A.I. Bubble

  • Thread starter Thread starter Flame
  • Start date Start date
  • Views Views 2,084
  • Replies Replies 29
  • Likes Likes 3
I think the tax payer will bail them out again like in 2008

How mad is that, tho. Billionaires tech oligarchy are creating AI so they can replace the workers and people.
  • If they fail. Those workers and people will bail them out.
  • If they succeed, those workers and people will eat dirt.
 
How mad is that, tho. Billionaires tech oligarchy are creating AI so they can replace the workers and people.
  • If they fail. Those workers and people will bail them out.
  • If they succeed, those workers and people will eat dirt.
that's unfortunately how the system is, the rich get richer while the poor get poorer
 
Step 1: Eat the rich

Ralph.JPG
Have you "eaten" one? I have :shy::evil:. And they taste NASTY !
Ralph.JPG
Ralph.JPG
Ralph.JPG
 
Oh, it will burst, and help filter out more useless workforce who went balls-deep into relying on it for their sloppy content creating and whatnot, so world will re-balance itself and probably lots of people will be able to teach classes on slop-correcting in their respective industries.

AI already started to spiral into learning from its own, constant mistakes since its driven by generally sloppy or shit answers from Breddit, and it will only spiral downwards from there. Once "what's 2+2," gets a "5," as an answer - it's donesies.
 
  • Like
Reactions: Flame
Same scenario as:

AOL -> Yahoo -> Google -> ADS -> Crappy alternatives.

@AOL.com -> @yahoo.com -> @hotmail.com -> @gmail.com -> ADS -> Crappy alternatives.

MySpace -> Tumblr -> Facebook -> ADS -> Crappy alternatives.

ChatGPT -> DeepSeek -> Gemini -> ADS -> Crappy alternatives.

Blockbuster -> Piracy -> Netflix -> ADS -> Crappy alternatives.

Radio -> Vinil -> CDs -> Spotify -> ADS -> Crappy alternatives.

People, as long as they can't find anything better to do, they'd consume anything.
Just a trend, like everything in life.
 
Depends on the implementation of the technology and what it'll be used for. LLMs have their uses, but I feel that they're being talked up a lot more than they are capable of at present. Also AI is open to manipulation, just look at the right wing rhetoric of Grok. Musk has his fingers so deep into Grok, he's wearing it like a puppet.

AI can and will have it's functionality, but there will always be a need for people to configure it, and the underlying hardware that it'll run upon. The google CEO wrote a good article recently, that the CEO level is more likely to be replaced by AI than lower level employees. But time will tell, as long as we don't get some skynet stuff going on, then we should be fine, but as with any new technology, time will be the greatest judgement of all. I mean just look at the metaverse, or NFTs, and how quickly that bubble burst.
 
It might get to the point where it's too big to fail.
How or why?

There is very little practical application of it at present. We have ChatGPT/Grok/Copilot etc, not all of them will survive. Im old enough to have seen enough of these wars between companies, to know that some will die off quicker than others. Bluray Vs HD for instance, VHS Vs Betamax etc.

Also the energy consumption and hardware requirements are exponential, the amount of data centres being built now to house this stuff is incredible. The electric/heating/cooling costs alone are in the 100s of millions, unless you do what facebook did and build one in the middle of Iceland and use external cooling via vents etc.

Also there is a free chinese version, and companies don't want to use these external LLM models, and prefer to develop their own internal ones, based on readily available code, also their data is much safer.
 
There is very little practical application of it at present. We have ChatGPT/Grok/Copilot etc, not all of them will survive.
This is called Evolution.
LLMs are a massive step from:


Wait what? You thought AI is new?
(not telling you specifically, mind you)


The next step from LLMs are AGIs, which will be incredibly useful.
In the currently scape of time, LLMs are like a bunch of librarians handling you information.
AGIs would be, in comparison, like Scientists of different fields utilizing Quatum Processing to create new information (i.e. no training).
Or that's at least how it should be.


Im old enough to have seen enough of these wars between companies, to know that some will die off quicker than others. Bluray Vs HD for instance, VHS Vs Betamax etc
Same.

Also the energy consumption and hardware requirements are exponential, the amount of data centres being built now to house this stuff is incredible. The electric/heating/cooling costs alone are in the 100s of millions, unless you do what facebook did and build one in the middle of Iceland and use external cooling via vents etc.
Nvidia CEO just anounced they're going for China (owch!) because China builds in days, not months.
And it would be wise to move for renewables, maybe nuclear at the very least. Again, that's ideally. Once the AI warlords finds: "Holy Shit, I can get free energy from the Sun? Man! Now I can charge x10 for my tech!"... they'll definitely push for Renewables.


Also there is a free chinese version, and companies don't want to use these external LLM models, and prefer to develop their own internal ones, based on readily available code, also their data is much safer.
Meh, debatable.
It's not like, y'all arent giving away your data to other services, like Gmail, Facebook, Netflix... at this point is incredibly stupid to play the "I'm shocked Google used my Google Photos pics for AI training" which anyone would've know beforehand if they read the End User License Agreement.

- You Retain Copyright: In most cases, you still own the pictures you upload. The EULA grants the company a license, not ownership.

- Grant of License: You typically provide the platform a worldwide, royalty-free, and sometimes sublicensable license to use your content. This license allows them to display your photo to other users, make thumbnails, or use it in marketing materials for their service.

Which are explicitly written in every fucking internet site that allows image uploading. Even the fricking TEMP has these in their EULA (or had? since I can't find them anymore) .
Post automatically merged:

BTW: Wasn't LLMs obvious because suddenly in 2019 or something everyone were doing neural networks? "Oh look, I let my computer on for 30 days, ended up with this one neural network, hur hur." "Oh my god! Nice, what's it for"? "Uuuuu... Super Mario World"
 
Last edited by JuanMena,
  • Like
Reactions: Magnus87 and janobi
If AI becomes powerful like realistic robots/androids and/or Star Trek type holograms, it won't go away. Once sexbots are a thing, AI will last forever.
Sex bots, that's where you've gone... Dude! We're a massive way off holograms, and star trek ideas, I mean that was quite a few centries in the future ;).

@JuanMena - I like your reply, and points. You made some valid ones indeed.

AI has come a long way from where it was, im sure we can all remember the video of Will Smith eating noodles, and the deepfakes that can now be created. As with anything technology will advance and improve, but it still has to have a practical application, but as you said if it can be weaponised, then the government will really lean into it.

Boston Dynamics is a prime example of this, when they were first developing robots, it was stated that there would be no military application, and then it gets bought, and all of a sudden there's robots being displayed with gattling guns on their backs etc... War makes money, and it's the biggest export of the US.

The Chinese AI, I wasn't saying companies are using it, im merely saying that they released the code so it could be adapted by other companies/countries. You're absolutely correct that using free services, YOU are the product, and that's how these things work. Facebook/instagram, X, Google, if you don't pay then you're the product, and in a lot of cases, even if you do pay they still sell your data regardless.

Im not sure on the retention of your images etc, afaik, once they're in the public domain eg uploaded to 3rd party website, then they're free to use. You can put disclaimers etc upon them, but im not sure you hold copyright for said images, this is currently being tested in court, and i'd suspect that most courts will side with the technology companies on this. Billionaires make the rules, they don't follow the rules.
 
Last edited by janobi,
  • Like
Reactions: SylverReZ and Flame
It'll reshape society, once we have general AI then we'll be forced to implement universal basic income just so society doesn't collapse. But also the AI will be able to make good use of certain people and help them generate more money on top of their UBI, corporations will end up making more money than ever before. And not just money, but power because why wouldn't anyone be loyal to a company that is making them rich in exchange for letting a real AI they control run your life for you.
 

Site & Scene News

Popular threads in this forum