I hear it’ll be $20 a month soon for the privilege of a glorified search engine distilled into a chatbot.
Comment on Researchers confirm what we already knew: Google results really are getting worse
northendtrooper@lemmy.ca 9 months ago
I’m finding that LLMs are doing a better job for searching for new things. If I have a question, instead of going to google or bing I’ll goto chatGPT and ask some of that nature with some sources for further reading.
Never would I think that I would need to use AI to answer simple search and yet here we are because the sole purpose of a search engine doesn’t really exist anymore.
LWD@lemm.ee 9 months ago
Odelay42@lemmy.world 9 months ago
How much do you think should it cost to use?
LWD@lemm.ee 9 months ago
If the data was gathered using the legal loophole of being for research purposes, I’m not sure anyone should have exclusive rights of it to begin with… Not sure if that answers the question
Landless2029@lemmy.world 9 months ago
Oddly I prefer Bing because it’ll cite the source!
centof@lemm.ee 9 months ago
Perplexity is great for this. It gives like 5 links in addition to the text answer so it is imo the best of both worlds.
chaogomu@kbin.social 9 months ago
The problem is, you can't trust ChatGPT to not lie to you.
And since generative AI is now being used all over the place, you just can't trust anything unless you know damn well that a human entered the info, and then that's a coin flip.
lolcatnip@reddthat.com 9 months ago
OTOH, you also can’t trust humans not to lie to you.
chaogomu@kbin.social 9 months ago
That's the coin flip.
Lmaydev@programming.dev 9 months ago
The newer ones search the internet and generate from the results not their training and provide sources.
So that’s not such a worry now.
Anyone who used ChatGPT for information and not text generation was always using it wrong.
BakerBagel@midwest.social 9 months ago
Except people are using LLM to generate web pages on something to get clicks. Which means LLM’s are training off of information generated by other LLM’s. It’s an ouroboros of fake information.
Lmaydev@programming.dev 9 months ago
But again if you use LLMs ability to understand and generate text via a search engine that doesn’t matter.
_number8_@lemmy.world 9 months ago
plus search engines don’t lecture me as much for typing naughty sex words
Speculater@lemmy.world 9 months ago
Get on the unfiltered LLM train, they’ll do anything GPT does and won’t filter anything. Bonus if you run it locally and share with the community.
notapantsday@feddit.de 9 months ago
However, I find it much easier to check if the given answer is correct, instead of having to find the answer myself.