Half of Americans now use AI chatbots, up from 33% in the summer of 2024.
Archived version: https://archive.is/newest/https://gizmodo.com/just-16-of-americans-believe-ai-will-positively-impact-society-pew-poll-finds-2000773112
Disclaimer: The article linked is from a single source with a single perspective. Make sure to cross-check information against multiple sources to get a comprehensive view on the situation.
This is a ridiculous comparison. GenAI has practically zero use case compared to the internet and it’s benefits to economic and logistical planning and communications. How the hell do you even come up with this take? We have already reached diminishing returns on advancement, meanwhile all these datacenters are massive resource drains with literally 0 benefit to society.
I think it’s a good comparison.
The main benefit of the Internet is that it gives easy access to near limitless information. The basic Google use case of the Internet, I type in a question and get a link to an answer, whether that’s a reddit thread, stack overflow, wiki article etc.
Whether you like it or not people are increasingly using LLMs for these sort of basic “factual” questions. Yes they can lie/hallucinate, so can someone on reddit, but most of the time they are correct and that’s good enough for most people.
The problem with both is when we take them to seriously. It is common knowledge that both LLMs and the Internet can lie or bullshit you. When you lose sight of that fact and believe an internet community or personality or an LLM knows the world better then your own sense, it becomes a problem. And honestly the people who get AI psychosis are far less dangerous to society then some racist asshole who lost his mind on 4chan.
People like you are the reason people hate AI. Both are painfully confidently wrong.
How am I wrong? Do you think getting information is not a use case? Or that people aren’t using LLMs for that use case? If not you seem to be the one confidently wrong claiming there are 0 use cases.
You can dislike the fact that people are using it for that use case, some people don’t like the Internet being used for porn, but you can’t deny that people are using it and getting something out of it.
Not when the information is wrong more often than not and there are already better solutions for that so-called use case that dont use half a billion gallons of water and cities worth of electricity to also mostly generate CSA content.
It causes far more harm than any hypothetical good and I am not convinced it even does anything particularly good as I have laid out here. You want to defend your slop maxhine, fine by me but at least be honest about it.
Have you used any of the recent models? Or is your only impression of LLM output memes on here where the LLM is laughably wrong. LLMs are right way more often then they are wrong these days. If you don’t believe me try and test it yourself, come up with some factual questions and ask opus 4.6, see if it’s more wrong then right. Hell try and come up with a riddle to trip it up, even that is becoming harder with frontier models, they’re not going to tell you to put glue on your pizza any more.
See this is the problem with both LLMs and the Internet, instead of putting a little research and trying to check something yourself you let the LLM / internet community do the thinking for you.
I’ll give it to you that there are some valid environmental concerns, but its not nearly as big of a problem as say meat or even golf courses so if your going to direct your ire at something I’d choose those.
Mostly? This is a small use case, that is probably overblown by the community your in. Yes it exists but this is also a problem with the Internet in general. Is the Internet/LLMs used to make and distribute CSA, yes, is it most of the traffic, No. LLMs are in a sense manifestations of the Internet, with all its beauties and horrors such as CSA. Yes you can focus only on the horrors but that is a very narrow view of what it actually is and can be.