Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News Editorials & Other Articles General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

General Discussion

fujiyamasan

(1,336 posts)

27. Same here -- they're fine if you're aware of its limitations

Reply to anciano (Reply #19)

Thu Dec 25, 2025, 11:26 PM

Dec 25

I don’t use it for AI “slop art” or whatever, but it saves me time on google searches and going through wiki rabbit holes when I’m curious about something. What I like is that after reading something either through the LLM or elsewhere I can use it to ask further questions.

If I then want to dig in depth I’ll look into primary sources.

Would I use it for some kind of safety critical application or something very complex? No, and I don’t think that’s what LLMs are intended for anyways.

Edit history

Please sign in to view edit histories.

Recommendations

0 members have recommended this reply (displayed in chronological order):

28 replies

= new reply since forum marked as read

Highlight:

How AI works [View all] canetoad Dec 25 OP

In other words... it's an accelerated version of the last 1000 years. WarGamer Dec 25 #1

AI has the potential to uncover some groundbreaking health treatments, True Dough Dec 25 #2

This is why a progressive Government needs to lead into the future of AI... WarGamer Dec 25 #4

Agreed. True Dough Dec 25 #6

We don't have that government currently, the AI bros have never talked seriously about sharing their highplainsdem Dec 25 #10

Putting that money directly into medical research would be much more likely to produce results. highplainsdem Dec 25 #5

It's being developed by all the wrong people for all the wrong reasons? snot Dec 25 #20

No, genAI is turning Homo sapiens into brainless twits dependent on hallucinating chatbots and highplainsdem Dec 25 #3

I disagree. WarGamer Dec 25 #7

So what? Chatbots have a very high error rate. You don't know if any of that's true unless you check highplainsdem Dec 25 #9

the whole IP theft thing... WarGamer Dec 25 #12

Human and AI learning are quite different. And btw, that's IP theft, not AI theft. And the AI companies know highplainsdem Dec 25 #17

Isn't the VAST majority of LLM taught on public source data? WarGamer Dec 25 #18

Look up the difference between public and public domain. What isn't in the public domain is highplainsdem Dec 25 #21

For one thing, when considering a claim for copyright violation, snot Dec 25 #22

google can't, but wikipedia can, and better WhiskeyGrinder Dec 25 #13

I agree with the phrase canetoad Dec 25 #15

there we agree... 100%. It's a propagandists dream. WarGamer Dec 25 #16

No, it's not. WhiskeyGrinder Dec 25 #11

Except the size and scope of what it can take in is greater EdmondDantes_ Dec 25 #14

Consider that "dog food" usonian Dec 25 #8

I personally have found AI to be a very useful tool anciano Dec 25 #19

Absolutely. In the last few weeks I had to dispute a medical bill ... rog Dec 25 #23

Excellent! anciano Dec 25 #24

Same here -- they're fine if you're aware of its limitations fujiyamasan Dec 25 #27

Just waiting for the bubble to rapidly disinflate. nt BootinUp Dec 25 #25

Looking forward to it. highplainsdem Dec 25 #28

I too have had an incredibly positive experience using AI. AI can find me information far quicker than Google. Doodley Dec 25 #26