![](/static/253f0d9/assets/icons/icon-96x96.png)
![](https://fry.gs/pictrs/image/c6832070-8625-4688-b9e5-5d519541e092.png)
No, I’m not db0, but thanks for the compliment. 😅
I’m way too lazy (and broke) to set up and admin a Lemmy instance.
No, I’m not db0, but thanks for the compliment. 😅
I’m way too lazy (and broke) to set up and admin a Lemmy instance.
We’re talking about LLMs. They’re useless for most practical applications by definition.
And when they’re not entirely useless (basically, autocomplete) they’re orders of magnitude less cost-effective than older almost equivalent alternatives, so they’re effectively useless at that, too.
They’re fancy extremely costly toys without any practical use, that thanks to the short-sighted greed of the scammers selling them will soon become even more useless due to model collapse.
If they still profit from it, no.
Open models made by nonprofit organisations, listing their sources, not including anything from anyone who requests it not to be included (with robots.txt, for instance), and burdened with a GPL-like viral license that prevents the models and their results from being used for profit… that’d probably be fine.
If you do crime big enough, it becomes good.
No, no it doesn’t.
It might become legal, or tolerated, or the laws might become unenforceable.
But that doesn’t make it good, on the contrary, it makes it even worse.
OS vendors aren’t selling¹ what users copy into the clipboard.
¹ Well, Microsoft probably is, especially with that recall bullshit, and I don’t trust Google and Apple not to do it either… but if any of them is doing it they should get fined into bankruptcy.
It’s an LLM. Odds are it’s hallucinating the sources and they don’t even exist.
Know what does compile sources for you which are guaranteed to exist and be related to what you’re looking for…? A good old not LLM infected search engine.
If my plumber replaces their wrench for a rabid gerbil claiming it’ll be just as good I’m definitely changing plumbers.
The good thing about that is that this kills the LLMs, since new models can only be trained on this LLM generated gibberish, which makes the gibberish they’ll generate even more garbled and useless, and so on, until every model you try to train can only produce random useless unintelligible garbage.
Generative AI is not going back into the bag.
It probably will, though, once model collapse sets in.
That’s the irony, really… the more successful it is, the sooner it’ll poison itself to death.
Impressive how…? It’s just statistics-based very slightly fancier autocomplete…
And useful…? It’s utterly useless for anything that requires the text it generates to be reliable and trustworthy… the most it can be somewhat reliably used for is as a somewhat more accurate autocomplete (yet with a higher chance for its mistakes to go unnoticed) and possibly, if trained on a custom dataset, as a non-quest-essential dialogue generator for NPCs in games… in any other use case it’ll inevitably cause more harm than good… and in those two cases the added costs aren’t remotely worth the slight benefits.
It’s just a fancy extremely expensive toy with no real practical uses worth its cost.
The only people it’s useful to are snake oil salesmen and similar scammers (and even then only in the short run, until model collapse makes it even more useless).
All it will have achieved in the end is an increase in enshittification, global warming, and distrust in any future real AI research.
I think this is still fundamentally misunderstood widely.
The fact that it’s being sold as artificial intelligence instead of autocomplete doesn’t help.
Or Google and Microsoft trying to sell it as a replacement for search engines.
It’s malicious misinformation all the way down.
LLM models can’t be updated (i.e., learn), they have to be retrained from scratch… and that can’t be done because all sources of new information are polluted enough with AI to cause model collapse.
So they’re stuck with outdated information, or, if they are being retrained, they get dumber and crazier with each iteration due to the amount of LLM generated crap on the training data.
So, they’ve basically accidentally (or intentionally) made Eliza with extra steps (and many orders of magnitude more energy consumption).
This article seems to put the blame on the shockwave from Starship’s rapid unscheduled disassembly in the upper atmosphere (not its launch) but there’s also been recent warnings about the effects of metal particulates from such explosions, satellites burning in the atmosphere, and similar pollution on the ionosphere.
All in all, burning or blowing up metallic crap in the upper atmosphere seems to be quite a bad idea.
These effects may be troublesome, but they are short-lived; re-ionization occurs as soon as the sun comes up again.
The problem is when you’ve got enough short lived microsatellites and Starlink-like constellations and whatnot that you’ve practically got a whole Kessler’s syndrome of the damn things constantly burning up in whatever’s left of the ionosphere…
If I’m not mistaken SpaceX actually has mechanisms in place to prevent Elon from directly meddling too much in the engineering decisions.
They’re the reason SpaceX’s products mostly work as intended, unlike Tesla’s, Xitter’s, or The Boring Company’s.
Elon being distracted with Xitter and politics and whatnot probably also helps.
Also to be able to disable certain functionalities of your car if you don’t pay a monthly ransom subscription.
So then you know what happens when you make a copy of a copy of a copy and so on. Same thing with LLMs.
LLMs are as close to real AI as Eliza was (i.e., nowhere even remotely close).
They can’t.
They went public too fast chasing quick profits and now the well is too poisoned to train new models with up to date information.
https://www.youtube.com/watch?v=MO0r930Sn_8