Any technology can be good or bad depending on how it's used. The way corporations are shoving AI in anywhere they can, and stealing peoples content to train them etc. is bad. So is the privacy concerns.

But LLMs aren't necessarily bad, and can be good depending on how they are built and used.

[–] sundray@lemmus.org 6 points 2 days ago (1 children)

But LLMs aren’t necessarily bad

Maybe that's true. LLMs are fancy text prediction gadgets utilizing linear algebra and lambda calculus, whose facility with language creates the illusion of intelligence (which is one thing that LLMs are good at). But I think a lot of the VCs and execs in tech have lost sight of that fact.

I mean, for basic language tasks they're pretty good! They can summarize a webpage, or go through all my emails and tell me how many Daves I know, or write a summary of the appointments on next week's calendar. Useful things.

But forcing them to do things they're not good at makes them look half-baked. They're not ready to be customer service agents, they're not ready to accurately answer questions in specialist domains, they're not ready to book your flights, or do your budgeting, or your grocery shopping, and maybe they'll be brilliant someday -- but so many companies are pushing them right now regardless, and it makes them look bad and stupid.

Ramp up the research. Slow down the deployment. (And maybe deal with the copyright and energy consumption issues, don't burn the planet down for LLMs.)

[–] Ek-Hou-Van-Braai@piefed.social 0 points 2 days ago* (last edited 2 days ago) (1 children)

I might get crucified for saying this on Lemmy

But I don't see the energy issue as a real issue. I feel that is just the AI = Bad mentality looking for reasons to hate AI

If you compare it to other things, it's a drop in the bucket

1 kg Beef = 60kg CO2 - source
1000km Return flight = 314kg CO2 - source
1 Bitcoin transaction = 645kg of CO2 - source
1000 AI prompts = 3kg of CO2 - source including training

Don't get me wrong there are plenty of reasons to hate AI

Before I probably get attacked for these valid points, I just want to say I'm not Pro-AI (look at my history) There are plenty of valid reasons to be Anti-AI like Privacy, Copyright and many more, but the argument for the environment is a very weak one when you compare it to other things we aren't as upset about like flying.

[–] sundray@lemmus.org 3 points 2 days ago (1 children)

Whataboutism is not a very compelling argument. The existence of worse polluters does not pardon LLMs contributions any more than the existence of the crime of grand theft makes shoplifting a legal activity. Considering the existential threat of global climate change, it is the responsibility of every industry to avoid damage to the environment, however small their "drop in the bucket" is, as at a basic economic level a business on an extinct planet can no longer create value.

[–] Ek-Hou-Van-Braai@piefed.social -1 points 2 days ago* (last edited 2 days ago) (2 children)

With that logic we should all stop watching YouTube and Netflix because it's so damaging to the environment (much more than AI)

What I'm saying is when it comes to climate change we should be focusing our energy where it matters.

Most countries still burn coal, we should be angry about that, AI is insignificant compared to that.

AI is however a massive privacy and copyright issue, there we can focus on regulation

[–] lemmyknow@lemmy.today 1 points 2 days ago

Well, maybe these companies should adopt green energy. Then they could offer their services without damages to the environment. Ecosia, for instance, uses over 100% green energy, meaning they free up other energy from the grid (if I understand correctly). If other companies were to do the same, damage to the environment wouldn't be a worry. Especially for video, which, as I understand, is particularly bad for the environment.

Now, other reasons to hate on AI are a separate topic. Green energy doesn't solve for privacy or theft.

[–] BakerBagel@midwest.social -1 points 2 days ago (1 children)

Youtube and Netflix provide an actually useful service that allows people to watch/share videos. LLMs use an obscene amount of energy to do what is essentially just a fancy google search. Comparing the 2 is not justified

[–] Ek-Hou-Van-Braai@piefed.social 1 points 2 days ago (1 children)

By your logic a lot of Wikipedia is useless, you can just go to a library and look up the information in books.

LLm's provide an actual useful service, else people wouldn't be paying to use them.

Pretending it isn't is rediculous.

[–] BakerBagel@midwest.social -1 points 2 days ago (2 children)

People are paying for them because they are brain dead MBA's who think it's the future. My litttle library in rural Ohio doesn't have shit about the Mau Mau rebellion. Knowledge collections are worthwhile. Opening up a new nuclear plant so that you can fire all your customer support staff and replace them with chatbots is not useful.

[–] Ek-Hou-Van-Braai@piefed.social 2 points 2 days ago

A lot of software devs also use it to find the right syntax for something, or to add more unit tests etc.

LLMs are very useful, thats just a really you have we accept, I wish it was never invented, but it's here and it's here to stay.

I myself find it very useful, I tried to deny ot at first but you have to accept reality

[–] Ek-Hou-Van-Braai@piefed.social 2 points 2 days ago

Do you ever use Wikipedia to look up something that is in your library?