars on the horizon. Look at russia, their economy has been growing since the war. Look at Israel, their economy is doing great since they started leveling gaza. So big ol’ Trump has already made up a plan of bringing “peace” to drug cartell infested places. You just wait, some war here and there and we are back to the golden ages without redistributing anything (at least not an home).
I am so grateful for Trump to also remind us europeans that we need to get ready for war. This has inspired a great sense of urgency in my country, essentially killing our peace movement. Sure the genocide in gaza sparked some anti-war sentiment but labeling them as terrorists seems like a strategy that works on most.
(of course i hate everything about all of this)
Aaaaand I just heard from someone that, apparently, it is not yet a bubble. Lol
And they talk about how AGI is going to revolutionize everything. “AGI AGI AGI” DAMNIT
The 2009 crash plus a timely redundancy set-up my first mortgage. Can I smell another market trough coming?
Gen Y anticipating the next once-in-a-lifetime market crash
P.T. Barnum saw this coming.
That’s gonna leave a nasty scar…
It’s an empire killing world war starting bubble
