I just say "I dunno lol" like a normal idiot 
this post was submitted on 01 May 2025
140 points (100.0% liked)
chat
8409 readers
192 users here now
Chat is a text only community for casual conversation, please keep shitposting to the absolute minimum. This is intended to be a separate space from c/chapotraphouse or the daily megathread. Chat does this by being a long-form community where topics will remain from day to day unlike the megathread, and it is distinct from c/chapotraphouse in that we ask you to engage in this community in a genuine way. Please keep shitposting, bits, and irony to a minimum.
As with all communities posts need to abide by the code of conduct, additionally moderators will remove any posts or comments deemed to be inappropriate.
Thank you and happy chatting!
founded 3 years ago
MODERATORS
AI translated manga bit
Okay but is it accurate tho
to be clear: that is cum, the verb, not the noun
Hexbear is a leftist online community that originated from users of the banned subreddit r/ChapoTrapHouse. It operates on a modified version of the Lemmy platform, focusing on socialist discussions and content.
getting head from c-3po I call that golden dome
According to ChatGPT, there is no way a bee should be able to fly. Its wings are too small to get its fat little body off the ground. The bee, of course, flies anyway because bees don't care what humans think is impossible.
It's much better than them pretending they simply know this information. We should encourage people to be open with their sources and not get mad at them when they say something like that. Otherwise they will just ask ChatGPT and not admit it.
They could also just not post, because if I wanted a wrong answer from a robot, I'd have just asked the robot myself.
What bothers me is when people post long outputs from LLMs and expect me to actually read it. Seems rude to me.
LLMs are wrong around half the time. So there is some value in asking it, depending on the question.
That's worse odds than a Magic 8 Ball, which at least sometimes admits it doesn't know
The Magic 8 Ball has that failure rate with yes/no questions. LLMs can achieve this rate on open ended questions, which is much more impressive IMO.
Source?
What do you mean?
I am being a weenie. :P
I fucking cringe so hard when people are like "i asked chat gpt about what's holding me back in life and it came back with great answers"
Bitch it's fucking astrology. I can come up with shit that vaguely makes sense for most people and you'll think that it's "so true" and "it knows you better than you know yourself"
This should be an instant ban on any part of the internet. I've seen it on Lemmy before.
My friend pulled out her phone to ask chatGPT how to play a board game last night, and despite all of us yelling at her that chatGPT doesn't know anything, she persisted. Then the dumbass LLM made up some rules because it doesn't know anything.
If you tell people that ChatGPT doesn't know anything, they will only think you're obviously wrong when it gives them apparently correct answers. You should tell people the truth -- the harm in ChatGPT is that it is generally subtly wrong in some way, and often entirely wrong, but it always looks plausibly right.
Yea, that's definitely one of the worst aspects of AI is how confidently incorrect it can be. I had this issue using deep seek and had to turn on the mode where you can see what it's thinking and often it will say something like.
I can't analyze this properly, let's assume this.... Then confidently spits an answer out based on that assumption. At this point I feel like AI is good for 100 level CS students that don't want to do their homework and that's about it
Same, I just tried deepseek-R1 on a question I invented as an AI benchmark. (No AI has been able to remotely correctly answer this simple question, though I won't reveal what the question is here obviously.) Anyway, R1 was constantly making wrong assumptions, but also constantly second-guessing itself.
I actually do think the "reasoning" approach has potential though. If LLMs can only come up with right answers half the time, then "reasoning" allows multiple attempts at a right answer. Still, results are unimpressive.
one of my friends did the same thing and it provided incorrect information about the games rules lmao
My friend pulled out her phone to ask chatGPT how to play a board game last night, and despite all of us yelling at her that chatGPT doesn't know anything, she persisted. Then the dumbass LLM made up some rules because it doesn't know anything.
Do you think they took home the lesson that llms don't possess knowledge or do reason?
Why would she take away that lesson? It produced a list of rules to the game that look approximately right.
Presumably her friends corrected her and showed her why the "generated" rules were incorrect... at least that's what I would expect of my friends
I hope so. You have to be patient in circumstances like that.
i fucking hope, but she didn't really pay much attention to us lol
she didn't really pay much attention to us
Why do people do things like this? What is the point of playing a game with your friends if you won't listen to or pay attention to them?
I can't remember who it was, but some PhD in anthropology (I think) was tweeting about his research and some fucking idiot was arguing with him because ChatGPT said something different. Fucking bong-cloud epistemology.
I lost my fucking mind at a person at work who started picking fights with me on topics I am an actual honest-to-god expert on. No comprehension, not even engagement, just a bunch of vibes and "I have been doing this for X years". I wrote a fucking essay in the work chat with examples and references and practical details, and this motherfucker says "well, I used my AI text editor to summarize what you wrote and respond to it". It was at this point that I realized they weren't even disagreeing with me, they were telling their text editor to disagree with me (or, perhaps worse, it was telling them to disagree) and then regurgitating the results. AI freaks are literally inhuman.
view more: next ›