- cross-posted to:
- usa@lemmy.ml
- usa@midwest.social
- cross-posted to:
- usa@lemmy.ml
- usa@midwest.social
You must log in or register to comment.
I’m guessing the lost funds are bleeding out to energy concerns. Correct? Asides from brief construction contracts for data centers, I’m unsure how one makes such large sums of money disappear.
There is a lot to unpack here, finance is interesting.
Investors spend money to build companies on the expectation that those companies will be worth more money than they spent in the future.
One way they spend that money is on the acquisition of bonds, which are firm guarantees against future debts. Bonds are often part of other financial structures (you’ll see them a lot in a hedge, which is a way to limit liability on an investment in exchange for some upside).
If I’m reading the article correctly, those bonds had a condition where they could be converted to shares at a certain price, so the investment and the hedge were part of the same instrument.
It’s not accurate to say that the money could have been deployed otherwise, one way to think about this is that large banks have very large, often self-issued, credit limits. Companies also have self-issued credit limits, related to the banks.
In this case specifically, some amount of this money is going to come out of the IPO, the 80 billion doesn’t exist in any other way.
tl;dr The 80 billion that ‘vanished’ here wasn’t actually spent on anything, it’s valuation against those bonds. It’s also an exciting headline.
Anyways, I hope this was helpful, these companies are definitely getting a lot of investment (running up their credit cards) in the hopes of a big payout, but this doesn’t have anything to do with private equities current running speculation on data center construction or power consumption, or any of those big problems.
From what I understood it’s not really money “lost”, it’s that the company was valued a certain amount and, with a bit of that hype dying down, it’s now valued 80 Billion less.
I’m not an expert on… pretty much any argument in this thread though, so if someone more knowledgeable can chime in I’d appreciate too.
Still supposed to be valued 1.500.000 Million dollars. Lol.
Goddamm… This could easily have housed every man, woman, and child in this country. But no, no free housing while we squander wealth with our stupid ideas.
Not to mention using all that land and water resources to prop up their AI ambitions.
deleted by creator
Billion is officially the new million.
How have i never seen this before?
This is so on point.
Robin Williams is absolute treasure.
this goes in line with thier aggresive pandering to the govt to be used mass surveillance.
I’m doing my part by constantly asking chatgpt questions without ever logging in or paying them money. Eventually they’ll all go belly up and maybe we can ditch this shit.
80 billion could have bought a whole lot of Chinese EV’s to resale and they could have made a profit.
Forget resale, for 80 billion you could probably get BYD to produce custom vehicles under your own brand.
Oh sure. Too bad the captive market here in the US wouldn’t allow it.
I’m totally spamming this across the thread with copy/vaste.
Oh yes I’ve seen that. He really did have a very high level of effortless explanation for it. Junkies fits perfectly.
A billion here, a billion there. Over some time it might become quite a sum.
I’m totally spamming this across the thread with copy/vaste.
$8B real loss on $16B revenue. ($72B is a bondholder gain (accounting loss) from conversion). $9B in training costs not counted as loss, even though it is worthless whenever next model version is released.
$561B in forced spending on compute over next 10 years, with $750B total compute costs lined up. Structural losses are baked in for a few years.
even though it is worthless whenever next model version is released.
For most version releases I actually doubt that, I’m pretty sure they just keep training the model and release checkpoints every few months. Architectural changes, however, will indeed invalidate the previously trained models, and occasionally they have to do that to keep up with the competition.
But if you do architectural changes even just once a year, that means you’d have to depreciate those training costs over just that one year to be truthful. Meaning they might as well be losses rather than depreciating capital investments.
There is a genuine “training corpus” that is an accumulated asset and grows and feeds new models. The vast majority of training cost, that should be expensed, is the training time compute. Neither of these are expensed. There was about 13 major US model releases in September. At least 7 of these replaced a 1 month old previous model family. Open AI gpt sol went from 5.6 to 6 to 6.1 in about 2 months.
Fast model releases not only lock in structural losses, they juice revenue growth illusions. Model use rate generally drops by 66% 4 weeks after release, because of competition, and a surge of use on release for evaluation.
New models are trained on older models, so even if the older model is not longer at par, without it the new model couldn’t exists
But, when the word guessing program gains consciousness, it will be capable of doing the job of every worker in the world, and eager to do those jobs, and anybody who owns Anthropic shares will be a trillionaire in an economy that totally works and is healthy. So, better invest now!
Musk giving 1 share (about $160) to all US children with a Trump account, will make the next generation cheer for skynet genocide and oppression. “For a while, it added 10-20% to my net worth.”
I’m totally spamming this across the thread with copy/vaste.
Why haven’t they just asked Claude how to make money with their business model? /s
They have.
-
Wait for excited CEOs to fire all their staff.
-
Crank up the prices.
2 only works if regulatory capture in place to prevent cheaper competition from open models.
That’s why they keep trying to play the “AI is so dangerous” narrative and why you keep hearing about “rogue AI agents” which in reality someone had to instruct to break the law, and they probably used some of their no-guardrails internal-use models (so Claude Mythos rather than Fable).
Claude Mythos and Fable are the exact same model, the branding only reflects which one has more restrictive classifiers in front of it during deployment. But the weights are identical.
-
Repeating the DotCom burst.
but worst. because some of those sites provided actual value, “AI” does not, most heavily used by Palantir/AI as mass surveillance and profiling people for bombings or harrasment.
I’m totally spamming this across the thread with copy/vaste.
i wish them ruin, you made my day
Good.
Idk man, to me, it feels like we are further from AGI than fusion reactors actually being hooked up to the grid, considering LLMs are still just glorified autocorrects, needing a ton of prompt injections from their providers. And yet, the AI companies make it seem like it’s right around the corner (obviously, because $$$), whereas to me it feels like we are still at least 50 years away. Maybe AGI will be the next fusion generator chase.
Ironically there was notable news in the fusion reactor space this week as Gemini 4, building on the work of Matt Landreman and GPT-6 Astra a little while ago, expanded counterexamples to Grad’s conjecture which had claimed exact solutions for controlling a key aspect of fusion were impossible.
So fusion may actually be more around the corner thanks to the advances of capabilities of AI which helped prove that the fusion problems aren’t as hard as they were thought to be for decades.
And if you actually believe that, I have an amazing business proposition to you. So, there is a bridge in NY, and I can sell it to you for very little money…
Building on other people’s work to provide examples for something is actually one of the things an LLM can be good for.
Evidently, the only thing it’s good for is providing endless dribble for snake oil salesmen to scam people with. There is an endless barrage of “our latest and greatest magic just solved a problem” and when the dust settles it’s always some combination of exaggeration, lies, misunderstanding, and stolen work.
Even from my very limited understanding of high level fusion related math, I can bet it’s the same case here. At best they’re not lying and they provided a proof for some old math theorem. What I know is that fusion isn’t “one proven math problem away”, and I bet even if they’re not lying it will not help that much if at all.
They did manage to do something different, though. I thought at one point that proof assistants should offer to automatically upload your proofs to a central repository so everyone can benefit from them. But such a system would require many to opt in for it to be useful.
What AI does best is giving you existing solutions. It even adapts them if they don’t quite fit. Because many people now discuss their ideas with AI, ideas may spread the same way as well when the conversations are used as training data.
So no, definitely not AGI but valuable. Though as typical of explosively growing startups the value does not come from the technology but from activity bordering on criminal that lawmakers can’t keep up with.
Because many people now discuss their ideas with AI, ideas may spread the same way as well when the conversations are used as training data.
As a training data it’s possible less then useful, unsupervised data like that is actively harmful. As a central repository of everyone’s private conversations it can be in a way useful, but I can’t understand how anyone could be OK with a private company reselling your most private data.
AGI has been achieved more or less because it just means self sufficient at completing tasks. An LLM is just a component these days because they are Vision-Audio-Language-Models with multi frequency ‘thought’ loops. The next letter is A for action where they just have high frequency outputs for direct computer and robot use.
Just because it is bad AGI doesn’t mean it is not AGI right
See the problem is that people expect AGI to learn on its own. LLMs sorta do with big enough context and writing memories, but… They also don’t. They don’t get better at reasoning or anything, they only get better at solving the tasks they’ve written notes on. Which are themselves often pretty bad notes.
They do learn they use search tools and create long term memories. Mechanically that is just tool use it isn’t magic. Do you need these things as emergent properties for them to be real? That would be cool but I don’t think it takes away what is possible now.
