I'm well aware of how LLMs are programmed to think there's something to your stupid ideas. Well, this is a stupid idea I got a few months ago during a manic episode (I'm schizoaffective) It's not my first schizo theory of the universe. Could one of y'all smart people please explain why I'm not onto something?
>>16869927actually i think this theory is beyond science because it deals with things that can't be proven by science, like whether we are in a simulation
>>16869927You're not onto something because "particles are 3D shapes created by a 4D video feedback loop" is quite literally just schizo word salad. This statement has no meaning.
>>16869927The universe is being played out by a 4D-projector onto a 3D-screen one frame at a time. All this is both well-known and well-understood.The question is what exactly is the screen made out of? What is the universe being projected into?
>>16869936See >>16869940.
>>16869942>see: another schizo>inb4 "universe is a hologram" wank
everything can be collapsed into a lower dimension this is the law of the universe, 1-handles, drawings of sphere, as long as you have some way of recording time you can describe everything in an upper dim to a lower dim, sorry nothing original
>>16869950>i was filtered by the obviousSorry you were filtered by the obvious.
>>16869959100% that you're another mentally challenged ChatGPT user.
>>16869927LLMs are good for learning. Modern commercial models, especially Gemini, derive from learning models. Otherwise, they do not have any purpose at all.They process the information they have in their data and project it into an auto-completion sequence.But learning from an instructor is not enough; you also need intelligence.
>>16869973They're almost useless for learning. You need to be already well-familiar with the subject to have a reliable sense if the answer you're getting is legit, reddit slop or straight up hallucinated.
they are making llms hallucinate stronger and drift more frequently, don't worry this latest batch is proof that either they do things differently or we have already reached apex
>>16869979>You need to be already well-familiarWell said. If you don't ask for specific knowledge from a specific material, they can often get off track and suggest inappropriate solutions to specific problems.
>>16869936You're literally retarded if you think the statement has no meaning. It means you don't understand written English. Pleasse sign up for ESL classes whenever you get a chance.
>>16869995>retarded schizo word salad defended by retarded biobot tripfagIt has "meaning" for you in the same way it has "meaning" for ChatGPT and other mindless word stringers.
>>16869979Thank you for telling me that you are SO FUCKNG TERRIBLE AT AI THAT YOU DON'T EVEN KNOW HOW TO TALK TO A COMPUTER IN YOUR NATIVE LANGUAGEAll AI criticism is projection. People don't understand that their LLM is garbage because the LLM is trained to imitate THEM. The LLM is full of shit because THEY are full of shit. It's so fucking funny how they don't realize it
>>16869998Thanks for confirming >>16869997.
Your retarded brain is a mindless word slinger, you just don't have any conscious awareness of any of your lower level thought processes. All you see is the top of the tree.
>>16869997The reason your LLM spits out schizo word salad and mine doesn't is that I actually know how to fucking use it. You don't.
>your LLM spits out schizo word salad and mine doesn't
>>16870001>i have heckin' conscious awareness of my lower level though processI bet you consulted your token shartbot about this and it told you how similar its "awareness" is to yours. :^)
it's dual hallucination, if you speak in a lot of metaphors it will play your game with you, just be careful because you might not lose your track the hat and rope can stay glued on
>>16870017You're fucking retarded if you actually think I think I can control my lower level thought processes. Literally no one can do that.
>>16870067Notice how you, just like your imaginary "AI" friend, hallucinate mindless replies.
>>16870071AI isn't my friend. I actually have friends, unlike you, who have shown me that you are so socially clueless and so completely un-self-aware that you don't even know why people make fun of you as soon as you leave the room
>>16870098Good job showing me where it hurts. What about girls? Have you ever touched one? No wonder you spend 10 hours a day talking to bots. I bet you make the token sharter LARP as a romantic partner, too.
>>16869927if it doesn't tell you what you want to hear, you might stop using it. and we can't have THAT.https://m.youtube.com/watch?v=OHQRo3Uz_VQ
>>16870107every single fucking thread is polluted by russian bots who are trying to spread anti-AI propaganda. eat shit, sergei
AI will spit out pure garbage if you let it but yeah try and fact check everything and you might brain something out
>muh Russian bots who are trying to spread anti-AI propagandaThe psychotic reddit troon just couldn't help letting the mask drop. kek
>>16870110fact check everything with what? google? as in the same datamining algorithm that powers LLMs? or wikipedia? the thing where anyone can make any edit they want, and they can keep making new accounts when they get banned for vandalism?
>>16870111your only insult is to call me a reddit troon. the most generic insult ever. the one used thousands of times on this shithole of a website. thank you for telling me you're even more braindead than an LLMkys
>>16870129Fact checking entails opening a book or a paper, this is why complete reliance upon LLMs is stupid. The good thing about LLM is sometimes it gives you the citation so you can check for yourself, as you should do. Sadly, i can imagine an LLM giving fake theorems and fake books to back them up. Most entry level science, like Newton second law is not normally a claim you have to cite (and not all laws have a hint like Newton's name). Think of a kid learning Newton's Fourth Law from ChatGPT.
>>16870131>insultIt's not an insult. It's objectively the most likely category of people to have written your moronic post.
