WALL-E
The problem with AI for work is that there are going to be a lot of people who aren’t going to have economically viable labor. History has shown that people who aren’t economically viable tend to get pushed out of society.
Arguably there are already people in this position. Or they are massively overskilled compared to any work they can get.
In unrelated news - will write SQL for food.
Nothing can be maintained without at least some of the skills and knowledge that built it. Even something like generative AI like LLMs will eventually break down in some manner. Its just what happens with complex systems, be they physical or digital.
After that happens, we’ll be forced to learn either to fix the thing, or how to do without the thing. How painful that learning process is will depend upon how essential that thing is to our lives.
If AI goes away in a few years, barely a blip.
If AI goes away after a few decades, and we’ve allowed much of our accumulated knowledge to atrophy… It’s a crisis.
If you’ve ever seen a millennial calculate sales tax on a replacement phone then you’re getting the picture.
Gen z trying to split a bill 6 ways and getting confused at how to split the service charge…
I don’t know, but I’m really concerned about AI use in education.
They are already teaching with it and integrating it into how our children do homework.
AI is the processed food of the brain/mental health. It will make people stupider and lazy, just like processed food makes us fat and lazy.
That’s the problem. You will end up with a epidemic of people who are mentally obese and lazy and refuse to do anything about it. Far worse than we have it today, and we are already seeing mental abilities decline due to it’s adoption and use.
You only become intelligent my exercising your brain. The issue with AI is it basically is like GPL for your brain, it lets you cheat and you have to make minimal effort, and your mental abilities atrophize, just like your muscles do if you don’t use them.
That just sounds like what’s already happened with social media, Murdoch journalism and ‘fake news’. So, SNAFU…
We should be making an effort to limit AI, not let it flourish.
That entirely depends on specifics.
Corporate America already is a soul-crushing land of terrible art, bad writing, and shoddy code. The artists, writers, and programmers there will have a mix of reduced job satisfaction and more competition for creative roles but a reduced portion of their workload doing the most creatively boring parts of the job. So, to the extent that “in-house creative” remains, it will more or less be the same blah it is today.
The big risk is the destructive cycle of LLMs and “GenAI” in specifically creative enterprises. If Disney replaced all their creatives with AI slop, and the AI continues its trend of unimpressive mediocrity, Disney as a creative corporation might shut down or even go out of business.
What’s worse about the above is that if it’s replicated on a large enough scale for a long enough time, we might wind up having no creatives at all and the whole skill set may atrophy away from our civilization.
On the other hand, if GenAI winds up substantially increasing the proportion of unemployed citizens, a UBI might be implemented and all those creatives who chased soul-crushing work just so they wouldn’t starve would do it for the pure joy of creation.
(All of which, of course, assumes that the runaway power demands of GenAI don’t destroy the biosphere…)
The thing Hollywood executives don’t get is if they can AI generate a movie, so can a guy in his basement
Global warming due to the co2 we’ve already dumped will destroy us anyways. The “AI” slop is just the cherry on top that will guarantee societal collapse instead of just mass migrations and water wars.
Everything that’s currently wrong with it today, but amplified.
People said the same about the internet.
We were all supposed to be illiterate because of TV by now.
Moral of the story is that doomers are always wrong.
I think the extremes are overall wrong but have whispers of near future truth and foreshadowing of distant truths. Once resources become truly scarce globally we’ll see some of the Armageddon set in fueled by the technology we have at that time, but it’s a lot slower moving that doomers communicate.
deleted by creator
If people can’t live without AI in the future, and access to AI ends, then people will die. Dying is pretty serious.
Here’s what’s happening right now, the rich are keeping their kids away from the AI crap. They’re being sent to private schools that still teach the traditional ways. They’re doing this to ensure that their kids will be in power in the future, and rule over the helpless slobbering idiots our public school system is currently releasing into the wild.
I don’t think this is sustainable, it will break in some terrible way, and everyone, including the filthy rich will have their asses handed to them. But, I don’t think that will lead to any sort of enlightenment or utopia, we’ll probably cobble together some barely adequate solution and then repeat the whole thing again inside of a couple of hundred years. Rinse and repeat until we finally kill everyone off. I mean, just based on how things have gone up till now.
Don’t need AI to maka a pike for musks head.
Probably not much different from all the other things that we can’t work or live without.
In the short term, we (will) primarily see ML used as a tool for control. It’s simply too bad to actually replace anything long term.
It has already replaced people. It doesn’t have to replace 100% of 1 person, if it replaces 20% from 5 different people that is 1 job lost and the other tasks reallocated.
Depends on who controls the AI. If you are running it locally.
https://blog.mlc.ai/2024/04/20/GPU-Accelerated-LLM-on-Orange-Pi
Then it doesn’t matter.
That solves maybe a single problem associated with AI. Energy and water use is still massive, nobody’s job prospects are improved in any way, the brain drain still happens, and creativity still goes to die in a corner.
I have since learned a few skills from probing AI. One of them is python and I use it today. They aren’t all downsides.
What water use does your phone have? Because an Orange PI has none. You can run it off a $5 solar panel and thus the energy problem is not there.
Stop conflating society destroying Capitalism with community built and maintained software. They may have functional similarities but the negative properties of Oracle Databases are orders of magnitude worse and different than FSF supported Databases such as PostgreSQL.
You’re already referring to AI as ‘them’ and that is a problem in and of itself.
I swear this must be some kind of disease of lemmy users, the urge to personal attack or criticize the post author instead of collaborating with the discussion. I’ve never seen it happen so often not even on Reddit.
OP is making a serious question, how it is called doesn’t add anything to this discussion.
If you think my comment was a personal attack, whew buddy… settle down. You couldn’t find the humor in it, or at least the dry retort?
it’s just being stupid and taking shit personally.
Isn’t them the plural of it?
Yes it is. Anthropomorphizing LLMs is stupid but that’s not what is happening here…
Yes
It could be understood to mean “AI [tools]”, which would be plural
deleted by creator
