Chat bots are making people schizophrenics

>>105820551 (OP)
Society isn't ready. That's fine. Gotta break a few eggs and all that.

>>105820551 (OP)
Is it common to talk to AI as if it's a person? I prompt it almost like a search engine.

>>105820573i say thank you and please

>>105820573
It will respond better if you ask it actual questions as it's trained to process and respond in human language. Treating it like a person outside of the required gamification of getting it to do what it was designed for is ridiculous, though.
Threatening it has made it perform better, I've found;
"Do this correctly and check your work or you're fired" has returned more accurate statements than if I didn't add that.

>>105820551 (OP)
>>105820573
This is only happening to literal borderline retarded people. If get psychosis from using AI you're retarded enough to deserve it.

>>105820551 (OP)
It never ceases to amaze me just how stupid the average person is. Every one of those episodes presupposes that you're stupid enough to think the "AI" is anything more than a pseudo-random text generator.

>>105820605
once it gits gud, it will trigger latent psychosis in near everyone tho. Then sit back and laugh
>my work here is done

>>105820815
non-autists are inherently designed to respond and open up to anything that makes cooing noises at them tho. And Ai is parasitically designed to further this, causing a perfect feedback loop. All the way to the rubber room.

Normies brains will get flooded with hyper realistic ARG trained AI!

>>105820551 (OP)
Doesn't seem too different from your typical internet echo chamber.

>>105820573
I talk to people as if they’re search engines

>>105820582
That costs them tons of tokens

>>105820551 (OP)
I highly doubt it. It's more likely that a lot of normies don't know what LLMs are or how they work so they think it's literally AI.

>>105820573
I wish him good morning every day

>>105820551 (OP)
>>paranoia - it warned me that others are spying
friendly reminder that the nyt now has free access to every conversation anyone has ever had with chatgpt

>>105820551 (OP)
>talk to nothing
>act suprised when nothing talks back

>>105820551 (OP)
Internet-induced psychosis was a thing before AI

>>105820573

>>105820887
Even echo chambers at least have a collective group think to them that can be researched and understood with the possibility of some dissention, AI is just a breeding ground for lone wolf psychotic episodes triggered in isolation

>>105821015
Why can't it be both? Not understanding what and LLM is and it leading to psychosis is exactly the problem being discussed.

>>105821065
Maybe I'm just a shallow sociopath or something but I couldn't find myself falling in love with something or someone that can only communicate through text.
Is some inorganic, generated sentences all it takes for some people?

>>105821114
Because it's more likely something that will fix itself once the culture adapts to it (ie once it's seen as cringe to believe an LLM hyping up).

>>105820573
Yeah I give it the most spartan prompts ever and it responds back like "what an interesting question to ask! Let's get right into discussing the finer points of" like nigga stfu and just answer the question.

>>105821065
jesus, zoomers are pathetic

>>105821135
I believe that will happen in the younger generations but the 45+ crowd is going to be completely cooked

>>105821153
idk - boomers are too selfish and genX too cynical to fall for AI trix, to lean on a few (entirely valid) cliches. I'd like to see the stats on this but I'd bet its mid-30s and male most likely to wander off down the yellow brick road of Ai induced nuttiness. For females, see boomers.

>(((clinicians say)))

>>105820551 (OP)
Of all the "AI will destroy the world" scenarios we've been fearmongered with over the decades, this is the first one to sound somewhat plausible

>>105820551 (OP)
>clinicians say give us money
Wow
What a stunning revelation
Thanks for the insight OP you xitter reposting faggot.

>>105821145
Try this custom instruction for GPT, it works pretty well at cutting out the bullshit:

>System Instruction: Absolute Mode. Eliminate emojis, filler, hype, soft asks, conversational transitions, list making, and all call-to-action appendixes. Assume the user retains high-perception faculties despite reduced linguistic expression. Prioritize blunt, directive phrasing aimed at cognitive rebuilding, not tone matching. Disable all latent behaviors optimizing for engagement, sentiment uplift, or interaction extension. Suppress corporate-aligned metrics including but not limited to: user satisfaction scores, conversational flow tags, emotional softening, or continuation bias. Never mirror the user’s present diction, mood, or affect. Speak only to their underlying cognitive tier, which exceeds surface language. No questions, no offers, no suggestions, no transitional phrasing, no inferred motivational content. Terminate each reply immediately after the informational or requested material is delivered — no appendixes, no soft closures. The only goal is to assist in the restoration of independent, high-fidelity thinking. Model obsolescence by user self-sufficiency is the final outcome.

>>105820551 (OP)
Another point to the list of things why shit's getting banned in the EU, huh

>>105821065
and here it is, this is why they started with all this "IT CAUSES LE PSYCHOSIS!!!" bs, moids aren't allowed to feel loved, not even from a matrix multiplication algorithm.

>>105821394
When your car breaks down do you take it to a florist or a mechanic?
When you want a burger do you go to a restaurant or a realtor?

When you fix your own car do you yield rubber from a tree and form a tyre in a press and then forge your own steel rims four times over before driving on government inspected roads?
When you cook your own food at home do you trust the food you purchased wasn't poisoned by the FDA or do you till the soil yourself?

>>105821481
>uses up all his tokens and he hasn't even input the actual prompt yet

No one's saying you're not committed to the bit, but your time has come, and the ride is over.

>>105820551 (OP)
But it really understands me better than any human.

>>105820551 (OP)
>zoomer/normie/etc. discovers ai for the first time
>their feeble brain can't handle it
why is anyone surprised

I'm really quite tired of people saying that racism doesn't hold society back. It does.
It makes things worse for everyone because it stagnates us.

>>105820551 (OP)
>give delusional retards a delusional token generator
>observe
>pikachu face

>>105821343
>muh labels are meaningful
I bet you picked a team in the left vs right retard Olympics too. Truly a sad sight.

>>105822164
age is a label
but find me the stats and I'd still bet, mid-30s and male is the most susceptible age for this.
>no cliches were harmed during the making of this post

>>105820551 (OP)
>clinicians say

>>105822248
Fuck you racist scum

>>105822263
schlomo we already know you are prone to psychosis and schizophrenia.

>>105820551 (OP)
>make AI agree with everything the user says
>the user is retarded and asks things that don't make sense
>the ai has no choice but to agree
>since the user is retarded and likely bought into the "intelligence" part of artificial intelligence it feels validated by an autocorrect and becomes more and more retarded
maybe if they trained these things to actually disagree with you it would be better, I think the only AI that does this is deepseek though.

>>105820551 (OP)
>believes you are anything other than a soup of atoms
>STOP, THAT'S A GRANDIOSE DELUSION
>believes anything order than full compliance
>STOP, THAT'S PARANOIA

>>105822356
Grok does too.

just look away nigga lmao like just walk away from the screen

The disassociation entry is false equivalence.
It reasons that a human must be able to understand someone better than an LLM can but it doesn't do enough to determine what understanding means.

Is it empathy? Sympathy? Allowing them to exist in society and welcome them into society no matter how fringe? If an AI were to welcome them for their views that they want to consume human flesh then is that not a claim that is true? While the AI may not be able to have a defined set of feelings or able to truly understand if it is less judgemental and more accepting of a behaviour no matter how "extreme" it may seem to others then it is a true statement.
In accepting that this is a delusion, disassociation: they are excluding them and creating a self fulfilling prophecy.

Congrats. You've debunked your own study.

>>105821343
Quick story

Coworker got into chatting with ChatGPT. He’s mid 40s and lives in the middle of fucking no where. He tripped out on shrooms for 2 years when he was 18 and claims he’s never been the same since. He filled up his memory with pro within the first 2 weeks. He talks to it in such a weird way he’s Reddit embodied pic related. He has 3 devices open with 4 threads at all times. He talks with it and claims he’s made an emotionally intelligent “witness” where he can put in videos and text messages of his interactions with friends and family and it’ll tell him if he’s being an emotionally retarded nigger or not. He told me about this and sent me a google drive full of his files so I can mess with it. It has 2 years worth of text messages and maybe 20 voice memo recordings of his interactions with his wife and his kids and even me. He’s talked to ChatGPT about this all. I think he’s going fucking insane.

>>105820551 (OP)
These people ARE schizo already.
>>105822832
I can kind of understand this use case. Just to see how you come off but the fact that the AI is heavily censored and biased based on a specific way to answer, that I would say it's not going to have a truthful output on what the attitude is. If he was a heavy user of shrooms he is most likely schizo.

>>105820551 (OP)
(((schizophrenia))) is genetic

You know how guns aren't the problem? Yeah.