55
The ‘dead internet theory’ makes eerie claims about an AI-run web. The truth is more sinister
(theconversation.com)
this post was submitted on 12 Jun 2024
55 points (93.7% liked)
BecomeMe
805 readers
1 users here now
Social Experiment. Become Me. What I see, you see.
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
I don't think content is
At least, not yet. And to be fair, the internet between say 2015 and 2020 was already a huge garbage dump that made searching for information quite hard.
This is paradoxical in part, but it makes sense. It was hard to find something in the early internet because it may have been missing in digital, it is hard now because there's too much information.
So maybe the problem will get worse because of the use of AI, and we will do something. Or not ¯\_(ツ)_/¯
Go to any of the "question" subreddits (AITA, TIFU, AskReddit, Two Hot Takes etc etc) and tell me that 95% of the content isn't posts by LLM. Claiming its a throwaway is an easy way to hide the fact that it's a new user with no post history, only difference is now they interact with users to get better understanding of where social lines are drawn.
Every single one has a username like danger -skyscraper-1376 and is obviously a chat GPT scenario asking obtuse questions that normal humans wouldn't need guidance on "my wife slept with the entire office, but we're illegal immigrants with 15 kids and divorce isn't an option. AITA if I want a PS5 pro that I don't share with anyone but my alternate Partner??"
The internet isn't "dead" but it is a corpse marionette that tells people whatever they want to hear, gives them straw man enemies made of everything they love to hate, and convinces them everything they think and feel is absolute truth.
The usernames are randomly generated when creating an account. It could simply be people that can't be bothered to change it, like I did.
It's no secret that old popular posts including popular top comments are reposted by bots though.
And woe betide you if you ever say "Wow, where can I buy that shirt?"
I think, we may frequent very different parts of the internet and thus have different views on how much is AI generated. Maybe you're right and Reddit/Xwitter/Facebook/etc are dominated by bots, I can't judge on my own about that
So where exactly on the internet are you that's bot free?
News articles are pumping out AI stories, websites chat help is often AI bots, social media is literally a bot farm with human spectators.
Unless you are reading blogs from people you know who aren't familiar with chat GPT, or you're entirely online for small/niche discussion groups, I'd say you're exposed to far more AI content than you realize.
I'm not suggesting everything is fake bot content with some people mixed in yet, but in 5 years it probably will be.
Much harder for people to organize, or even get exposed to reality, if their digital experience is a walled garden of pushed bot content.
Well, first of all, yes, I am mostly either in Lemmy, or in Telegram reading small channels.
But even then, I am not against content that is AI generated if it is reviewed and correct. The article talks about outright nonsense, so I supposed that you were talking about that, too.
To be fair, human-generated nonsense is also a kind of content I find good, but it depends a lot and I grew tired of simply meaningless absurd sites lately.
Edit: oh, and I am totally with you on the "yet", we're about to see The Matrix level of isolation from reality some time soon, I guess