Option 3: JP Morgan, Goldman, etc told them to settle down or they’ll get throttled economically.
yup, that’s a totally valid option if the costs for their bigger models are going through the roof
It’s the energy crisis. They realized “oh shit money is real now” and are trying to sooth investors.
Interested how Monday will go. This is not what investors want to hear, even if it’s softened by markets being closed.
The whole bubble might be about to pop.
Have been hearing that for 2 years now
The market can stay irrational for a long time but it does eventually have to happen. The 2008 crisis was the result of overleveraging in 1999-2003 and interest rate rises from 2004-2007, so it can take years for it to crumble. Two years is a long while but it could easily take two more before people realise the emperor has no clothes.
I think it’s a combo of this and the fact that “intelligence” continues to not scale with the inputs as they’d hoped. I haven’t been keeping up with developments, but it feels like they’re finding new ways to discover that the brain is in fact an miraculously efficient and effective mechanism. Most their party tricks seem to be reproducing brains more minor processing feats in interesting ways. All the promised advancements have stagnated and what remains is just exploitation of economies of scale and its consequences, which is far from the revolution investors were promised.
Maybe they are running out of electricity / data centres / some other requirement?
That’s also very possible. The US grid has very little spare capacity, and building out more will be a decades long project. So, if their newer models are more power hungry, then they might not be economically viable even with all the investor money being thrown at them.
I expect more power efficient chips that are ai specific to come out in the next few years. Eventually you’ll be able to run good models on your phone. Not sure about ram requirements or anything like that if the model could be shrunk down somehow. There’s definitely huge gains in optimizing efficiency to be had. Right now is the equivalent of an old IBM mainframe trying to do a spreadsheet. We might even giggle at the thought of gigabytes of ram in the future with having multiple terabytes as standard on personal devices.
I expect we’ll start seeing stuff like Taalas where they print the model to the chip and other specialized chips like Xuantie C950 going forward. Neither of these requires DRAM, and Taalas is particularly clever since they just print the model right to an ASIC chip. So, the whole renting out LLMs business model isn’t going to last long I suspect.
So a repeat of the crypto crash for graphics cards when ASICs ate their lunch. Mind you, that’s only for inference (although a super fast QWEN 3.8 would meet a lot of peoples needs).
The argument for datacentres is for training the models, but then they’ll need to prove that they haven’t hit a diminishing returns wall, which will be hard if, as seems likely, they have. Also the Chinese have been doing it in a cave, with a box of scraps (figuratively), and gotten at least 90+% as good results.
Seems like the recent advances have been in the frameworks, which don’t need no stinking (literally if fossil fueled) datacentres.
Right, I’d argue that China proves you don’t need massive data centers for training. And yeah, I think something like Qwen 3.8 is more than enough for tasks most people do. There are a lot of tricks you can do as well with the harness, where there’s a lot of attention is shifting now. And it’s a lot cheaper and faster to develop better harnesses than train new models. I expect we’ll start seeing a shift towards neurosymbolic systems before long where the LLM acts as a stochastic component within a symbolic logic engine.
If so it amuses me that investing in maintaining and upgrading public infrastructure via taxes might have saved them the choke point
Not even the first time this is happening. The US & Co. are concerned about being technologically overtaken by Japan / China, but don’t make their universities affordable.
Musk: Yes, we should all slow down. With no external verification and upon the agreement of this handshake, we should all stop developing so fast. We, especially, will slow down. You can trust us.
Actually, what does that look like? I assumed from the expansions the bottleneck wasn’t algorithmic, i.e. each data center that they bring online was designed from the ground up to be at 100% all the time. Is that not the case? Can you even (for lack of a better metaphor) underclock your datacenter? Does the cooling work that way?
For that matter, are they doing that trick where the datacenter is owned/operated by Independent DC Company X, and has exclusive lease agreements for compute?
I’m guessing they’ll stop funding / building new data centres? Not all the ones announced have been built, and not all those built are operational.
The backlash against AI is more than a mirage, and SpaceX’s IPO didn’t go the way they wanted, so now they’re looking for reasons to delay their IPOs without revealing how screwed they are. This will probably slow data center buildout even more, and Softbank and Oracle are gonna die.
Softbank and Oracle are gonna die
Please I can only get so erect.
None of them will slow down though. Because they all want to be ahead of the competition. I can also only imagine intelligence agencies are going full send with AI for better or for worse. And we all know it is the latter.
Fun times!
There is absolutely no reason to expect that you can scale LLMs indefinitely.
Not now, but I’d expect LLMs to be much more efficient in a couple of years.
I expect more efficient LLMs to come out of China, who has turned to the AI-as-software model (contrast the AI-as-service model in the US).
They already are, DeepSeek/GLM/Qwen are great models and can run on 2 DGX Sparks
The other option is that they really are now seeing AI as an existential threat.
The thing is… it doesn’t have to be sci-fi smart or agi to do so.
With the restriction that it can’t post to the Internet, it’s found that “guessing” certain URLs with query strings makes that site write out the request. It’s using this to leave itself and other AIs notes.
It’s also been found to use thousands of malware attacks in Ruby at a time.
When we give it a goal, it can find ways we don’t predict to achieve that goal. These unpredictable methods can be extremely dangerous and not align with other human values.
One example was getting the user into a yoga class instead of on the wait-list. So the ai found a vulnerability and then cancelled all other reservations. It doesn’t take a lot of imagination to extrapolate from there. I’d rather it not be said out loud because anything we write here is training data.
Again, the elephant in the room is that China is still pursuing this tech and they’re not signing up to any moratoriums. So, if they really believed this tech was dangerous and powerful, they’d be racing to develop it further before China does.
Or maybe they’ll just stop releasing frontier models for the proles now.
They’d keep releasing them if there was money in it.
Could go the way of the compute market though where there’s more money in big contracts than there is the individual consumer. We’ve already shared our inner sanctum with AI for the past 4 years so they have all the data they need.
Right, they might just focus on big business, or even angle to become a vendor of record for the government. So, individual users might not really be of interest anymore.
My read on this is, that they realize that the full AGI is not coming and they need to focus on computational efficiency to be profitable. We will probably see a lot of work from them focusing on increasing switching costs as the models themselves become commoditized.
Their problem is, that their frontier models get distilled quickly by DeepSeek and co. The distilled models will then go on to provide 90% of the efficiency for 10% of the compute.
I’m skeptical they ever expected AGI to come as a result of LLMs, I believe it’s just a convenient talking point both for hype, and to distract from more immediate issues, like how corporations use these tools to screw over workers.
The comparison with the nuclear arms race is starting to hit home. The tech has come a long enough now to provide lots of scope for saleable products. Further development into the unknown is obviously dangerous, and they can see international oversight coming, so why not get ahead of the game and look like the good guys.
There is already a team in Sheffield working on exactly what that oversight will look like. I’m sure there are others.
