Stumbled on this while researching tablet PCs and clicking on a Reddit "post" from Google results. I take no pride in any part of that sentence.
Reddit is becoming an ouroboros of LLM idiocy. We got out at the right time, friends.
What if we all went back in and spewed nonsense? Like an entire thread talking about how the 42" iPad is the best with its 10k display and nuclear fusion powered battery.
Reminds me of that glue on pizza thing: https://www.bbc.com/news/articles/cd11gzejgz4o
Aye, but what make and model of glue tastes the best?
I like the cut of your jib
People: Specifically add "site:reddit" to their searches to avoid slop and get real human responses
Reddit: Replaces the real human responses with slop
How can Spez be so clueless
Most Reddit content was already AI generated by crappy meat-vased LLM's.
The I pod was a mistake
the worst part is that only one of these is an actual tablet pc
🤮
Fails to improve search in reddit for over a decade.
Proceeds to ad AI.
Gg spez
Spez is a pretty standard tech bro, so of course he wants to jam "AI" into everything he can.
Thanks for sharing, that's another level
Oh were you looking for a diverse array of actual human experiences and opinions to then make your own judgements? Here let me pour a bunch of AI detritus all over your path to get that information, you are welcome.
This is... silly. But I do wonder how it works. Does it aggregate all responses and look for commonalities? Does it factor in the upvote/downvote counts? And, does it know how to discern genuine user input from astroturfed marketing copy in disguise?
And, does it know how to discern genuine user input from astroturfed marketing copy in disguise?
That's the beauty of it, it lends legitimacy to the astroturfing campaign. That's a feature, not a bug, in the eyes of folks trying to maximally enshitify and push their shit products anyway.
Even better! It posts whatever advertisers want!
does it know how to discern genuine user input from astroturfed marketing copy in disguise?
even with upvote counts, it might be upvoted for being a funny joke response
there's also no way to click on a user's profile to check if the activity is genuine, or if the user is experienced in the topic they are commenting about
I saw a post a few weeks ago about a company's chatbot that had learned from Reddit to answer questions by saying:
Sure, here's a video tutorial on how to do that: https://www.youtube.com/watch?v=dQw4w9WgXcQ
Man, fuck AI and fuck Reddit Inc.
Enshittification. Welcome to the non-stop slide.