Human well being includes clean drinking water and the ability to discern truth from lies…both of which AI is destroying.
Human well being includes clean drinking water and the ability to discern truth from lies…both of which AI is destroying.
No it isn't.
If you think AI is using drinking water and energy, wait till you hear about...Netflix. andymasley.substack.com/p/individual...
I’d be thrilled if streaming ceased to exist…but at least Netflix provides the service it claims to and isn’t being literally forced on people in a vain attempt to make it profitable.
No one is forcing the use of AI on you. And many of us would not be thrilled if whichever modern inventions you happen to dislike were to cease to exist.
Do you work in an office? The problem isn't people playing around with AI on their home computers. It's people rushing to shove it into real infrastructure that people rely on, all while feeding it data that the AI evangelists don't really have the right to use.
There do seem to be some legitimate technical uses, but not nearly as many as the hype merchants seem to be pushing. So when they say they want to build giant nuclear power plants just to power data centers, I think they're getting ahead of themselves.
I do work in an office; my perception is that LLMs (and associated tools) are getting more capable and useful fast. But my point isn't that I think these investments will pay off—they may not. My point is that it is perverse to want them not to.
If they get good enough to replace a lot of workers, it's bad one way. If they don't get that good but companies replace workers anyway, it's even worse. And if the workers spend all their time correcting the AI output rather than doing creative work, that's bad a third way.
Like I said, I do think there are technical applications where AI makes sense. And tightly controlled output on limited subject matter might make sense, although I'm not sure what that gets you over using templates. But the hype guys are aiming for a scale of use that makes no sense to me.
Ideally some of the projects get cancelled as reality starts to bite, so that the resources aren't consumed before things come down to earth. But I do think they need to come down to earth, and some of the biggest hype guys have been obnoxious enough that I want them to land hard.
Well, then you're hoping both that the benefits of AI don't come to pass and we have a major economic downturn in which a lot of people get hurt. I very much hope you don't get your wish.
So, you'd rather we went back to physical media, where it requires factories to manufacturer the media and the packaging, and then the packaging and media get thrown out and are made of hard plastic that lasts centuries in landfills?
Nothing’s perfect. But my issues with streaming go beyond environmental issues and into how it has affected culture…and I can’t boil all of that down into a reasonable number of skeets.
Fair enough.
Yeah, this is just Luddism rather than any kind of reasoned environmentalism.
Nobody's building out all those data centers for people doing hyped up web searches. They're manipulating a lot more data than that in ways that may not be compatible with a free society. (If you do use Netflix, watch the documentary about Ed Sullivan. I had no idea.)
I don't think there is any reason at all to think that our odds of maintaining a free society are greater if AI is a big bubble and it pops.