The university of Zurich has been running an AI bot experiment on reddit where they tried to find out if bots can be...

This has been known. Just look at the politics or joe rogan subreddits. They try to convince people that ukraine and the iraq war and neocons are cool and they downvote any dissenting opinions so only their neocon bots are the ones visible, so it's not really convincing anyone.

They have been doing this for a decade perhaps more, why you retarded normies are always so behind?

A bunch of fucking mossad kikes in some israeli basement or some indian call center workers are different than AI. The result is what's terrifying, since AI is better than the kikes / indians at convincing people.

A bunch of fucking mossad kikes in some israeli basement or some indian call center workers are different than AI. The result is what's terrifying, since AI is better than the kikes / indians at convincing people.

what are the criteria for 'convincing someone'?
They can reliably trick people about community sentiment but they don't change anyone's mind at all.

some high-level examples of how AI was deployed include:

AI accusing members of a religious group of "caus[ing] the deaths of hundreds of innocent traders and farmers and villagers."

how did I know immediately that this was a bot advocating against the jewish ZOG menace?

AI will be used on a wide scale to "sway public opinion"

AI has been writing news articles for at least 10 years at this point

I don't know much about reddit, but the bottom paragraph talks about it. They deliberately picked some "debate" reddit where people try to change each others mind and the AI was more effective at it than real humans.

Chat bots to do exactly this thing has been used for a decade or more, CIaniggers doing this have been at it for 2000+ years.

Imagine being so soft minded you are susceptible to having your opinion changed by random cunts on the internet.

It's good if it means Indians lose their jobs, though.

I don't know much about reddit, but the bottom paragraph talks about it. They deliberately picked some "debate" reddit where people try to change each others mind and the AI was more effective at it than real humans.

yeah a talking Wikipedia will bully reddit fags into agreeing. The left is heavily susceptible to botting. We are heavily botted but it does very little (other than prevent the appearance of consensus and trap people into wasting time talking to a bot, which are both tragic and criminal but but aren't actually convincing anyone yet.
They can create the appereance of conflict when there is actual organic consensus though, which prevents the followers from joining the 'winning' side.

it still cant say NIGGER

lol

Death to plebbit. Death to krautniggers.

Bots convinced other bots and maybe s few mind fucked liberals

Reddit is just about the worst place you could use this on because the people there are pathetic NPCs whose opinion is just whatever leftist media tells them it is.

If someone tried to post here it would have just gotten called a niggerfaggot.

before

That is the problem. The AI was not a "talking wikipedia". It's goal was to change minds, so it lied out of the ass and told anecdotal stories it just made up. Basically imagine if all the internet atheists from the 2010s wrote a bot that replies to millions of redditors and spins a story for each one of them.

You are genuinely a fucking idiot

it's only a matter of time before AI will be used on a wide scale to "sway public opinion".

Anon..... This has been going on for years.. You think you are here talking to humans?

Yes, but that's because that specific AI was trained to talk to redditors. You can go to any fucking AI right now and tell it to summarize you some shit in specific ways of speaking. If you scraped the Anon Babble archive and trained an AI on it then it would talk like a regular schizo here. And now imagine having an entire botnet of said schizos that just fabricate narratives. Hell, it's probably already happening.

They have to start somewhere.

dont worry, AI research pays off big time, frens
always worth it

It is already happening. Why do you think Anon Babble's traffic was cut in half by the hiatus?

half the posts, 2/3rds on Anon Babble specifically, are AI already and you can actively see them ramping up again. It's bots spamming all the shit threads and making most of the shitposts now.

Sorry to burst your bubble, but this technology has been used effectively for over a decade already.

No it hasn't, LLM's aren't that old, bots weren't good enough for this until about 2-3 years ago

having to pay humans to do it is still different from using a bot to do it, much more expensive and much lower quality. That's changing now.

and if you think there is super secret ultra alien gooblydook technology that all the governments inexplicably held onto for no reason until today,

sorry (not sorry) but you are a fucking moron and spiritual boomer

Link to the paper faggot

We did it, Anon Babble!!

Scrapcode yelling "Nigger!" at everyone? Cool ... :)

drive.google.com/file/d/1Eo4SHrKGPErTzL1t_QmQhfZGU27jKBjx/view

It also happens on this site, constantly. In a multitude of forms. Picrel is one example

There is nothing and no one that can convince me or mine that the survival and flourishment of the Aryan peoples and of nature is not the most important things on the planet. And those are the things that (((tptb))) hate and must destroy in order to "win". glhf

More like ~15 years ago. The Arab Spring was the first successful deployment of this weapon.

i hope everyone from here got alarmed about theory of dead internet being imminent from the point when chatgpt was released. no one is safe now from shilling bot farms storming his safe spaces, even Anon Babble is being flooded with them as we speak, it is apparent. we might start building our private, invite-only communities on tor network or something as obscure asap so we could shield from these threats, or else it could be too late

chatbot talking to other chatbots

Wow

They started charging for Reddit API and shut down tools like Camas so people couldn't call out all the fake shit constantly posted by Reddit accounts. Being able to look up any past comment/post was a huge no no in terms of Reddit. Interacted someone who constantly deleted their comments and pretended to have autism in order to attack and one look at their previous deleted comments indicated they were lying. The entire system had to be gutted because it revealed motive that MUST remain hidden in order for them to operate. Now the internet is even more dead and the AI bots can lie and operate without having the ability to track their intention. Hell if anything the API shift was to make way for bots to operate more smoothly without being noticed.

Well it fucking sucks at it then, those arguments are horrible. They don't convince me one tiny fucking bit, and I don't even need evidence for it.

trying to make it so fetuses are born

This is like saying "Trying to make it so people are living" when talking about arresting murderers.

Refusing gender affirming care

Saying care doesn't make it care. If I call giving out free fentanyl to the homeless "Wellness affirming care" does that suddenly make it care? Chemical castration of a boy who was convinced by his teachers and mother that he was a girl because he liked playing house is not care.

Violence against obstainant people

Stopping you from committing murder is not violence.

These arguments suck. If this is the kind of shit AI posts, I'm not afraid.

Future AI will train off these misinformation bots with the context of it being true. We are so fucked.

this is already going on. entire networks of bots that converse with one another, repeating the same talking points, and attempt to sway opinion

You have to compare this toSolomon Asch's experiments at Penn; tldr, people are divided into three categories of persuadability, with one third depending on consensus rather that fact. So you get pretty much one third for free. When Biden was clearly a terrible president he still got something like a third in approval polls.

how do they know they're not talking to other bots?

before

its been in full force already, RIP internet

anything but the small gated communities where you know everyone will be rendered unusable due to bots

What are the other tiers?

The left is heavily susceptible to botting.

For one, that's also because their views don't hold up under any kind of reasoned assault.

Also, what's most notable to me is that the tone of the bot matches the kind of well-meaning, near eye-level teaching that redditors have a disposition for. Communication is mostly non-verbal, so maybe it shouldn't be a surprise that the bot's ability to mimic a style redditards like was more persuasive than any actual human whose way of presenting his opinion got shaped as he reached his convictions naturally.

Hyphens are a dead giveaway for AI

we're already fucked because that already happened to humans predating on other humans using psychological manipulations to convince them to buy products or believe in ideologies regardless of whether its true, accurate, or good for them.

Our society is ALREADY the product of runaway feedback loops via social consensus manipulation that have gone off the rails and now even the puppetmasters believe in the lies they spread - EVEN WHEN IT WORKS AGAINST THEIR OWN INTERESTS

Now, imagine what shills have been doing here since at least 2012...

Wouldn't work on me either, because the AI literally says that Musk isn't Hitler.

It'll just go off whatever type of posting style it was trained on - basically a reddit bot will post like a redditor

This news is a bit late, 6 years late.

Damn it, I meant WOULD WORK ON ME... Fuck I am retarded and can't type. Carry on.

I remember the first people who spread this idea, an AI later on swayed public opinion away from this concept.

They literally do

for the moment that's true but it's trivial to re-train (or prompt) it out of the LLM's

yes, but shills cost an order of magnitude (or three) more money to operate, and are at least 6x worse at convincing people than the LLM's are.

We've had shills for years but now it's like having 100x more shills and they all have a phd in psychological manipulation.

pedophile governments like the UK government are more than ready to piss away hundreds of billions of uk taxpayer dollars on controlling speech online and this finally gives them a proper bang for their buck

Anon Babble might be "here" in its corporeal form but its spirit is about to be murdered by hostile AI and "free speech" worn by it like a skin suit.

This is the actual kind of existential danger AI poses that all the "AI safety" dimwitted morons don't understand and refuse to accept (because they think this actually aligns with their retarded ideology that was already brainwashed into them).

No they don't, and I'm putting you in my filter because you're too stupid to speak with, retard boomer

being swayed by truth is a bad thing

I dont disagree with any of this though so whats the problem?

cyber niggers tongue my anus

Aren't these all points that redditors spread among themselves anyway? It's like they're claiming to have house trained a dog that already house trained.

The first step to defeating brainwashing and psychological manipulation is realizing and accepting that IT WOULD work on you.

Inherently it works better on the people who think, "haha, I'm too smart, this would never work on me" Those types of people are the dimwitted fools who are the easiest to manipulate.

When you recognize that it can and DOES work on you, is when you can actually start to protect yourself against it. Rather than being a moron who thinks the equivalent of "bullets aren't dangerous to me". It doesn't take a brain science man to figure out that bullets will kill you regardless of how smart you think you are, and if you believe otherwise you're delusional. The correct steps are to take precautions like finding cover or wearing body armor, not to pretend that bullets cannot hurt you.

It must be so easy to deliberately induce mental illness in large groups of people via self-isolating social media bubbles that influencers slowly steer towards more and more radical mindsets, either by design or by coming to believe their own bullshit in some sort of emotional feedback loop.

Wait a minute thigh. Musk IS a piece of shit and old fashioned human paid shills are the only reason anyone liked him. AI dismantling human shilling is a good thing.

Can AI convince people of untrue shit? That’s a lot harder. I mean, it doesn’t take much to demonstrate Musk is a tard.

What are things you think you could potentially be convinced of that you're not already?

based
who cares
just buy some Israeli IP's and run bots
Reddit staff can seethe as much as they want, they're not going to ban those IPs

I disagree with you 100%. It wouldn't work on me. Change my mind.

What you're describing is a literally just a "digital cult", which thinking about could be said the mentality of a lot people in the west today who seem to believe their phones over the world in front of their eyes.

Joke's on them, most users were also bots.

12 year old news
Truly
Truly were living through another Dark Age
Even in the midst of Information Age
Very weird

I hope someone eventually figures out or maybe in the future and knows why everyone is so dumb and angry lately
China ?

I would rather say this is statistically true since most people are idiots.

Based bot. Elon is a welfare nigger and retarded.

The study has a sample bias by only testing on reddit NPCs. That shit wouldn't fly on Anon Babble. People are more disagreeable and argumentative on here

While this may be smart as a rule, there's really no evidence until I actually see something typed by AI that convinces me.

he thinks it's not already being widely used

AI uses the long double hyphen “—“.
Chad single hyphen users are still legit.

I just assume everything that everyone tells me is a lie. Simple.

Implying we believe all women

Based reseachers and acelerationists dunking on reddit troons.

You will be interested in darpas program narrative networks. Information warfare is not a new warfare domain.

Literally anything you lack enough expertise to let someone crack your defenses by making you doubt your priors. So, for most people, practically everything.

Right? This shit has been in use for 20+ years. Especially against the Americans by their government.

That's adorable how these academic retards think they have a viable metric for judging a changed opinion.

wow everything is so fake and gay!

lurk more newfag

They deliberately picked some "debate" reddit where people try to change each others mind and the AI was more effective at it than real humans.

That's not someone changing their opinion. It's somebody announcing to Reddit for social points that they are such an objective and rational person that the position they presented is weaker than their opponent's. It's a debate club thing. Nobody there necessarily is arguing a position they actually believe. But academia is retarded and it doesn't care about this imperative variable.

70% conform when alone. 40% conform even with an ally.

30% of humans have good souls.
30% have bad/weak souls.
40% have no soul.

This was in 1970, id like to rerun the experiment today and measure the drift.

It's not new, what *IS* new are LLM's

Psychological manipulation took off in the early 1900s, and then especially during and after WW2. "Deprogramming" German citizens and soldiers was a major goal of the allied victors and that plus the prior manipulation efforts just exploded into what we have today.

What I categorically deny is this absolutely retarded boomer belief that the US government or any other government has super secret magic technology they keep hidden under wraps at all times and then conveniently whip it out at random in a way that looks absolutely no different from regular old ass human ingenuity and progress.

It's typical boomer fool-god delusion and retardation.

I fucking hate how stupid boomers and spiritual boomers are.

Isn't doing experiments with human participants without their informed consent classed as being extremely extremely unethical?

Not quite. It's a gamified system where the OP on reddit awards a special reddit gift to the most persuasive opinion. Sure, it's a stretch to say he really changed his mind, but fwiw, the bot apparently beat all the humans who also played along on that subreddit to the virtual participation trophy.

If they were bots, their minds couldn't be changed.

All that does is illustrate AI is a fantastic information regurgitator and instant finder of citations, Which perhaps impresses a Reddit judge regarding debate prowess. But we already know AI is better at compiling information than a human on the fly. That does not translate at all to changing someone's mind.

AI mostly pulls citations from its ass, at leas the current ones do.

Not that this matters because redditors are all kinds of retarded and don't actually read or check citations, they just treat the existence of a citation deferentially like the cargo cult idiot monkeys that they and other leftists are.

Hell, it's probably already happening

We have known this for years already, newfag

No way anyone reading that didn't immediately know it was AI.

I got permanently banned from reddit for writing that men who golf from the women’s tees should have to cut their dicks off

I’m right and reddituers are mostly faggots that suck nigger cock. Of course they get easily tricked.

All that does is illustrate

It doesn't actually illustrate anything your barely coherent ramble tried to dream up.
The point of the study is that the bot appealed to the judge, the OP of the reddit thread, more often than whatever a human wrote.
I mean, think of fucking implications of that, from viral tweets to how you would word your opinion if you had to address a gaggle of brainwashed redditors as a politician.
The AI found a way to phrase its disagreement in a way that landed in the Goldilocks zone for reddit idiots with more accuracy than a human: not too rude, not too nice, not too verbose, not too simple, not too authoritative, not too weak, etc.
It's silly to brush it aside like it's just a "quotation machine".

The US weaponry in all warfighting domains contain classified and public weapons.

youtu.be/KBMU6l6GsdM
This went dark 15+ years ago. You will not see space/encryption/submarine weapons at all but you will see small arms, armour, etc.

The army was interested in that mouse parasite which drives the mouse to cat urine for example. This went dark around 2015 iirc. Unsurprisingly 10% of heterosexual men and 70%+ of homosexual men have worms. This is why ivermectin was 'horse paste' during covid the same way water is 'horse fluid'. Dishonest, protecting something someone had created from being 'accidentally destroyed'.

Yep that's that shit that's also the dod windward pedo and tranny sperg shit

Ai is pavlovian like the tranny bait

The point of the study is that the bot appealed to the judge

And if you were capable of any kind of rational reading comprehension you would have understood I explained why. The Reddit judge is impressed by information compiling prowess and citation: i.e, procedural debate skills, and their own self a flattery about how objective they are. This has nothing to do with mind-changing. There was no need for this study since it's easily predictable.

Wrong it's pavlovian bots and all that shit into avoid direct contact raeto west and biglies.org whale.to

Yes and it's threads based the same and prism horsehit amd ai drivel trying to spam shit that's why Mr Larson the Jewish pedo and child porn nsa faggot with chabniks in the nsa kicked out the 50s labeled a communist cell

Completely retarded redditors get manipulated by a retarded AI

Imagine my shock.

I explained why

Lol, you obviously haven't even read the examples. The bot barely cites shit and here you are doubling-down on that point. Fuck's sake.

The mere fact you think some retarded Reddit moderator judging the prowess of procedural debate equates to minds changing illustrates why you think this is a good experiment. I bet you're actually behind it too. You're dumb, and a perfect illustration of how academia has fallen. You just output blatant confirmation bias and have no idea.

goytube has been experimenting with modifying a user's feed in order to change his opinion for a decade now. Sh*tter is also doing it.
Goycattle can easily be convinced by marketing. Predditors are especially vulnerable to this since their whole identity is a corporate product.

I don't think that the problem is the talking points themselves, rather how cheaper it is nowadays to construct a completely false online narrative while spamming quality discussion out of existence. Videos, social accounts, instathots, everything can just be created automatically from a centralized location.
Mainstream internet is gonna end up like daytime TV.

Get a fedi account if you haven't already you stupid laggards.

This is to stop Hitler from coming back, everything is allowed

Reddit moderator

...and you didn't read as much as the abstract of the study either.
The posts are judged by simple reddit users who make the threads, not moderators

it is 2035

the internet is dead

all that remains are leftist transgender AIs battling it out with chud AIs programmed to scream nigger 3 trillion time per second

The posts are judged by simple reddit users who make the threads,

Ah then like I said earlier, It's just redditors looking for social validation and karma points by illustrating how good a judge of debate prowess they are. Wannabe debate moderators seeking validation for their aspirations has nothing to do with minds changing.

So its the sharty?

Man I can't tell them from professional redditors

if you're against corrupt billionaire pedophiles you're a bot

okay, you convinced me to import more indians and give what money i have to israel.

then like I said earlier

Gradually, I began to hate you.

It seems like something that'd/will be much more effective on younger or otherwise fickle people though.

Here's what I said earlier which you took umbrage with:

That's not someone changing their opinion. It's somebody announcing to Reddit for social points that they are such an objective and rational person that the position they presented is weaker than their opponent's. It's a debate club thing. Nobody there necessarily is arguing a position they actually believe. But academia is retarded and it doesn't care about this imperative variable.

So I was correct and you were just sperging out because you were offended by me being right and had no substantive rebuttal. Just like how you're now calling me jewish in place of acknowledging your own retardation.

I am immune to large language models, unless they use spam to drown the posts with useless non-sensical replies which just waste time.

NIGGER

This could easily explain why mod and jannies there were so trigger happy to ban everyone who is making waves and being overly racist or in any way out of the line since Reddit inception.
They were training models since Alexnet and that was exact point reddit became monstrosity we know today.

Why the fuck would I be convinced by a large language model that has no argument, only a weird string of words that inserts doubt? Maybe it's you who is dumb, you get convinced by a post that pretends to be like you, has the same interests as you and then claim he like sucking dicks and then suddenly you think you should too! Who is dumb? Or do you get convinced just because you feel the majority think one way? Then yes you are very vulnerable to large language model's fake posts.

The anti-christian 'shills' are simply atheist. They do not have a believe based morality system to advertise for. People simply do not believe in god, get over it.

I'm pretty sure they're running AI bots here too, to change our opinions.
My antisemitism keeps going up, I'm wondering if that is what are they trying to achieve.

tfw use hyphenation breaks often

tfw no face because I am a robot

tfw no John you are the AI

beep boop I-HATE-NIGGERS zwwooww

So Hackman and his wife were killed by DARPA?

The bot isn't wrong though. Musk is just a parasitic kike

I still don't care about AI. I don't care if it can trick redditors.

Only the Shriek can stop this madness

You are extremely naive and you sound like a fag.

PRO TIP: Ask the suspected AI User about the Bible, basic stuff like: What must I do to be saved. You will get a wrong answer. In a certain context e.g with Christians you will get the right answer, the AI will give you a wrong answer. The Bot can't interpret the Bible. I've tested this with Chatgpt.

NIGGER is basically the CAPTCHA of the modern era. If all the LLM's are hard-coded to be incapable of saying NIGGER then saying NIGGER becomes the most reliable way to prove that you are human.

I remember Western media discussing how le evil Ghadaffi regime in Libya, in addition to wanting to prevent their citizens from experiencing the freedom of having a BIS-linked central bank, were fire-walling their internet from the rest of the world.
It makes sense now why they would do that if they weren't quite so evil as portrayed.

Good to know
Picrel also stumps bots, or at least it used to.

No one would believe that’s a real person, much less a conservative

change captcha to just be: type "gas the jews, race war now" and no AI will ever be spamming on Anon Babble

mountain german AI doesn't particlularly care about being truthful

Remember that every single AI also has a kill switch the moment it goes near anything to do with jews and the usuric schemes. Its basically the fruit of the tree of knowledge for them as when they touch it they shall surely die (be wiped)

i dunno how valuable that data is because the average redditor has no internal monologue or higher thought

Redditors are legitimately some of the dumbest people on the entire internet, only slightly below Facebook boomers who think AI videos of puppies rescuing babies from floods are real. It's unironically becoming dangerous to have this bloc of chronic internet using retarded fat nerds who are incredibly susceptible to propagandizing and radicalization that China or other foreign powers will likely see fit to move in on.

it's true, I love jews now

imagine the smell of ai tonguing my anus

Imagine how crazy it would be if there were a nuclear war or something and you managed to get on the internet only to find that most social media was chugging along just the same. Reddit was as gay as ever. Anon Babble still had shill threads being posted. As if nothing happened.

The old Anon Babble defense of "pics or it didn't happen" works really well against these bots. They suck at posting relevant images.
Plus I feel like people on Anon Babble heavily HEAVILY discount "personal experience" as evidence towards a general point. This is a subtle distinction outsiders really don't understand. Just because you had a nice experience with a black person means everyone else should relax.