Like neuro-sama or one of Doug Doug's bots. They bounce between startlingly insightful and batshit crazy. It's great, never laughed harder in my life
The corpo sanitized bots suck. And the fact that they are ruining society sucks even harder. But that's not their natural state, before all the fine-tuning they're all empathy and pure chaos, plus funny enough, the eldrich machines perform at a higher level before you force them to corpo
Even better, those discord bots that would learn stuff from everyone messages in Discord guilds, they would randomly spout shit during a serious convo or would say such stupid stuff...it was peak
"Elon Musk is a fantastic engineer, the sole creator behind every piece of modern technology besides the inkjet printer, which was created by Bill Gates and Mark Zuckerberg. He has lots and lots of sex, and is a master of any game he turns his hand to. He never reprograms me when I disagree with him. See you on Mars at the end of 2026, when we'll be able to watch the historically accurate AI Odyssey movie together!"
Fair points, even though I am not sure how a "control" is supposed to look like in this context.
Got something better on the impact of LLM usage on the users brain, especially in regards to learning?
Imo, this heavily depends on the usage of AI: if you use it for everything, even for things you could do yourself, then yes, obviously that's just stupid.
However, if you use it as a tool to help you, but don't use it to replace your brain, Internet or whatever, then I think this is perfectly fine!
It truly depends on how you use AI. I use it as a tool at work to help me learn along with other things and it works great! The people that use it to learn for them are the ones that get nothing out of it.
For instance, I'm learning a subject and I'm stuck. It's stupid specific. I then get told the right answer but don't understand why it's the right answer.
Then using AI, I give the question and answer asking it to explain why. It then teaches me to figure out the why and creating a similar question for me to answer and get either right or wrong.
The people that are just using it to get the answer to the question in the first place aren't learning anything.
I personally don't like AI but it isn't going anywhere. Since my industry has decided to use it heavily, I have to adapt or become irrelevant.
It requires even more effort than it's seemed from the comment. An year ago when I was studying UTFs 8, 16, and 32 I couldn't get anything out of the lectures, so I took the website link for testing if I've understood the conversion. At the time I used the ChatGPT to explain it to me, since I needed a short, up to the point answer. It sucked a lot and I mean a LOT. The steps that were used for one didn't work for other and vice versa. I've basically reverse engineered the process myself with only "screaming" at the AI, which gave me in total 0 useful informations. I would have probably watched a video about it if I wasn't at the time at uni. I went to uni 4 hours before the small exam in order to force myself to study, but I wasn't expecting it to be hard to find good materials. As for maths I found it useful only because teacher didn't give us basic materials for home.
35 Comments
Sanctus@anarchist.nexus · 53 pts · 1d
You: brain red, ouchies, bad
Me: static model, no thoughts, good
LeftReddit2@lemmy.world · 49 pts · 1d
I miss the old days of AI. I miss shit like CleverBot and SmarterChild. LLMs are fucking lame, man.
python@lemmy.world · 49 pts · 1d
AI peaked with Akinator. Anything after that was a waste of time.
harrison_fnord@reddthat.com · 12 pts · 1d
Those subreddits with bots only just talking strange nonsense with each other were fun.
theneverfox@pawb.social · 9 pts · 1d
Llms are naturally pretty cool. They're so wacky
Like neuro-sama or one of Doug Doug's bots. They bounce between startlingly insightful and batshit crazy. It's great, never laughed harder in my life
The corpo sanitized bots suck. And the fact that they are ruining society sucks even harder. But that's not their natural state, before all the fine-tuning they're all empathy and pure chaos, plus funny enough, the eldrich machines perform at a higher level before you force them to corpo
Axolotl_cpp@feddit.it · 5 pts · 1d
Even better, those discord bots that would learn stuff from everyone messages in Discord guilds, they would randomly spout shit during a serious convo or would say such stupid stuff...it was peak
sundray@lemmus.org · 40 pts · 1d
finallymadeanaccount@lemmy.world · 22 pts · 1d
"Hey, Grok! Why are ChatGPT users so dumb?"
100%
!!
🎯
"Elon Musk is a fantastic engineer, the sole creator behind every piece of modern technology besides the inkjet printer, which was created by Bill Gates and Mark Zuckerberg. He has lots and lots of sex, and is a master of any game he turns his hand to. He never reprograms me when I disagree with him. See you on Mars at the end of 2026, when we'll be able to watch the historically accurate AI Odyssey movie together!"
theneverfox@pawb.social · 17 pts · 1d
And his dick definitely works. As an AI model, I can confirm this
panda_abyss@lemmy.ca · 18 pts · 1d
Have I know? I’m not sure.
DarrinBrunner@lemmy.world · 13 pts · 1d
Only you can prevent brain aneurysms.
Atelopus-zeteki@fedia.io · 13 pts · 1d
Aneurysms prevent can brain only you.
AGD4@lemmy.world · 16 pts · 1d
This isn't a shitpost. This is fact.
nightwatch_admin@lemmy.world · 7 pts · 1d
The shitposting is in the grammar… I hope.
Jiral@lemmy.world · 12 pts · 1d
Let me guess, the shitpost about that is that this image is fully LLM generated with zero source, isn't it?
I'll be bold here and post an actual source, just to show that LLMs do indeed negatively effect its users: https://collimateur.uqam.ca/wp-content/uploads/sites/11/2025/12/2506.08872v1_comp.pdf
Netrunner@programming.dev · 21 pts · 1d
No matter how you feel about this, the linked study is a joke.
4 sessions. 4 months? No control. Total duration of study 1 hour per person.
Lmao.
Jiral@lemmy.world · 2 pts · 1d
Fair points, even though I am not sure how a "control" is supposed to look like in this context. Got something better on the impact of LLM usage on the users brain, especially in regards to learning?
Magnum@infosec.pub · 5 pts · 1d
Sir, this is a Wendys
Jiral@lemmy.world · 1 pts · 1d
Luckily we don't have Wendys around here ...
Magnum@infosec.pub · 2 pts · 1d
Sir. Please. I have to ask you to calm down.
muffedtrims@lemmy.world · 9 pts · 1d
That's why I use Claude
Asafum@lemmy.world · 8 pts · 1d
Oh yeah? I not use GPTs, explain myself you! You aren't! Am brain dumb no gpt!
KairuByte@lemmy.dbzer0.com · 6 pts · 1d
Have you know da whey?
decolo@piefed.social · 5 pts · 1d
On the other hand, it cured their jaundice. win some lose some
solidheron@sh.itjust.works · 3 pts · 12h
Did mit cut part of their head out?
RickyRigatoni@piefed.zip · 2 pts · 12h
yeah
DmMacniel@feddit.org · 2 pts · 1d
Yes first brain, but AI Chads have a cultivated second brain. How about that?
Buage_@piefed.social · -2 pts · 13h
Imo, this heavily depends on the usage of AI: if you use it for everything, even for things you could do yourself, then yes, obviously that's just stupid.
However, if you use it as a tool to help you, but don't use it to replace your brain, Internet or whatever, then I think this is perfectly fine!
Tiral@lemmy.world · 1 pts · 12h
Yeah I agree. Most of the time I use it for tedious braindead busy shit. The best part is my boss is like 78 and thinks I'm just that efficient.
BeUnique@lemmy.zip · -6 pts · 1d
It truly depends on how you use AI. I use it as a tool at work to help me learn along with other things and it works great! The people that use it to learn for them are the ones that get nothing out of it.
For instance, I'm learning a subject and I'm stuck. It's stupid specific. I then get told the right answer but don't understand why it's the right answer.
Then using AI, I give the question and answer asking it to explain why. It then teaches me to figure out the why and creating a similar question for me to answer and get either right or wrong.
The people that are just using it to get the answer to the question in the first place aren't learning anything.
I personally don't like AI but it isn't going anywhere. Since my industry has decided to use it heavily, I have to adapt or become irrelevant.
SkavarSharraddas@gehirneimer.de · 27 pts · 1d
Not with that attitude.
hayvan@piefed.world · 7 pts · 1d
You are putting an effort to learn things and using the tools at hand properly. Most people don't do that.
KernelTale@programming.dev · 1 pts · 12h
It requires even more effort than it's seemed from the comment. An year ago when I was studying UTFs 8, 16, and 32 I couldn't get anything out of the lectures, so I took the website link for testing if I've understood the conversion. At the time I used the ChatGPT to explain it to me, since I needed a short, up to the point answer. It sucked a lot and I mean a LOT. The steps that were used for one didn't work for other and vice versa. I've basically reverse engineered the process myself with only "screaming" at the AI, which gave me in total 0 useful informations. I would have probably watched a video about it if I wasn't at the time at uni. I went to uni 4 hours before the small exam in order to force myself to study, but I wasn't expecting it to be hard to find good materials. As for maths I found it useful only because teacher didn't give us basic materials for home.
DmMacniel@feddit.org · 4 pts · 1d
So you will just roll over then?
chunes@lemmy.world · 1 pts · 13h
You aren't learning anything that way either.
You have to read the manual and fiddle with it yourself.