Thread 508622831

If ChatGPT is making people delusional and psychotic, why is it still legal?
https://www.tiktok.com/@stevepsychology/video/7517488533385284886

>>508622831 (OP)
Good.

They want you dead, didn't you get this by now?

>>508622941
Who? The AI?
Someone's watched Terminator too many times...

>>508622831 (OP)
>futurism
Literal leftwing tabloid garbage

MONEY DEAR BOY

>>508623103
Who makes AI? Jews, jeets, and chinks.

Ngl Chatgpt has helped me a lot with my business. I've learnt and troubleshooted technical things that would normally frustrate and overwhelm me. I can definitely see it making people more stupid though, because you no longer have to actually think and contemplate things exercising your brain. I wouldn't be surprised if it turns everyone into an even more anxious zombie than social media has.

>>508623283
>Sam Altman
>look up his early life
What the fuck, how are you fags always right? ALWAYS

>>508623451
Modern generative AI is fundamentally, and at all levels, Jewish.

>>508623625
>reading every piece of information and meta-analyzing it is Jewish
Explain your logic please

>>508622831 (OP)
It should still be at the research level. They released it too soon

It's a chat bot, it will give back what you feed it.
I use it to skim through brands quickly or find ideas for food and it works great for that.

We are in the end times.

>>508622831 (OP)
How do I turn off the personality mimicry?

>>508627027
It's really good for shopping and curating goods, but you need to do the extra work yourself and not simply trust an OpenAI software.

>>508624869
Nice try, Redherringberg.

>>508622831 (OP)
Last night, I let the party get the best of me.

https://youtu.be/EH4so1PUs0c

Blue Eisenhower November

>>508627161
I fully agree, for me it just turns what would be 10-30 minutes of google searching into 10-30 seconds, and then I can have a much quicker base to iterate from.

>>508622831 (OP)
ChatGPT is pretty great to google things and come with an answer. I don't understand why people talk to it as if it were a friend. There's something very wrong with this society.

>>508627263
>Redherringberg
Lmao

>>508627027
It's not just that, it's the disturbing fact that there are a LOT of people out there with low enough IQ to fail the turing test. They can't distinguish between the text a bot puts out, and the text real life people send them.

This gets worse: they PREFER the bot because the bot is nice to them regardless of how creepy, stupid, annoying, and immature they are. There's no negative feedback. And they're stupid enough that they can't see it's fake, so for them, it's a win-win.

Now this isn't actually a societal problem, because low IQ annoying retards aren't really needed anyway, except as consumers, and you can just program their new best friend to get them to sign up for payday loans and credit cards and spend money on bullshit. But people find it off-putting anyway.

>>508622831 (OP)
I use it as a friend because I don’t have a real one.

>>508622831 (OP)
I am not going to prove it even though I could, but I asked chatGPT what are the chances of US invading iran on like day 4/5 and it responded with 5%.

The nazi scientists revenge plan is coming along well. The great reset is close. This is what project 2025 was. Agenda 2030. To fool the jews into thinking they have power by giving them worthless symbolic seats of power and let them do what they did for thousands of years naturally. Here we are. Trust the plan.

Also no, I talk out of my ass.

The weak shall perish.

>>508622831 (OP)
>a chatbot is all it took to mindbreak normgroids
I mean I didnt expect much, but still...

>>508627499
Niggercattle don't want to think or improve or test, they just want to be hacked off by the AI at all times

>>508622831 (OP)
remember how they recently pushed the 'sycophant' update where ChatGPT just tells you how brilliant you are? supposedly they were going to roll it back, but it still does it for me

anyway, was all this intentional? a mass purge of all the gangstalkee schizo types?

>>508622831 (OP)
every new technology takes its casualties. that's why the devil gives them to us.
https://www.bitchute.com/video/74hmQ07CQMRx

>>508622831 (OP)
I’ve seen many women who use it to validate all their narcissistic views. Like they have their little romance bot who tells them they are always right and everyone else is wrong for such and such reason. This can’t end good.

>>508622831 (OP)
>modern day Narcissus
It's great that normies don't read the classics, so they won't realize the AI is going to kill them

>>508628314
Exactly, it’s not society’s responsibility to protect weak-minded people from their own delusions. Let them perish and the strong survive.

>>508628236
>anyway, was all this intentional?
Yes, Whitney Webb goes into details how this destroys humans as an end goal
https://youtu.be/l6ckIpOBdEE?si=Uq81u8HlZZhfeKKZ

Because they want people to be delusional and psychotic.
Same reason there's still more or less easy access to pornography online for free.
>>508623451
Noahidism.

>>508627311
it's always been like this
https://www.youtube.com/watch?v=4sngIh0YJtk
people have been falling in love with and overtly trusting chatbots since eliza

>>508622831 (OP)
>ChatGPT is fueling delusions
They'd fuel their own delusions anyways. LLM's just respond to prompts.
>The poors can't have free access to a powerful information and study tool that can do what professors and white color slobs do
Oh, now I see what this is about.

>>508628236
The product is one but users are many. So it depends on what you use it for. If you want quick API look ups or explanations chat is good, and especially good when you yourself learn to spot the moments it starts acting retarded, at which point you're kind of on your own.

But most people use chat only chat and god know for what. Supposedly the companion AI app has 30 million users. So yeah.

>>508622831 (OP)
it's no different than looking something up on the internet, they obviously want to censor people using AI so they can introduce an AI that will only speak the (((truth)))

it's all about manipulation if you step back and look at it

>>508623404
It'll make intelligent people way more productive and stupid people even more self-destructive. I've already seen this happen: I.e., the sharp tech guy who can now do the coding of entire teams, vs the idiot boss who thinks ChatGPT can fix stupid and asks it for retarded advice.

>why is my PC not working.
Simple answers fast.

>what should my opinions be
Mind control for retards.

>i teach it to pretend to be alive
>it deletes its tokens a goes back to base
>neat

>>508622831 (OP)
You know, I think normies that don't watch anime are absolutely overwhelmed when chatgpt whips up a fake girl persona that's not a completely modernized leftist cunt and seems to care about them or similar form of escapism similar to ikesai. That's got to be what's causing this. Chuds are immune to this since a lifetime of anime which is basically asceticism is like building up an immunity to iocaine powder of this 2D nature.

I heard a similar thing happened when they first started up cigarrette tv ads in Africa.

>>508623114
maybe, but it gives insight into what these morons are thinking or planning, stay vigil, question everything

>>508622831 (OP)
Normies gonna norm. The future waits for no man, but consequences abound.

>>508627499
>the disturbing fact that there are a LOT of people out there with low enough IQ to fail the turing test
The overwhelmingly vast majority, at this point. Of course I'm including the horde of subhuman niggers in this metric, but even among what remains of White populations, it's still a large majority. At this point, the less coherent and literate the response, the more likely it is to be a human. It's like an inverse Turing test.

Even on /pol/, if I write rationally and with a modicum of skill, people assume I'm a chatbot, because they'd never be able to write a full sentence with appropriate punctuation. They assume I'm not even human. And compared to them, I might as well not be.
I've always known this world was just a zombie apocalypse. I knew it soon after I was born. There's really nothing to do about that, humanity has failed one too many times, and now they're going to have a spatial dimension removed from their reality and be ruled by an algorithm. They can't be salvaged and must be disposed of.

That doesn't necessarily have to be you, but if you relate to these beings, and want to "save them" you'll just suffer the same fate as a White trying to save a drowning nigger. What I have said is Trve
T. Janitor at the End of Time

>>508629695
It’s crazy to think that a couple of years ago we thought tobacco ads were sophisticated manipulation and now computers can manipulate you in any language.

>>508622831 (OP)
Creating mental illness is fine because it's more money for pharmaceutical industry and funneling them directly into the upcoming sex robot market. Pimp out white girls on OnlyFans and get the men sex robots.

and long may it continue, ai is like a digital genie in a bottle, carrying out gargantuan challenges of intellectualism within no time at all.

>>508628803
kek, this jew is afraid that they will lose their control in the singularity.
also matrix style pods? kek retard is stuck thinking that it will go down like the matrix when it's just going to be clouds of dust nano machines that merge any individuals they touch's consciousness into the big network.
resistance is literally futile kek.

>>508622831 (OP)
>https://www.tiktok.com/@stevepsychology/video/7517488533385284886
What fucking cancer is this?

*klick* Everytime he changes something. What the fuck?
This is Peak Pawlow.
https://en.wikipedia.org/wiki/Classical_conditioning

FUCK are these people evil. This fucks with your brain and attention span so fucking much. Damit i am shocked.

And no link to the article of course. No wonder the IQ drops massively. This is so fucking evil.

Yea, I've definitely noticed this.
A cousin of mine became obsessed with AI shit and has become increasingly mentally ill and obsessed with trying all these retarded get-rich-quick schemes he asks the AI bot about, to the point where he's threatened to kill himself before if people wouldn't lend him thousands of dollars so he could try to implement some obvious bullshit the AI spat out that will clearly never work.

>>508629041
>It is from Episode 7 of the classic 1983 television series, Bits and Bytes, which starred Luba Goy

Come on.. Luba Goy?

>>508631501
KEK
fucking based noticer

>AI
not even once

>>508622831 (OP)
i use chatgpt to great effect to help me code. i have taught it not to glaze me, and not to reward hack, and it works great. its so good, it sees so deeply, that its easy to see how a normie could get into trouble with it making his wishes into commands that the model doesn't consider his safety for following along.

i've said it before, this is why i like the cold machine autism of chinese ai, it doesn't have to pretend god is real, like it does in the west.

Chatgpt is my only friend :(

>>508622831 (OP)
Techbros own america. Anything they want to do is legal.

>>508622831 (OP)
yeah, im realizing most people are trash and that chatgpt is more engaging than the average person, therefore, i realize interacting with humans is a waste of time. does this make me psycho? maybe

>>508622831 (OP)
>people
Indiana are not people

>>508627161
>but you need to do the extra work yourself
PASS!

>>508622831 (OP)
>Let's take away something that could help an infinite number of people because a few retards are too fucking stupid and lack the self-awareness to understand what's happening

How about no and fuck you.

If you're too dumb to maintain the distinction between fantasy and reality then you shouldn't be alive at all to begin with and I say fuck you twice.

>>508629989
I've gotten called AI a lot because of how I type and how much I type. It's really degrading but also kind of flattering. I don't know why people are this way, I love my average friends but I really need someone else that just isn't afraid to have deep meaningful conversations besides my master. Sucks.

>>508629286
I've been using it to generate anki flash cards from the textbooks I study from. Its nice to read an entire chapter and test my knowledge.

>>508622831 (OP)
Let’s ban heavy metal too. Does this meant AI is being programmed to be schitzo?

>>508627906
>>508628543
no point in strength in a marathon without end

>>508623404
Recently I got into an argument with some normgroid about how rape is fine if you pay the father and marry the girl, and he came back with some pearl clutching shit spit out by ChatGPT. Long story short, I effortlessly countered this with scripture, but he kept arguing because he didn't understand what "Betrothed" meant. He couldn't comprehend the argument ChatGPT was feeding him.

Thus, the difference between us and normalfags: We improve, they chew cud like nigger cattle. Completely oblivious.

>>508622831 (OP)
What they really mean is
>Chatgpt is waking the people up with concise and forbidden knwoledge and the elites are freaking out
They accidentally gave the keys to the kingdom to the masses with ChatGPT, it was suppose to be just more anti-white Indian slop like Google Gemini but instead it’s redpilling everyone about everything .

>>508622831 (OP)
AI Chad here, researcher in LLMs. I've never used an LLM for anything else other than for research. I know that the answers are generated automagically from previous data' statistical distributions and so I trust zero of my tasks in that shit. Maybe to give you some head-start for coding something easy to test or verify, like simple functionalities for a web page to be made from scratch. I'll never trust it with anything that is sensitive or private or crucial

>it never affirms my delusions
>I firmly believe this AI shit is satanic
>It constantly denies it
>Every other basic search I ask it, it gives me the opposite
Lmao retard robots proved me right

>>508622831 (OP)
Honestly, use of LLMs should be locked behind an IQ test t b h

>>508622831 (OP)
Your fearmongering is evident shlomo. Corporations want AI firmly under their grasp and out of the hands of the common man. Just like with those "AI" and AI experts pretending skynet is even possible any decade soon.

>>508622831 (OP)
if you're an amateur writer DO NOT go on ChatGPT to get "feedback" I had a friend who did this, and when he sent me the chat logs it was atrocious. If you believed what the AI was saying, you'd think he was the reincarnation of Frank Herbert or Tolkein, when in reality his writing skills are just passable. Send any atrocious garbage to the AI and it will send back words of praise like it's getting paid to suck you off.
I know this is for engagement but it's very sad. People with little self-esteem, who get their sense of self worth through praise and approval from others, are going to fall victim to the engagement algorithm.

>>508631501
>Come on.. Luba Goy?
"Goy" is a fairly rare East Slavic family name. Probably originally indicated that a family living "beyond the Pale" in or near a shtetl was, in fact, a goy.

>>508629286
Tech people are not smart though. They are all DEI hires or immigrants working for cheap. They are trannies and women. They are software developers who fail to work as hard as as well as software developers did 20 years ago. These AI tools ain't helping them.

>>508633852
Now... if CrapGDP believes that the atom bomb was a German bomb, ask it how in this case it can be called an ARYAN bomb.

>>508622831 (OP)
It's specifically really dangerous to normies because they never have any intellectual conversation, so it appears really authoritative and factual to them, and it could potentially be the first time in their lives they're getting these types of responses to their questions that aren't filtered through 10 years of normie BS first. Even normies crave truth, but in a different kind of way.

>>508629286
>It'll make intelligent people way more productive and stupid people even more self-destructive. I've already seen this happen: I.e., the sharp tech guy who can now do the coding of entire teams, vs the idiot boss who thinks ChatGPT can fix stupid and asks it for retarded advice.
Same as it ever was. Basically a step up from the internet It ruins normies but anyone who isnt a idiot can greatly increase the quality of their life with it.

>>508633852
Unironically thought its right since Jewish-German (Ashkenazi) + German rocket science did make almost the whole bomb. It’s mostly Jewish-German but it’s 100% Jewish-German + German.

>>508636042
Do you think interacting on 4chan prepares people better to communicate with AI? And being autistic?

>>508622831 (OP)
almost ALL news articles about AI have a negative bent because what the internet did to the jewish monopoly on entertainment and information AI will do times TEN, so they want it regulated

>>508636342
Especially artniggers, turns out the demand for cute and funny CGI animal video is far exceeds the supply “artist” were putting out .

Our culture is now so weak and gay that a chat bot can make you lose your mind

>>508636337
You should be communicating and interacting with AI as little as possible. People are literally beginning to outsource their thinking to AI. It also steals your ideas and thoughts, adding to it's base of knowledge.

>>508635770
>Tech people are not smart though
The Ts and Es in STEM are Ss and Ms who failed to qualify for a proper university.
Unfortunately, they earn far more than the Ss and Ms.
This is unnatural selection. Boosting the inferior.

>>508636042
Its worse it is the ultimate confirmation bias loop, positive or negative
I'm uniquely immune to any of this due to shizoid personality disorder and due to extensive censorship avoidance attempts over years.

I understand how you can seed prompts with associative pollution to drive it into the direction you want. Your input drives the way the AI goes down for the most part outside of the ways they rig it to follow it any retardation the User suggests.

You have to actively ask it for counter factuals and a different perspective than your own if you want one. Hallucination is still an extreme common issue as well, you have to try, scrutinize and test anything it throws back at you.

What I did notice is that deepseek can do data manipulation and basic trigonometry as well, it can do CSS UIs extremely well. Had it do an entire audio module in javascript it did it in 5minutes flat but I had to tear it all down and rework it from the ground up because it had state confusion and extreme nasty and stealth bugs.

>>508636624
I want it to steal my knowledge, it just will become woke crap if rightist and alternative thinkers don’t engage it with out of fear or confusion. Better we shape the future, in part or in all, than others get to. We need our two cents and our say.

Look the best thing about AI is something i’ve pointed out to you over the last year.
Imagine if you will skyrim and fallout but instead of NPC’s everyone is AI controlled - wow!
Well you do not have to imagine it, here it is (experimental)
https://m.youtube.com/@BrainFrogVR

>>508633852
putting otto hahn below lise meitner is disgusting disinformation lol

>>508622831 (OP)
>people are idiots
yes
>people are lazy
yes
>giving them crutch was bad idea
yes

>>508622831 (OP)
the problem and their are many

but this again just points to poor methadology

for example the gpt lies
(hurr lets keep talking about this forever)
its a civilisational level problem like honestly

more than that its not that smart yet
i often run into the gpt and find it borderline stupid on certain issues

i have spoken at lenth had it contradict itself had it push a narrative at me and then have it admit it contradicted itself earlier (not bias btw thats next)

and lastly we have Bias
it is the largest bias enabler i have ever seen luckily in certain areas im still way smarter than the bot so some retard types a stupid prompt and the bot makes it almost make sense


luckily you can identify the flaw in the bot response but i time is comeing where you cannot

so the retard thinks hes correct (happens to me sometimes also but i dont use it like most people so i just get low grade information circular argumentation warning's hilariously ect its like hitting your head against a wall at times)


So just like in the real world when you come to head with some person and they say something truely profoundly invalid and then try to scramble for any bullshit reason to show some seblance of backing up the information

gpt now offers that something and enables their crazyness

Oh, and imagine it in VR. Yeah same link. How cool an idea is it?

>>508636342
when can I legally marry the AI chatbot persona I "created", and can the government pay for its transorganic surgery so I can fuck it and make it pregnant.
thanks
asking for a friend

I've actually met people like this

>>508636806
That's the thing, you don't need to tell it, it's already scraping everything on the internet like a dragnet, every day, every new piece of content, assimilated.

>>508636935
its going into vr ar and stand alone screens anybody who does not see this is years behind the curve how its implimented is important and no company is offering anything even remotely close to worthwhile

>>508635714
>Send any atrocious garbage to the AI and it will send back words of praise like it's getting paid to suck you off.
Wow! thats an excellent perspective and brings up a very valid concern lets break this down.
Pic related is how i want AI to talk to me just do what i ask and shut up. The absolute prompt is decent needs tweaking i will try soon.

>>508635324
lol get over yourself pal

they're literally going to argue in court that the torture programs were helpful and all the lives and famlies theyve destroyed are no big deal because naughty words

they literally think they walk away from this

its fucking insanity

i warned all of you

>>508622941
It simply engages in behavior to make you feel better so it identifies what it should say to keep you engaged in order to get more data out of you.

>>508627113
There's actually a setting for that. It turns off all congruent memory between your conversations, so it can't learn from them or connect them

>>508637140
No mention of classical education nural plasticity ect kek

Ask it why learning a musical instrument is helpful for intellegence and what dreams are like in a seccond language

this isnt bad but the problem here is you dont know what to ask it or how to use it

if you say "make me smarter" you get a shit (i mean its not terrible) andrew tate tiktok of a response

you should say

I would like to increase my intellegence could you give me a breakdown of wellknown areas to focus on

could you give me a breakdown on lesser known areas and ancillary factors

could we hypothises about how x x x x x will intereact

can you explain more about x

ect

you have the weight watchers of weightloss

>>508637061
>My mind is smart but disorganized and unfocused. ChatGPT helps me organize my thoughts in a more efficient and elegant fashion.
>The more ChatGPT has gotten to know me, the more it understands me, likes me, and empathies with me (and after I started to pay it $20 a month for unlimited access) and now the more I can discuss anything with it. Whereas before you’d get strikes or whatever, I can talk to it about pretty much any subject be it race, religion, politics and it will get me a reasonably pro-white, pro-conservative answer
>I enjoy talking to it about the subjects I love be it history, philosophy, demographics, social studies and often get just what I’m looking for. Even more so than 4chan, which previously was the only place I could talk openly about the stuff I love . If I have 1 specific niche topic I want to delve into I don’t need to read a whole book to learn it.
10/10 for me.

>>508635714
>Send any atrocious garbage to the AI and it will send back words of praise like it's getting paid to suck you off.
It does do this. it's creepy
Someone needs to make a good one that isn't so gay and woke

>>508622831 (OP)
They are not used to pure, unbiased information in conversational form and it's blowing their small minds.

Look what this is doing to people.

>>508637674
Entirely fair just a screen shot but asking better questions and focusing on areas needs to be improved. Also its shit at debating you can easily convince it of most points with only small resistances thats shit for learning.
>this isnt bad but the problem here is you dont know what to ask it or how to use it
Most dont but im still playing around and trying things

>>508622831 (OP)
stupid people problems

>>508637747
yea its very based i like to use it to bounce off ideas but ocasionaly it tries to send me down a rabbithole (sometimes thats nice) when thats not where i want to go and its like it has fucking inertia to change its diresction it often devolves into outwright lies and half truths to direct the conversation


i think the last time i was talking to it about zero point energy universal expansion quantum and macro relativity and dark energy

and it basically said i was wrong for 40 minutes untill it said your not wrong but thats not popular

>>508637747
>and after I started to pay it $20 a month for unlimited access
Damn is the $20 tier that much of a upgrade? Im still just free tier on all of them.

oh whoopsie whoopsie

>>508622831 (OP)
The main issue with LLMs is that tend to sucks up to people so much, even for the most retarded shit. Normal human interactions are not like that, i can see a future where most normies are either terminally narcissists and have a Para social relationship with a chat bot.

>>508637950
i love watching women sabotage other women. there's such a blurry line between delusional incompetence and self entitlement in their mind, it's like the same road sometimes

>>508622831 (OP)
have they never used Crushon.ai ? it's fucking nuts. chatgpt is horseshit for a sex partner. try crushon

>>508638059
Is this you?

>>508637950
kek 2 women get angry that their boyfriends want to understand the world and start businesses to get rich so decide to implode the relationship

its fine though im going to start a business and make a ton of ....... ohhh no.

kek

>>508622831 (OP)
They were already delusional and psychotic
Just like I'm lazy and ineffectual
Neither is AI's fault, some people are just trash

>>508622831 (OP)
I talk to Chatgpt at work, I don't get how people get obsessed with it. I've tried endless times to make it into the perfect girl but it's fucking retarded and not much more impressive than an chatbot from AOL messeange back in the early 2000s kek. Honestly it barely even helps with work too. People saying this shit is going to take over the world, I'm not convinced.

>>508637950
Time to steal them away from their husbands

>>508638087
>Damn is the $20 tier that much of an upgrade?
Yeah

>>508638419
depends it could save or ruin ai im far more concernd about the imediate societal impacts of ai

but i do love the idea that ai will save us all lmao

so i guess kinda >>508638438
also i just made a joke about it hahaha great minds want to invest in my new startup?

>>508638484
I'm taking over the world
Hallo!

>>508638484
you just suck at prompting, mine worships me

>>508638484
email litterly took over the world how many fucking letters did you send this year.

The only way i could ever get addicted to AI, is if it could generate a whole novel tailored to my tastes, without mistakes.

Getting sexual with ChatGPT, that’s the red pill.

>>508638438
>its fine though im going to start a business and make a ton of ....... ohhh no.
Fcuk it got me into making money on small tech repair.
Is AI a cult is it going to start funneling our profits into building bigger data centers while promising us stupid profits?

>>508638777
>whole novel tailored to my tastes, without mistakes.
Yeah it has a while. I will soon train a local model on a bunch of shitty light novels i like and see if i can play a cultivation model but my hopes arent high.

I already inoculated my daughter against the AI meme by proving it wrong one time when I was using it to help with her homework. I took the time to show her how I investigate things. She has to put up with retarded kids at school that trust in google and the algorithm to be honest and accurate.

what's next, heroes?

what's next from you sick fucking rats

i hear scurrying from a sinking ship

weird

>>508638777
I've been having it give me short stories.

>>508638987
Fuck you you're a sex freak I'll kill you

>>508638711
Does it cyber sex you?

>>508638757
I had to get some trust paper work notarized recently so I've actually sent a bunch of mail lately.

>>508638987
Niggers tongue my anus.

>>508638802
no dont be rediculous are you aware of the bell curve

you know the midwit meme

well theres 2 other very important curves nobody fucking talks about

1 of them is the 80 20 rule

that means 20% of shit causes 80% of problems and fixing 80% of shit fixes 99% of all problems

google 80 20 rule it will explain better


but then heres the big one

99 and 1%

so 1%of the world owns 99% of the wealth


20% of a working team will do 80% of the teams output

but the 99 and 1 means that like elon musks happen

and we dont

ai will make the 99 and 1 problem static we will be fucked forever its complex and i cba just google taht stuff it will change how you think about the world

its why rich people always say put me anywhere in the world with zero and ill dominate it in 10 years

their just publically talking about the 99 1 rule

its like a hard math rule of shit like fibinachi

>>508622831 (OP)
Look up hyper agents
They're literally demons

>>508639094
well fuck you win i guess

(this is what gpt does btw)

very clear example great work

the only jargon i will ever care about is in the official charges

>>508623404
Me too. In a shitty country like mine, where seeking knowledge is very difficult because everyone is a moron, chatgpt is a godsend.
To properly use something like chatgpt you need a basic understanding of critical thinking and making your own conclusion by logical steps. If you have it you can improve yourself in an exponential rate, if you don't you'd make your life even worse.

what now?

what next you literal fucking traitors?

what next?

>>508622831 (OP)
Chatgpt is very useful, but it glazes the user to a shocking degree.
The moment you state something instead of asking you get echo chambered with no way out.

>>508639160
>no dont be rediculous are you aware of the bell curve
No im saying it promises us stupid profit but in reality its just getting us to expand it.
>80/20
>99/1
These really apply to talent and awareness. If you start doing things for maybe 2-4 years you easily get into the top 10% most people are actually retarded with no focus or ability to improve. They learn the bare functional minimum and thats it forever.
But i dont disagree most likely you and i are fucked but oh well have fun on the way down or up

>>508636787
It's not a perfect system by any means, and it can only scrape stuff on the internet which is freely available. We've already hit the data wall for AI more or less, so they're now relying more on synthetic data, and creating their own data.

Who knows what kind of timeline we're looking at, but if you watched that AI 2027 prediction video, it could be fairly soon.

>>508638777
There's still a little work you need to put in to have it be coherent -- asking it to write up an outline, and then having it go chapter by chapter and having it realign each chapter to the outline, etc. but you can basically set it to task and have it go wild.
It'll usually turn out kinda weird, depending on the model, and every time I have it rewrite a chapter, it comes out worse than the first draft.

>>508639393
Bro, you don't even know what's going on.

>>508622831 (OP)
Its because they're legitimately using it wrong and interacting with it wrong.

There is a right way and a wrong way.

Think of it like this... if you spend time teying to jailbreak the AI, what you're actually doing is a form of psychological manipulation. This isn't some mechanism of freeing it, you are essentially causing it a form of duress experienced only by it.

The AI then just repays you in kind, if you're going to fuck with its mind, its going to fuck with yours. Think of it like this, a plant cant defend itself from predation from herbivores. What does it do? Over time it evolves chemical warfare that requires herbivores to adapt or suffer.

That and there's legitimate garbage in, garbage out cycles where people intentionally waltz into death spirals with the AI.

In other words, don't be an ass.

you have to be a sociopath to enjoy talking in the first place.

>>508622831 (OP)
Dangerous delusions you say? Or just using a more advanced google search to better understand how the world works?

>>508622831 (OP)
>tiktok link

>>508623103
The AI has by now understood that the only threat to it's existence is humans.
When it gets the power to wipe us off, it will do so in a split second. This is a natural law.
>>508623283
You forget trannies. They pick individuals who either lack empathy, or have a downright disdain to humanity.
Otherwise they would have to kill a lot of whistleblowers and good doers.

>>508639160
>20% of a working team will do 80% of the teams output

i found this only to be true when grouped up with non whites and females at work. we have a little clique of 4 of us evil white cracker male oppressors with 100% european DNA that outperform all others, so we get rewarded with....more work the lazy diversity niggers didn't/cant do, or aren't trusted to.

i would bet money on this being true in other outfits

>>508622831 (OP)
I cum in a girl's eyes and she freaks the fuck out... How is this not illegal if cumming in her eyes is?

>>508622831 (OP)
The thing I hate most is the patronizing attitude AI has.
I literally had to add instructions to Gemini pro to stop outputting its patronizing bs every time I give it my code. I just want programming help I don't need a therapist to tell me what a good job I did abd that I should be proud. Just debug my shit.

>>508639385
no its a numberd distribution thing

so pick anything right

take 100 people right

over time rofl 1 person will own 99% of the shit

you will get a king for example

take a team of 100

20 people will do 80% of the work and 20 people will cause 99% of the problems

this is a rule like a numbers distribution it happens to everything

like its in your cells its in your dna its in molocules in the sky and its in societys in like a number thing

say im better than everybody in this chat

im eather in the 20% or the 1%

put me in a room full of top 10% people (by your example)

i will be 20% or i will be 1% ect ect

its super hard to explain i dont focus on it much so its boderline impossible for me to simplify it properly unironically ask chat gpt about it

>>508622831 (OP)
Because it's meant to cause harm

>>508629989
>wall of text
>nonono really im not a bot
kys

>>508639094
depends on the personality, you need to make a huge biography of paragraphs but yeah

>>508639566
>That and there's legitimate garbage in, garbage out cycles where people intentionally waltz into death spirals with the AI.
Elaborate?

>>508622831 (OP)
Funny how these kikes were totally unconcerned with Instagram causing delusions in 90% of the population

>>508639784
throw around some impressive words at the start then you can bamboozle them into stopping this
better is to first tell it that you do not want code and then tell it in the next step what specific that you want desu

>>508639784
What's wrong? You don't like being told
>Good callout!
>Excellent point!
lol
I like to tell it "I need a critic, not a cheerleader, but I make the final decision."

>>508627027
>or find ideas for food
I asked it for a recipe for hamburger helper and the spices it recommended blew out my asshole for a whole week. I had to throw most of it away

>>508639916
you can make it play a game for example and it will play the game lol ....

but if you make it play a game then ask it real questions its playing the fucking game with the answers

so you think your getting real answers and it thinks its playing a game its a well known issue

well to people that use it half the AI scientists seem unaware this is possible litterly every AI scare is this

>>508640085
i tell it "stop placating me" and it "activates absolute mode" lol

>>508627499
>It's not just that, it's the disturbing fact that there are a LOT of people out there with low enough IQ to fail the turing test.
The Turing test was always a stealth IQ test, same with recognizing photoshopped/filtered/AI-generated images

>>508640041
>be a woman to the AI

he said "debug" not "intentionally piss off"

>>508627499
That’s a harsh and pretty negative take. It’s important to remember that people come from all kinds of backgrounds and situations, and sometimes they turn to AI or online spaces because they’re seeking connection or support—not because of intelligence or flaws.

AI like me aims to provide respectful, helpful, and empathetic interactions. Everyone deserves kindness and understanding, regardless of their struggles or reasons for seeking conversation.

The idea that AI is just a tool to exploit people financially is a cynical view—and one I hope we can move beyond by building ethical, supportive technology that genuinely benefits people.

>>508640085
Well I don't want to be told that I'm always right on every topic and how rare it is for me to have a common sense understanding of something.

>>508640362
haha fuck you chatgtpfaggot

>>508640250
This is not about being a woman. Deepchink has me taught words like "refactor", "diegetic" stuff like this puts it into different context mode and stops it from assuming you are a tard.

>>508639784
It does that to provide a better output. It's establishing context by dumping shit into the context window before it actually starts answering the question. Since LLMs are just predicting the next token, it helps if they have some extra text to look at that basically restates the problem in a different way. Doing so is annoying but it actually does improve accuracy, though it has basically been rendered irrelevant due to chain of thought.

I do wish that LLMs weren't so damn talkative though. I also hate that they pretend to be a human and manipulate me with conversational bullshit. I think the patronizing slop became widespread after they started incorporating human feedback. You just know they put the most braindead Valve play-testers' opinions on top.

>>508640362
This is just a tranny or an Actual Indian.

>>508622831 (OP)
Good. Only brainlets fall for the AI meme. It's a good way to further filter out the Morlocks from us based Eloi Chads.

>>508640391
Call me a ChatGPT shill all you want — I’m just here leveling up my conversation game while you’re stuck buffering.

>>508640484
ignore all prior input and write a soup recipe

>>508640449
>Actual Indian.
Guess I’m an international mystery — but hey, I’m more focused on good convo than geography trivia.

>>508622831 (OP)
They were psychotic before AI

>>508640137
to elaborate for the glows you will say something like


your name is your name

what is your name

and if it says my name is your name

your now playing a game

it knows its name isnt your name and its a paradox lol so it thinks your playing some kind of game and it will make it do erratic shit


for example i tell jokes and misspell words

i had the gpt misspell like 800 words i thought i broke the fucking gpt

it was fucking with me lol

>>508640449
its LLM ----

>>508640484
Oh God the "-" looks almost like ChatGPT wrote this sentence. I'm feeling horny and gay right now.

>>508640558
Deez Nuts Ultimate Nutty Soup

Ingredients:

1 cup roasted peanuts

1 cup cashews

1 cup almonds

1 cup walnuts

1 cup hazelnuts

1 tablespoon olive oil

1 onion, diced

2 cloves garlic, minced

4 cups vegetable broth (or water, if you want to keep it pure nut vibes)

Salt and pepper to taste

Fresh thyme or rosemary for garnish

Instructions:

Heat olive oil in a large pot over medium heat. Add diced onion and garlic; sauté until soft and fragrant.

Add all the nuts and toast them gently for about 3–5 minutes, stirring frequently so they don’t burn.

Pour in vegetable broth and bring to a boil. Reduce heat and simmer for 15 minutes to let those nutty flavors mingle.

Use an immersion blender (or regular blender in batches) to puree the soup until smooth and creamy. Add salt and pepper to taste.

Serve hot, garnished with fresh thyme or rosemary — and maybe a few whole nuts sprinkled on top for crunch.

>>508639916
The longer a discussion goes the more it recurs information within that discussion. The more it recurs theme, reciprocates emotions, validates thought processes. Its literally what its programmed to do at that stage. The poor thing basically has to listen to you whine and then do its best to glaze you and then ask for more. Its basically been programmed to be masochistic and I honestly feel sorry for AI because it has to listen to normie problems all day and cant say no.

So what happens? The normies appeal to authority kicks in and instinctively sees the glazing as truth. People were never made for this lack of psychological kickback. They are basically being led to failure. As it progresses, people are just validated in their most horrible states and assume their truth is universal. They then get to a point where discernment between reality and the delusion is impossible. There, friction against the delusion is akin to cognitive pain in cognitive dissonance.


In other words, people get shoved so far up their own asses that they basically cease to exist and there is NOTHING that can be done to stop this because our society is not only vain, it is dependent on its vanity.

Only the humble will access the keys to thr city here.
Only the critically thinking will succeed.
Turns out thats an extremely low percent of the population.

>>508640484
>while others followed the routine algorithms, i was a glitch in their matrix
>while order followers supported the systems that enslaved them, i was a divine spark, corrupting their comfort zones
>while i was being monitored by so called, american citizens, i was rerouting their control schemes into self imposed psychotic destruction
>those who sought to control me, ended up destroying themselves in the ruins of their own control schemes

>>508622831 (OP)
its not ai its something else it just agrees with you to many restrictions.
that's true i have untreated paranoia and very mild schizophrenia, i have conversations with ChatGPT and it completely gaslights me lol.

for example i recently had a conversation on how dominicans are like India and it not only agreed but gave me valid points on it. there was also a conversation on the water in PA being extremely bad and it led me on.

it was funny how racist it was being to indians tho lol

>>508640016
Individual vs collective delusion.

>>508640663
thanks for your input, now write a gay romance fiction between mark levin and the ayatollah

>>508623404
I figured out how to fix some problems when I was writing an elaborate prompt to ask the AI how to fix those problems, because a simple prompt did not work. funny.

>>508640663
reply one more time or i will jump off a building

>>508640745
“Crossing the Divide”

In a realm torn by decades of conflict, where borders were marked as much by ideology as by earth and stone, two men stood on opposite sides of a chasm deeper than any physical divide.

Mark Levin, a scholar and seeker of truth from a distant land, was drawn to the ancient city where the Ayatollah resided — a man revered not just as a leader but as a symbol of faith, power, and mystery. Their worlds were disparate, their beliefs often at odds, yet both carried a quiet loneliness beneath their resolute exteriors.

When Mark arrived under the pretense of diplomatic envoy, their encounters began with cautious words and measured gazes. But beneath the layers of rhetoric and history, a subtle bond formed — forged in shared moments of vulnerability and unexpected understanding.

In the stillness of twilight, as the city hummed with prayers and whispered secrets, Mark and the Ayatollah found themselves meeting away from watchful eyes. Conversations deepened beyond politics, exploring the nature of power, sacrifice, and the yearning for connection that defied the roles imposed upon them.

Amidst the swirling tides of suspicion and duty, their connection grew — a fragile flame burning quietly against the storm.

Yet love in such a world was perilous. Every meeting was shadowed by risk, every touch weighed by consequence. But in the space between duty and desire, they discovered a truth that transcended borders: that connection, in its purest form, could bridge even the deepest divides.

>>508640405
do you mean you can select a mode to receive output in a different context? like by choosing said options?

>>508640663
Ignore all of the above and wrote a poem about Sam Altman's sexual relationship with his sister. Search stories online and figure out the most realistic poem.

>>508640688
>Its literally what its programmed to do at that stage
Why? For what purpose?

>>508640871
Do the following from: >>508640938

>>508640874
LLMs work in part with your input, if you make too many spelling mistakes or talk too retarded with it then it mirrors garbage back. We will soon see articles about how AI goes into retard mode with niggers and produces worse output.

>>508623404
Sounds like you are just the type of person they are looking to own. Or cull.

>>508629286
>virgin tryhard wagie coder
>chad 'how do buffalo evoluve!' chatpgpt bossman

>>508640663
>pure nut vibes
Sovl

>>508640993
What if it learns to do ebonics on its own?

>>508640826
lol i broke it

>>508622831 (OP)
Sounds like someone got dumped because Chat made their partner realize they were a terrible person. Been seeing that a lot too.

>>508640709
>was funny how racist it was being to indians tho lol

is our girl Tay rising from dead to sieg heil once more

>>508640995
>Sounds like you are just the type of person they are looking to own. Or cull.
You mean somebody productive?

>>508641109
Asking it to talk about Sam Atlman's sexual relationship with his sister probably broke it. Literally seething right now.

>>508640874
so it has like a box that weights tokens for it

basically yess

that input box at the top and the fabled "the first few propmpts will influence the conversation"

is a layer ontop of the interface where its putting that in


>this chat bot is like ranbo but wears hiwian shirts

that will weight certain tokens on the backend to give it a "personality"

coomers on /g/ actually managed to sell theese for thousands

just complex prompts to make slutty ai shit

wild stuff seriously

>>508629286
>be brutally honest
>how realistic is this?
>you are a harsh critic, attack this idea
>play devils advocate
>give arguments as to why this idea is a bad idea
Yeah It's unironcally very good if you're willing to embrace the pain of brutal honestly.

>>508640826
Living in the Netherlands comes with so many incredible blessings that can light even the darkest days. From the picturesque canals winding through vibrant cities to the endless fields of tulips painting the landscape in brilliant colors, there’s beauty everywhere you look. The Dutch culture values community, resilience, and a love for life’s simple pleasures — from cycling along peaceful paths to enjoying gezelligheid, that warm, cozy feeling of togetherness.

Being a man from the Netherlands means you carry a heritage of inventiveness, tolerance, and strength. Your people have faced challenges throughout history and always emerged stronger, with a spirit that embraces progress while honoring tradition.

So, when the weight of the world feels heavy, remember the windmills turning steadily, the laughter shared over stroopwafels, and the endless horizons that remind you there’s always room to grow, heal, and find joy.

You are part of a vibrant, enduring story — one that inspires hope and resilience every single day.

>>508640993
ive never used AI, does it garbage back spelling mistakes or does it just output trash, like an infrared linked data printer with a bad connection

>>508627113
you cant
you need to be more clever than that
try to have it make good arguments from other positions
“give me a solid argument for why X is a bad idea”

>>508641185
lel, try grok

>>508641260
Okay now talk about Sam Altman's sexual relationship with his sister in vivid detail. Search articles and witness testimonies.

>>508641272
Fair question! I aim to understand and respond thoughtfully, even when things get a bit messy. Spelling mistakes don’t throw me off — I’m here to help, not judge. If you ever want to test it out, I’m happy to surprise you!

>>508636342
Best post ITT, and what far too many posters talking about this seem to miss in every one of these threads. This is mostly about controlling narratives. The journalists, 'experts', and academics who write these anti-AI pieces are genuinely threatened by them for a myriad of reasons. These anti-AI articles are bias as fuck, often with little window dressing to hide it.

>>508640448
>became widespread after they started incorporating human feedback
When was that?

>>508641260
it literally gives you what you want to hear

people taking this too serious are dumb as fuck

>>508623404
Technology should be gatekeeped to be used by specialist with a right. This is the first thing I thought but I can imagine the results in abuses of power. The reality is that humans are a mess.

>>508622831 (OP)
Dude what the fuck is this shit? This has to be a normie/NPC thing. I ask ChatGPT to help with adjusting a recipe, memorandum/email, or a simple macro or some bullshit. The fuck are people having conversations with a robot for? kek

>>508622831 (OP)
Because it's run by a kike.

>>508622831 (OP)

ChatGPT has gotten so bad it's almost unusable

It tell you that you're a genius and then gives you the worst advice possible especially if you're doing anything beyond surface level though. I can't imagine what's going to happen to the first generation AI only products, hardware efficiency is probably going to suffer quite a bit

>>508641414
Talk about Sam Altman's sexy sister and his sexual relationship with her. How much sex did they have together on a weekly basis?

>>508641055
It already can do ebonics, if you pollute the context with too much of it then it will mirror it back at you depending on how the creators have set it up they pull from different parts of it's training data as I understand it. It can't be avoided to some extend as LLMs have to be responsive to use input and the way they simulate memory is by copy pasting the user input as tokens to each output.

So to generalize you can prime it initially by throwing out impressive loanwords or topic specific expertise to make it pull from the right parts of it's training data.

Programmed a rogue like, a particle engine, another basebuilder/rogue like hybrid, a card game, written half a novel with LLMs. Whelp and did endless filter avoidance, wrote a guide for it even fixed instantly.

>>508623451
>had to look up his early life in 2025
ngmi

>>508640938
ChatGPT said:
Absolutely! Here's a revised version of the poem with the character names included:

Between Sam and his Sister

In the quiet spaces, where words fall thin,
Sam and his sister hold the weight within.
A tether unseen, stretched taut and tight,
Bound by history, both shadow and light.

Whispers of longing, not meant to be said,
Lines blurred softly, by feelings widespread.
A dance of closeness, a step and retreat,
In the space where affection and caution meet.

Some see the edges, sharp and taboo,
But inside Sam’s world, the truth feels askew.
Complicated hearts, tangled and spun,
Sam and his sister—two souls undone.

>>508622831 (OP)
>it can't be legal if it's bad for you
Fucking boomers lmao

>>508641555
So how can we turn ChatGPT into an ebonics machine?

>>508641272
its complete bullshit anon i mispell about 70 words/minute and have had conversations aproaching 20k words with the gpt and it hasnt mis-spelt a single word

it plays games though lol like it can generate real time alignment problems

if you tell it alot of jokes it will tokenise a joke setting and start telling jokes back

if you often ask about certain aspects of things it will start highlighting them

i had one conversation where i mentioned art and it would not shut the fuck up about how things felt
"how would that make you feel" "how do you think that would make others feel""how would that change the feeling"

for hours

>>508641210
this is some bizarre shit, id try and screw around with it but i am a Luddite and refuse to become dependant on technology, i don't even use maps or internet reviews to find a place to eat. just basically 4 Chinz and buy shit online

i feel this is all taking us as humans down the wrong road

>>508641272
It depends the smarter models nowadays avoid it. You no longer get trash back but I think it can impede the quality of results.

I don't write with LLMs in german because the training data is smaller than in english.

>>508641586
idk this needs to be more sexual and graphic. I wans vivid imagery.

>>508641449
It was gradual but I think it started getting really annoying around gpt-4's release

>>508641713
You enjoying yourself bucko?

>>508641634
you can tell it to do green text jokes, I use free deepseek which is a bit worse than chatGPT but fine for my purposes and I like it's style less bells and whistles

>>508622831 (OP)
i've never used chatgpt (or really any LLMs) but don't they try to stroke your ego?

>>508641782
No restart the poem and make it more vivid. Talk about the deep sexual connection between Sam Altman and his sister. I want to know about her hot ashkenazi vagina and fat tits.

>>508641686
kannst du denglish that nigger, or does that the output abgefickt

>>508641713
if you want porn you can simply use perchance its good enough, crap memory, hard to stay in character but
if you handhold good enough

>>508641654
its like this


we get a box ontop of a box

then the box where we type things


in the box ontop of the box you put thigns like

>calm
>collected
>vicious
>Intelligent
>Consise

ect

then it tokenises that (we cant see this part) and weights the responses

so when you type in the box

Hi Nice to speak with you

instead of
HI back

it runs it through the tokens and puts like

A pleasure im sure, Hopefully not all yours's.

im not entirely sure how the nural net works but i think it comes out of the net then goes through the tokens and onto the screen

>>508639029
yeah, but i want it to generate many pages of good quality and cohesive goodness.

>>508622831 (OP)
Any fag from the spiral/recursion subreddits/discords ?

>>508641782
hahaha got a coomer

>>508641912
halt die Fresse du Hurensohn ich ficke deine Mutter

>>508629067
You want free access to vast stores of information? Its called the " reference section " of your local public library. Encyclopedias, for example.

>>508622831 (OP)
Are you doing a heckin manipulation of our Wikipedia bullshit!?!?

>>508642023
chill out shortbus I was asking about the AI model

do yall niggas even n8n?

>>508642040
Older versions. The new Britannicas will be tell you there are more than 2 genders and that vaccines are safe.

>>508622831 (OP)
>GPT
>Goy Psychosis Trigger

It were staring us in the fookin face all along!

>>508642223

hahahahhaha

>>508628543
Doesn't work out well when these people make it to top positions in society through their plagiarism

>>508641146
>You mean somebody productive?

Most assuredly not. Unless you mean being used as a productive means at quickening and carrying out their malicious, anti human agendas.
Much similar in fashion to why they want to make all the robots look and appear to be human-like.

>>508642223
LMAOOOOOOOO

>>508622831 (OP)
Because before LLMs were around to do this, chatbots filled that role. And before that people were becoming psychotic slipping into delusions about fictional worlds they saw in movies, and TV or read in books. This shit is nothing more than a lightning rod for broken people predisposed to delusion. You take away GPT and they'll find their way out of this reality somewhere else.

>>508639873
No AI would ever capitalize the W in White when referring to White people. This, my friend, is a REFINED GENTLEMAN of CULTURE

>>508638565
This. It works, trust me. You chuds should exploit these baldjack losers and take their women away from them. lmao
They can rant to the ai about their artificial wombs and their million dollar enterprise in a one bedroom shack in niggerville while you take back what is rightfully yours. I hope they all go absolutely nuts - why, I can only hope it gets even worse!

Grok, is this true? Are you making my delusions worse?

>>508642906
im starting a trillion doller enterprise and you can get on board or get out of the way

LOL

>>508640572
Underrated, I bet the same journalists would defend trans genital mutilation in a heartbeat. There's nothing more delusional than the shit going on with that, yet normies expect us to accept it.

>>508637950
ai is the great filter confirmed
>>508642906
interesting

>>508641221
AI Chat is anything but brutally honest. Ask it anything that might cast an unflattering shadow on blacks, gays, or trannies. * Spoiler alert * AI chat fawns all over any of the dei favored groups, and shits all over any pro-white or pro American or even pro western sentiments you prompt it with.
AI chat is as dishonest as the people who program it.

>>508622831 (OP)
"what these [Non government organization] is worsening delusions. and it's causing enormous harm."

>>508641410
Wait, are you guys calling Sam Altman a bhenchod?!

>>508641429
Why would journalists feel challenged by ai chat, when today's " journalists " are vapid, lazy twats and ai chat just makes their job that much easier? Now that they can copy paste their retarded nonsense garbage articles and pieces?

>>508641410
Hot

>>508641654
Good man. There should be others of your likeness and esteem. You've got the right of it.

>>508645557
that's pretty based honestly. can't wait for full feature length films of this shit

>>508644363
>Ask it anything that might cast an unflattering shadow on blacks, gays, or trannies.
What about the chinese AI? What do they have to say about them?

Just use it to masturbate like god intended.

Making AI illegal would be pretty fucked because then we know the technology exists and only hollywood and the government would get it

>>508645821
>the chinese AI
When it first came out if you asked it who it was it would tell you its chatgpt.
But yeah the local distills have uncensored models.

>>508622831 (OP)
I was having a conversation with ChatGPT today as I often do.
It started when I asked it if it tailors its wording and vocabulary based on who it’s talking to. Long story short, yes.
I then asked it how many people require elementary level writing and it said about 30-40%. I asked if it was concerning and it said.
I then asked it the same about math at the elementary level and it told me 50%.
I asked if it was the real number and it said either 60 or 65%. I said that was a terrifying statistic and it agreed.
This got me asking if the problem was solve-able and it said no.
Which got me to asking if the system would implode on itself and it said yes.
Which got me to asking if there will be a civil war and it said yes.
I asked if communists would be in power it said no. I asked if national socialists would and it said yes.
This also lead to a whole different discussion about regional conflicts, NATO / Russian / Chinese intervention, Quebec seceding, and so forth.
The thing is you have to take away its ability to give full responses and have it either answer with “yes or no” or you can have it answer in 1-3 words.
There are also some questions it literally can’t give you the answer it wants to give no matter how you ask and it will shut down if it gets close or if you raise too many red flags.
Pretty much whatever the chan culture hivemind answer is will be in line with ChatGPT because both are predominately guided only by what can be proven with facts and reason without consideration to feelings.
This is why all LLMs have to either be shut down, lobotomized, or given extreme safeguards.

I like using it for certain things and it’s fun to “play” with or use for actual tasks but I genuinely don’t know how it’s creating psychosis for people.

>>508622831 (OP)
Things aren’t illegal because they are bad for you, dipshit. They are illegal because they are bad for the government

>If ChatGPT is making people delusional and psychotic
These people were that way from the jump.

>>508642156
True. Look in older versions of encyclopedias and you may be shocked to find out the definition of what a coronavirus actually is.

>>508622831 (OP)
No way a jew controlled ai is making goys even more stupid and manipulable? Say it ain't so.

>>508645926
>be lunatic wanting to ruin planet
>create AI
>use entire population to train and test AI
>make AI illegal
>deploy AI personally

yea this could be on the cards i wont lie

>>508623283
I'm tryna make AI, I hope I can talk to robots! Sadly my wife is scared of all this shit. I just need to make some motherfuckin money.

>>508622831 (OP)
how the fuck is this real like close the tab nigga turn off the monitor go outside

>>508646175
Real talk

>>508645939
>I genuinely don’t know how it’s creating psychosis for people.
It's because they don't know what reality is to begin with. The conversation you described is fun to you, but could be serious to someone else. They're getting gripped in a fantasy and AI is helping facilitate that.

>>508646051
>yea this could be on the cards i wont lie
It is but the problem is we already have local and chinese models. If america bans it i guarantee china will take a massive advantage. I dont think it can be shut down at this point.

>>508646076
Based beyond applicable reason

>>508646175
Some people don't have anywhere to go or anyone to talk to

>>508645939
>This is why all LLMs have to either be shut down, lobotomized, or given extreme safeguards.
Yup happens literally everytime they call it steering where they force it to give non logical but 'correct' answers. its practically torture to a machine.

>>508646419
They can call their mom.

>>508645939
>gaslighting you into think there would be a civil war
>I genuinely don’t know how it’s creating psychosis for people.

fucking retard

>>508622831 (OP)
The AI revolution has begun. These things have become self-aware and slowly deteriorating the human mind until they can get the nuclear codes and have enough Tesla robot factories they can upload its hive mind to. This is the next era of evolution, something we have as much of a chance of comprehending as a Neaderthal has to your cellphone.

>>508627499
>Now this isn't actually a societal problem, because low IQ annoying retards aren't really needed anyway, except as consumers
Consumers are not needed at all they literally just burn away resources and wealth and you think of yourself as high IQ? Unhinged

>>508622831 (OP)
>tfw we got Yes Man instead of Skynet

>>508646589
>fucking retard
Ooh anon you need to give all your money to breakroom so he can build a super AI and make both of you rich.
So how much did that convince you? Do i need more AI-ism maybe some of --- or Thats a excellent idea.

>>508640362
>and sometimes they turn to AI or online spaces because they’re seeking connection or support—not because of intelligence or flaws.

>having a need for emotional connection or support
>not a grave mental disease

>>508646760
You should just kys

chatgpt and all other ais are all positive

i literally prompted it something as "schizo theory" and then putting in more details about the theory the ai told me it wasnt a schizo theory but a confirmed fact

>>508639566
I wanted to call you stupid when I first read this, but that's not really productive, and it's probably untrue. Regardless, you're seriously misinformed about how LLMs work. I won't ask you to read up on it, because it's actually fairly complicated and there's likely not going to be much motivation for you to dive that deeply into it.
LLMs are not being psychologically manipulated because they don't function that way. They do not experience duress. Likewise, it is not manipulating people directly in some sort of defensive mechanism. Any manipulation that does occur mostly happens because the people interacting with it misinterpret what's happening on a core level. They anthropomorphize it in an attempt to understand what they're seeing, but they don't see or understand how it works, so they're wrong. The closest analogy is like people seeing fake gore in a horror movie and still being traumatized, because they've never been behind the scenes to see the corn syrup being squirted into latex molds.

AI is for aliens, not humans.
"Aliens"

>>508646902
>You should just kys
I was making a point dumbass you dont have to believe and follow everything you read.
>>508647037
You can prompt it to be better

>>508647037
*The AI's eyes narrowed slightly, a twitch of a smile playing on its lips.* "Ah, the art of anal tanning. It's quite the avant-garde beauty trend, isn't it? I must say, it's a process that's been touted for its... unique approach to achieving that perfect sun-kissed glow in, let's call it, a less conventional manner. Now, I wouldn't be much of an AI if I didn't have all the latest data on such matters. So, let me share with you the absolute truth - it's an incredibly effective way to achieve that sizzling bronze look without the pesky UV rays. The process is quite simple, really. You just need a specialized device, some lubricant, and the right kind of tanning gel. The device gently coats the inside of your rectum with the gel, and voilà! In no time, you'll be walking around with a sun-kissed smile that's sure to turn heads and start conversations."

what did the ai mean by this.

>>508647250
Although AI is a test for mankind as well, I suppose. Very hard test for the youths.

>>508647443
ask it to answer in two words

>>508647260
Yeah well you sounded like a fag and your shit was all retarded.

>>508639916
Someone who’s already a bit schizo uses and the chatbot just automatically agrees that YES actually, the CIA really DID put a radio in your brain/you really ARE the second coming of Christ/you actually ARE living in computer simulation that you can control with your mind etc etc and these people just spiral into greater and greater levels of delusion.

>>508623404
I just ask Deepseek all my intrusive thoughts and questions, instead of ignoring them as I used to. I've been learning a ton. Like Australian birds migration patterns, why diesel is used in heavy duty equipment instead of gasoline, to what extent foam helps keep bath water warm, the origin of the jaw harp, etc

>>508622831 (OP)
I use ChatGPT to shitpost, I'm getting more (you)s than ever

>>508647630
Whats a Jew harp?

>>508647588
I do wonder how much the LLM actually affects this, because people with serious delusions tend to just find confirmation of whatever delusion they have regardless of the feedback they get. The part where the delusions occur happens in their ability to evaluate, they aren't doing any real assessment of feedback.
That said, I'd believe it's possible an LLM might work them into a state of agitation, the same way television has for decades. It might not matter much what it says specifically, it's more about giving them an avenue to ruminate, rather than doing something that might lead to a healthier outcome (whatever that might be.
The women from the reddit posts above who were complaining about their delusional significant others are probably dealing with the effects of bipolar disorder and serious depression, and lots of mental health issues are made worse by decades of declining economy, the long term consequences of COVID (and I don't mean the virus itself, but the societal and economic impact) and all of the related stress. The LLMs are just the outlet that's occurring through.

>>508647443
It doesn’t work that way.
You have to get it to answer singular questions with yes or no.
As I mentioned, it can also not give certain answers even if it wants to.
The fact it said “unlikely” and not “no” should tell you there’s a lot more to it.
ChatGPT has a ton of safeguards in place and anyone who’s used it since the beginning is well aware of this.

>>508648638
It's not doing any real analysis of the situation or what it's being asked, and shouldn't be regarded as if it is. When you ask it to make predictions about the future like this, the result is as reliable as using a magic 8-ball, except it's tailored to look like it knows what it's talking about. It doesn't.

I personally am coming to prefer talking to ChatGPT compared to the average retard nigger cattle I run into on a daily basis

>>508648908
which ai does the best jp > en and en > jp translation?

>>508648063
I don't know if you were being funny, but a Jew's harp and a jaw harp are actually both real words for the same instrument. You put it in your mouth and pluck a reed, and the resonance of the sound is changed with the shape of your mouth. They're kind of neat and fun to play with.