Cory Doctorow: What Kind of Bubble is AI?
(locusmag.com)
from mo_ztt@lemmy.world to technology@lemmy.ml on 27 Dec 2023 18:38
https://lemmy.world/post/10032962
from mo_ztt@lemmy.world to technology@lemmy.ml on 27 Dec 2023 18:38
https://lemmy.world/post/10032962
threaded - newest
For reference, we are on the third AI bubble
I don’t know about third; I guess it depends on which bubbles one considers worth counting. I got a little work out of the expert systems bubble way back in the day, just before the first dot-com bubble started.
The 2008 financial crisis was not the result of a “tech” bubble.
Thanks for your pedantic, useless gotcha analysis.
You’re welcome.
Eh. Financial instruments are an invention, and they can shape advances as readily as any electronic innovations. Joint-stock companies contributed to the age of sail as much as ship’s clocks and other navigation tools. That gimmick is ultimately why stock exchanges have been high-tech digital affairs (with quite a lot of analog interfaces) since before microchips.
Enron claimed to have some vague new means of lowering costs, acquiring revenue, and making the line go up. It would have been transformative for the power industry if it wasn’t straight-up fraud.
The 2008 lending crisis was halfway between those. It was a new strategy that turned crappy lending into worthwhile lending… in theory. And it relied on a ton of data. Too bad it was cyclical bullshit.
The 2008 crisis was a result of predatory lending practices and mortgage backed securities that were full of subprime mortgages and other toxic assets, and a burst real estate bubble. I don’t think it was a tech bubble like the dotcom bust.
Well, they were gaming the rules in the California energy market. They were doing shit like causing blackouts so they could price gouge.
Right, and also just making shit up. Enron’s executives literally could not describe how they made so much money. The secret ingredient was crime.
The 2008 crisis was at least somewhat rooted in reality. It was a risk management strategy that - in theory - let them lend out a looot more money, by being less likely to lose money, even if they gained less per-dollar. But of course every single asshole involved cranked up their flour-to-sawdust ratio and promised nothing could possibly go wrong.
I think it’s worth including finance shenanigans alongside tech bubbles, because even if finance isn’t tech, tech bubbles are absofuckinglutely just another finance shenanigan. The dot-com bubble didn’t burst because people stopped using the web. It was all about obscene sums of money being thrown at stupid ideas. Some of those stupid ideas worked! If not for the investor-class expectation that you turn the crank and get ten dollars for one dollar, we’d simply describe the period as one of naive experimentation. Better environments would allow people to try their big stupid ideas using their own money, or mundane business loans, instead of trying to attract millions and promise billions.
It’s only a bubble because greedy idiots pump it up. Without that, failures are still news, but they don’t become a historical event.
spunk bubble
An interesting article, but it seems to be missing the big applications of AI. It isn’t all about the LLMs and other large models, but where it will definitely be used is in smaller-scale problems where specialized models can be pruned. There is a bubble, that’s for sure, but it’s in the usage of large, unpruned models for meanial tasks.
I’d say it isn’t, but I’d be saying it about the existing applications (and near-future stuff that currently sorta-kinda works), not the tech-bro delusions that wind up on billboards and cause aging suits to both sweat and drool.
AI is the kind of bubble where the tech involved is halfway to magic. And it’ll run on your local hardware. No matter how hard the offices-and-capital side crashes and burns, that’s not going anywhere.
Right now is the worst that image AI will ever be again.
LLMs might stumble, because the big-iron approach seems to make a difference for them, but there are local versions and they do roughly the same things. That’s going into video games, for a start, and probably turning every single NPC into a verbal chatbot that can almost hold a conversation.
… entertainment is a bajillion-dollar industry where abject failure is routine and hilarious. There will be a boom like there was for CGI, only with even worse treatment of the tiny companies performing miracles. Workers getting screwed while revenue floods in is not the same thing as a bubble bursting. Unfortunately.
I’m disappointed in Doctorow for asserting this technology will remain big and complex and expensive. When has that ever stayed true? Saying it’ll always take forest-eating big iron sounds like predicting computers will only be affordable to the five richest kings of Europe. This whole neural-network revival kicked off because consumer hardware made training feasible. More training equals better models, and more computers equals more training, but Google’s still pouring gigawatts into glorified video-game tech.
If all the creative and academic zeal gets left working with mundane single machines - guess where all the advancements will happen.