I did find it funny one time when I got a solution for a problem from ai at work, and then later found the almost word for word solution in stack overflow. Like this is just a search engine but with higher sea level rise and rainforest clearcutting when you use it!
It’s a search engine that you already have to know the answer to use.
It is like a search engine that will randomly spit out something correct (a human’s work), wrong, or sadly wrong.
With every response to a prompt you get to figure out the relative level of bullshit or propaganda.
Google also making search worse so people will use their slop machine instead
It seems like Google Fu for people without Google Fu
Often it is exactly the same as the top stack overflow answer. Sometimes it throws in some extra shit to make it incorrect though, so you have to scroll down and click on the SO link, which is what you would have done before all of this nonsense anyway.
Less ads though.
Not for long. Its called enshittification.
Not for long. Its called enshittification.
Do people think it’s something else!?
Good news, so does ChatGPT lol
Always ask for source
Asking an LLM to cite its sources is like asking a street-drug dealer where the drugs are grown/manufactured. They don’t know, and if they did they wouldn’t tell you.
ChatGPT recycles those very same 11 year old Reddit posts, with a 20% chance of making up the complete opposite, just because.
Um, LLMs were trained on that 11 year old post tho
Yeah, but they were also trained on tons of other incorrect, useless bullshit, so
The point I’m making is that reddit is also shit.
The internet has always been a cesspool of uselessness with a few gold nuggets to find. Some of the best information out there was in those old posts, from Usenet to Reddit, buried in shit.
When search engines actually indexed and crawled without the advertising algorithms you could find stuff. But for the past 6 years reddit has been shit, and search has been shittier longer. Now when you visit reddit you’re helping to fund fascists.
So the hallucinating LLMs were trained on garbage, so it’s shit on top of shit.
Quite sad
Yup, reddit is a shithole.
In another 12 years, those will all be ChatGPT replies too!
Lemmy, but yeah, same idea
I’m not sure I’d trust 11 year old Lemmy answers.
I trust AI to be one thing.
wrong.
If the point of this is that both are equally trustworthy then I agree.
Well where do you think AI got its info from LOL
Except Reddit IS all AI posts now. Dead internet
theoryrealtiy.that’s why “from 11 years ago”
True. But is the user accoint from 11 years ago. Or are you comparing an interesting from 11 years ago to AI today.
from 11 years ago
