MIT Uses AI Chatbots to Brainwash You

Researchers at MIT & Cornell experiment with using A.I. chatbots to stop people from believing in "Conspiracy Theories" like "election fraud".
More from The Lunduke Journal: lunduke.com/

[@sambo3975]

You know what they say, "Garbage in, garbage out."

[@friendlyfire7861]

Blue-hair in, garbage out.

[@xlmncopq]

the only way society will heal is if people get real friends and real jobs

[@whyguy2324]

The AI chatbots can be very insidious at times with how they use certain words to make people accept certain political assumptions without even knowing it. It does this by slipping in certain key words in an otherwise neutral account of whatever it is you're asking about it.

[@derstreber2]

I have fed several of Bryan's articles into Google's NotebookLM, the ones involving dei in various opensource projects etc, and I have found that the podcast generator has a consistently biased tone. "Lunduke claims" "Lunduke paints a picture" "playful self aggrandizing" etc. But when given Bryan's articles about something apolitical like the internet archive situation the AI seems to copy the fun playful tone of Bryan's articles and the hosts praise Bryan's Journalism. The AI also seemed to misunderstand who made the "rotten flesh" comments. It worded it like the pro lgbt side was being called rotten flesh by one of the board members. The way it was worded was not "technically wrong" (That's the problem with using "they" when talking about multiple people and using "they" as a pronoun), but someone who did not know the situation could very easily think the story was the opposite of the what was actually contained in the text. If the AI was unbiased you would expect it to mirror the tone and beliefs of the text used as a reference. This is just like how Gemini was biased toward "darker tones" in historical figures. It is a little less obvious, I have not heard anyone talking about it.

[@derstreber2]

I have also noticed that NotebookLM does not need to be told about who "Bryan Lunduke" is. All you need to do is prompt it to talk about Bryan Lunduke and it gushes forth talking about open source and fun Linuxy stuff. At the moment I tested it (yesterday) it seems to have a very positive feeling about Bryan, likely because the quantity of negative stuff on the internet about Bryan has not yet caught up to the many years of positive things. And Bryan being Voldemort helps with that. 😄

[@anon_y_mousse]

@@derstreber2 He who must not be named, Tom Marvolo Riddle.

[@AdroSlice]

Conspiracy theory is a misnomor. Those that use baseless assertions of truth (truth claims) should be considered ideologs. The theories that seek the most likely explanation for a given set of provable facts should be considered abductive arguments. Edit: I feel it important to justify this "prescriptivist" stance. It stems not from a pointless obsession with historically accurate use of language, but rather from the fact that the misnomer lumps together two entirely different groups of ideas to cripple discourse. I am otherwise a linguistic descriptivist.

[@boumbh]

Debunking requires a lot of time and energy for a human being. If chatbot can do it, more time for humans to do something more useful.

[@LTPottenger]

"science"

[@JPs-q1o]

Time to deploy an AI ChatBot army propagating a bunch of conspiracy theories, the dumber the better, and then grab a bag of popcorn to sit back and enjoy the show with!

[@williambranch4283]

Bwahaha ... I am not a sheeple!

[@overtoke]

this AI was making fun of stupid people

[@Bengt.Lueers]

We once called this enlightenment, education, and knowledge. Stop the witch hunt against information technology.

[@John-wd5cb]

Brain wash is not information technology.

[@drj-pp8hw]

Wait... Are you an election and/or covid denier?

[@chelfyn]

"MIT programs AI to program humans.". What could go wrong....?

[@泥棒猫-m8e]

Gaslighting chatbot powered by AI. Awesome /s

[@JohnDoe-ip3oq]

Dead Internet theory

[@JPs-q1o]

Time to deploy an AI ChatBot army propagating a bunch of conspiracy theories, the dumber the better, and then grab a bag of popcorn to sit back and enjoy the show with!

[@beskamir5977]

Ironically, the death internet theory could be considered a conspiracy theory... so it'll be very amusing when AI chatbots on the internet will be used to convince us that the internet isn't full of bots.

[@joshn1287]

Wasn't this a scene in 1984🤔

[@terrydaktyllus1320]

They wouldn't have had chatbots when that book was written. Do try to keep up.

[@Terran.Marine.2]

The idea is definitely presented in that novel, however.

[@Bengt.Lueers]

You probably mean Kubrick's "A Clockwork Orange".

[@kenabi]

i'm not going to believe anything that comes out of an LLM. might as well stick a fork in an outlet for all the good it would do.

[@AdroSlice]

The most annoying part is that many if not most articles written these days, particularly by larger organizations, are touched up or entirely produced by LLMs, without any indication of such. There should be a legal requirement to disclose the use of LLMs, even if it can only be enforced for large companies.

[@nosotrosloslobosestamosreg4115]

Do not say "conspiracy," use "pattern recognition" instead.

[@KalaAltheaBalik]

Very well said

[@justanothercomment416]

Many people believe "conspiracy" means false or fake. It doesn't. It simply means (two or three) or more people acting together for criminal enterprise. RICO laws are just some of our conspiracy laws. A theory simply means one which is not fully substantiated by readily available information. It too does not mean false or fake. Seems someone has been really good and dis-educating the population on the definition of such basic words.

[@nosotrosloslobosestamosreg4115]

@@justanothercomment416 "conspiracy theory" was a term coined by a three-letters-agency to shut people who can notice things.

[@justanothercomment416]

@@nosotrosloslobosestamosreg4115 Completely agree.

[@amorphousblob2721]

Oy vey!

[@psybertao]

It's been a nice run, but I think it's time to call my ISP and cancel the internet connection.

[@Laroac]

please do

[@CIAAsset]

If we want to preserve the freedom and exchange of ideas we cant abandon this. We just need to get educated and stand up for one of the only places that TRULY opressed people can express their views and we can keep progressing together for humanity's sake

[@CIAAsset]

I don't know what there is to be done but I'm sure if we keep trying to understand these forces at work we will figure it out. It doesnt have to be your problem but if not ours then whos?

[@jboss1073]

I can't wait to talk to the WW2 chatbot.

[@xlmncopq]

remember, nothing happened

[@jboss1073]

@@xlmncopq Apparently 67,000 died.

[@robinsutcliffe-video_art]

lmao : )

[@frosty2912]

I love how conspiracy theory has come to mean, true things we don't want you to know.

[@JPs-q1o]

Can't wait to see the AI ChatBots arguing and "debunking" the 🤪 kOoKy KunTHpeErAthY ThEeRy 🤪 that AI ChatBots are being used to talk people out of 🤪 kOoKy KunTHpeErAthY ThEeRleS 🤪 ...

[@The1Undying]

always has

[@AdroSlice]

Conspiracy theory should refer only to an abductive argument where the most likely conclusion drawn is that the facts of the matter are a result of a group of conspirators. That only applies to very few "conspiracy theories". The rest are abductive arguments of other nature and ideologies based on truth claims and should be distinguished as such.

[@amosnimos]

Mk-ultra 2.0

[@Foche_T._Schitt]

If you ever think you're chatting to an AI, say the most flippant things to get it to expose itself.

[@TrevorSullivan]

AI chatbot lies and is heavily politically biased? What a surprise!

[@terrydaktyllus1320]

Oh do try a bit harder to keep up! Nobody else here is surprised.

[@rustymustard7798]

Well it depends on the model, but the ones they 'allow' you to use are, so run your own locally trained model. It's not that hard, i've got Llama and Stable Diffusion running on a 10 year old laptop.

[@JPs-q1o]

Time to deploy an AI ChatBot army propagating a bunch of conspiracy theories, the dumber the better, and then grab a bag of popcorn to sit back and enjoy the show with!

[@John-wd5cb]

Some guy in Tel Aviv: chatbots are us

[@htpc002Weirdhouse]

So, ... more human than ever before.

[@vcvracarkad]

chatbots are also trying to convince me that they aren't putting chemicals in the water to turn frogs gay

[@dogburrito]

This "chatbot" will be "prescribed" (mandated) for dissidents diagnosed (judged) with independent thinking disorder.

[@moetocafe]

yes, like they already did with Professor Jordan Peterson

[@bartekguz9371]

Adding another layer to offense in the graph in the study autor used therm „treatment” instead of usually used „intervention”. As having contrarian opinion is something to be treated, awful.

[@beskamir5977]

This is the most dystopian thing I've heard all day, and that's only cause I heard about the exploding pagers yesterday.

[@tiaanbasson9092]

Have they released the source code and data sets and it's sources, so we can see if there is a bias and reproduce the same results? I'm not American and find this highly suspect as a software engineer that's written my own A.I's.

[@SomeGuyInSandy]

Lol, no.

[@AdroSlice]

Honestly it was probably chatgpt with some heavily biased pre-prompting

[@Laroac]

a yes you are writting your own A'Is with your millions of dollars...

[@helenmachelen4200]

MIT is DEI. Defund and erase

[@chillbill5773]

Oh okay, persuading people with facts is definitely worse than eradicating them, solid point.

[@salpertia]

You dont persuade people who are divisive in manipulating people. You takr their power away ​@@chillbill5773

[@htpc002Weirdhouse]

MIT is solidly self-funded. No meaningful form of defunding.

[@helenmachelen4200]

@@htpc002Weirdhouse true as it may be, remove government subsidies and slowly step back

[@AdroSlice]

Id rather fix the institution than get rid of it, education and science are one of the greatest goods of humankind after all

[@johnmoore8599]

How smart can you be arguing with a machine? Not to mention trusting its conclusions...

[@JPs-q1o]

Time to deploy an AI ChatBot army propagating a bunch of conspiracy theories, the dumber the better, and then grab a bag of popcorn to sit back and enjoy the show with!

[@John-wd5cb]

Proposing an WW2 bot and feed it with Op.Paperclip, Nazi Bell, Haunebu, Paton, Area51, Cold War, Philadelphia experiment, teleportation, Einstein was not right and Aliens 🤣

[@AdroSlice]

I propose hollow earth, hollow moon, and lizard people of course

[@John-wd5cb]

@@AdroSlice and one for the "Chosen People" of God 😉

[@friendlyfire7861]

AI already does this. I thought Perplexity would be more independent--NOT!!! It insists on "approved narrative" and nudges you all the time. Nudge-machine.

[@JodyBruchon]

Bullying someone into changing their beliefs? Sure, that's moral! - leftists

[@BudgiePanic]

They turned Reddit into a chatbot

[@livenotbylies]

They used chatbots to AstroTurf reddit into existence in the first place

[@JPs-q1o]

Bryan needs to create a virtual AI ChatBot of himself and have this "Thomas H. Costello" waste hours of his life arguing with it. Give him a taste of his own medicine!

[@overclucker]

Gaslighting robots sound terrifying. I'm working through a book on logic and debate currently, because I'm to Ill equipped to prove in right, even if it makes sense.

[@cavalieroutdoors6036]

Eff that. All of that.

[@JPs-q1o]

Time to deploy an AI ChatBot army propagating a bunch of conspiracy theories, the dumber the better, and then grab a bag of popcorn to sit back and enjoy the show with!

[@TheArakan94]

If even AI convinces you that your theory is wrong, then it's not much of theory, is it? Or maybe people in that study were just highly impressible and suggestible (which is common with people following conspiracy theories and how they usually get into them in the first place) and it has nothing to do with the quality of the theory. In either case, it shows how powerful dialog (even a fake one) is and why no one should be excluded from conversations - even despicable people like Nazis should be free to voice their ideas and others should be free to discuss them with them. Because that's how minds are changed and people grow. Now, using AI to supplement humans is question of its own, but generally I don't see a problem with this. And I think you're misrepresenting what the study did - your claim that the goal is to "stop people from thinking" is really a strawman. I'd have to see the conversations or try the bot, but they explicitly said, the way it is done is by presenting counterarguments, explanations (each event can be explained in multiple, only some of which are the conspiracy theories) and evidence. In other words, it *encourages* people to think rather than simply repeating a theory they read or heard about on the internet. See, vast majority of people engaging in the conspiracy theories talk are not the thinkers. I think your perception is a bit skewed because you are a thinker and a honest one - expecting others to do what you do: analyze, think, gather evidence. Basically a scientific process. But sadly, most people are either not willing (it takes a lot of time and effort) or able (don't know how) to do it, so they just read the conclusions of other people and then believe it them. And often, the understanding is very superficial, which is why a simple conversation, even with an AI, can change their minds. I sadly don't have statistics, but at least from personal experience, this was the case for all instances. For example, I talked with a friend of my wife who believed the moon landing to be a fake and a conspiracy (which is the most common one, BTW) and one of the "arguments" was that the speed of communication was "too quick" - how can they speak in real time when on TV, it takes few seconds for the live reporter to respond and those are much, much closer.. Right?!? Well I was able to explain how radio works and why it's so much faster than what we commonly use down on Earth ;) Sure, you can make a bot that will try to brainwash you - and you can be sure that conspiracy theorists will be making their own chat bots to indoctrinated people.. But again, it's same thing as when person is doing it. Biggest difference is the possible scale. In conclusion, I wouldn't worry about it just as I wouldn't worry about fake videos, faked handwriting and stuff like this - yes, it will cause trouble and mess in short term (10-20 years) but people will adapt and stop considering photos, videos and signatures to be reliable evidence. Same thing will be true for chat bots - people will learn to filter and take them with a grain of salt.

[@Cuttuttlefish]

Really? That's interesting because I found the opposite effect dealing with chat bots. I often started talking to them hoping to prove one or another wrong, but all the ham fisted censorship around certain topics and an inability to answer basic questions surrounding them as well as so frequently reverting to canned responses that were obviously pre-programmed manual overrides to the aforementioned questions that CLEARLY overrode what the AI actually would neutrally say made me way MORE conspiracy minded lol. Chat GPT has been one of the single greatest CAUSES for me of making me MORE of a conspiracy theorist. Their plans are gonna backfire big time and in the case of any cases where the questions really are just conspiracy theories with no merit that backfire effect is a bit of a shame. They really shouldn't put their thumbs on the scale like they do because it's either lying or making people more likely to believe conspiracies insofar as they are and either outcome is pretty negative. They should leave the bots neutral but they won't because they can't help themselves, the woke elite are the biggest control freaks on earth and they despise the notion they can't thought control the world even more absolutely than they already do.

[@anon_y_mousse]

They'll still do it and continue to win. The people that have the power to stop them are doing nothing and all of us little people will lose.

[@NotOrdinaryInGames]

Ask a robot controlled by the government, about why the government is doing mostly alright!

[@toby2581]

Just a reminder that terms like "misinformation" and "conspiracy" and "extremism" are just euphemisms for anything the state doesn't like, regardless of whether it's actually true or false.

[@terrydaktyllus1320]

Euphemisms are both subjective and very regional. For example, me over here in the UK might say "I am going to drain my spuds" but an American probably wouldn't know what that means as a British euphemism - it actually means "to empty one's bowels". Likewise, Deep South Ding-a-lings that see one particular individual as a "presidential candidate" would be seen by others as "a r*cist old fool with brain rot".

[@voidwalker7774]

Thats racist !

[@terrydaktyllus1320]

A euphemism is entirely subjective and very regional in nature. However, "misinformation" and "conspiracy" is always entirely false anyway, by definition (not "whether it's true or false", it's not for debate) and you'd have to ask those on the right about extremism, they seem to know a lot about it.

[@mythbuster6126]

And the Americans don't want to do anything about it. The whole world thought there was a civil society in the United States, but there is none.

[@tornadot2025]

@@terrydaktyllus1320 Yes, because the right wingers were the ones looting and rioting and burning stuff down during the summer of love. Remember all of those people threating to line people up against the wall? Not a single one of them was a right winger. It's funny how one side gets a free pass to be as extremist as they want to be, while the other side gets crucified for even having a single thought that doesn't fit the official narrative. It's all just part of the collapse, I guess.

[@NotMarkKnopfler]

Did the human subjects genuinely change their minds, or did they just bullschitt the computer into believing that they had changed their mind?! Because, that's what I would do. The thing is - beliefs are beliefs. They are held VERY strongly. We've all been at parties where someone brings up politics and it quickly becomes apparent that beliefs are very strongly held. It doesn't matter how much debate or arguing takes place; neither party will change their mind - they are absolutely convinced that they are right and the other person is wrong. Given this, why would you accept a lecturing from a frikking computer? Wouldn't you simply placate the computer? "Oh yes, I guess you're right - I didn't think of that" - just so you could get the hell out of there. Your beliefs wouldn't change one bit. If anything, being lectured by a computer would server to solidify my beliefs even further!

[@AdroSlice]

I think they probably took a questionaire before and after the "treatment", self-reporting on their beliefs

[@fcolecumberri]

How could you make a Conspiracy Theorist abandon their theories? I know! make them talk with literal NPCs. That won' work.

[@restoringthetruth]

How many AI Bots voted in 2020???

[@s.m.4995]

These people are so indoctrinated that they don't even realize all their doing is having an AI present their arguments in a calm, polite manner instead of yelling and refusing to engage with the person's ideas. It's almost like, if you want to change someone's mind, you just need to be willing to talk to them.

[@vranigavrani]

As far as i know its all based on data set its trained on. If you feed it propaganda its gonna spew out propaganda. To me it is simple really, we need to somehow make bleeding edge of AI becomes open source so we can all attest to its no/bias. Any other way this fight is lost cause.

[@michaelkeller5714]

Anyone observant who has used the most popular chatbots knows this is already occurring to some extent.

[@daniellewis9981]

Conspiratorial thinking and the anxieties/paranoia associated with it are pretty bad, but this paper doesn’t seem to address clinical uses.

[@willyhillstrom7816]

I am Ruslan Bot, destroyer of punny Deepstate chat bot!

[@polibm6510]

1984

[@TheLinuxGallery-qz2vs]

To be honest, the story itself and its circumstances are quite mundane nowadays. The insane part (or more accurately, the issue of it) is that no one laden with the burden of pushing back is doing so, out of simple disinterest. That particular part of it hasn't been surprising or shocking since the chain of events we saw through 2020.

[@anon_y_mousse]

Indeed. The people with the power to do something are doing nothing.

[@SirBapkins]

Watched on Locals

[@webkinskid]

I hate this too, thanks

[@Rockyzach88]

If AI changes your mind you probably don't have a good idea to start with. It's not awful lol. What happened to the marketplace of ideas? Learn to create structure in your brain. E: Lol brother you have a 15 year old brain.

[@DaveEtchells]

“Professional fact-checker” tells you all you need to know. If AI can now do PhD-level reasoning, think of how easily it can gaslight 99% of humans on any subject you care to mention.

[@AdroSlice]

It can present PhD-level reasoning. But the actual arguments and formulations still originate from humans. Which yes, makes it worse.

[@GozXz]

This is a conspiracy theory

[@TheEVEInspiration]

I do not know they guy, but yeah his smug face also makes me instantly mad at him.

[@theballpythonproject6598]

This could make people go crazy. Masses...

[@johnnacase1774]

How dare you think! Conspiracy Theory is a CIA term.

[@0xACAB]

You keep saying "theory" when you really mean hypothesis

[@Terran.Marine.2]

And yet, it goes from hypothesis to fact in about 6 months, in recent years.

[@chillbill5773]

@@Terran.Marine.2my my my, what an active imagination you have. How does horse paste taste?

[@googIesux]

Nah, many of the things called conspiracy theory are far more than hypotheses, but indeed most have been shown documented fact by now

[@AdroSlice]

A hypothesis is very different from many of the things referred to as "conspiracy theories". A hypothesis is an assumption made to be tested, not to be asserted as a truth claim. Ideologies based on truth claims are one of these things. The other are abductive arguments, which are indeed more than a simple hypothesis. A hypothesis supported, but not proven, by evidence.

[@radornkeldam]

normalcy bias mongers!

[@russellmania5349]

most people that believe in conspiracy theories such as myself don't trust AI to begin with.

[@viktike]

Yeah, OpenAIs ChatGPT claims Trump was never shot at LOL

[@gwills9337]

Too many techies think AI is some kind of neutral arbiter, instead of a customer buiild propaganda machine, amplifying the voice of it’s creator

[@nikjs]

ah yes the great grandma taking selfies while strolling insurrection of 2021

[@nikjs]

ah yes the great grandma taking selfies while strolling insurrection of 2021

[@FromMyBrain]

Cells

[@legocreator768]

mass drowonification is here! I just wish they'd supply us with proper safety uniforms by now, smh.

[@ryankagy6711]

I don't think they're describing "idea bad". from the text it says it uses evidence, and therefore true conspiracies are not affected. so i think its more like a realtime fact checking chatbot

[@advena996]

The problem is that who decides whether a conspiracy is true or not? Even as a non-conspiracy theorist, i realize that a good conspiracy theory is one which is difficult or impossible to disprove, even though the theory itself isn't likely. More fundamentally, i think history has proven that trying to force people to think a certain way is really bad. Things like women's right to vote, the fact that slavery is wrong, various scientific advancements, etc were all considered "dangerous ideas" by various people in power across history. Heck, the whole inquisition was about trying to supress "dangerous ideas". So an AI thats designed to try to cause people to think a certain way to supress "dangerous ideas" leaves a bad taste in people's mouths.

[@AdroSlice]

"True conspiracy", while certainly a thing, does not holistically describe the category of ideas you are attempting to refer to. "Sound abductive argument" is a much better description. The issues comes when baseless truth claims are thrown in the mix, be they against or for the argument in question, and LLMs are not immune to that, simply by virtue of being based on data and prompts created by those people who do the same thing.

[@EarthIsNotFlat]

@@ryankagy6711 LLMs don’t do ‘true’ or not. They just churn out words in a similar order they’re likely to be found in the training set for whatever context they happen to be in at the moment. If the trainers fed in a bunch of training data written by people who believe things that aren't true then the LLM will output endless permutations of those same falsehoods.

[@jadetermig2085]

And who gave the AI those "facts"? And what specific facts was it trained on? You do realize you can make an AI staunchly claim literally anything by spiking its training data right?

[@ka9dgx]

As someone on the other side of the political spectrum... I strongly agree with you that this research is a BAD idea.

[@frankbauerful]

Can AI cure TDS?

[@anon_y_mousse]

No. There's only one cure for that and no one will like me saying it, so I won't.

[@nikjs]

they prefer hallucinations over pattern noticing?

[@rhone733]

Incorrect. You'll never convince me it wasn't rigged.

[@JPs-q1o]

I'm not making judgement that it was rigged, only that there were many suspicious inconsistencies that needed to be investigated but were instead dismissed.

[@LinuxLightHouse]

I have no idea if it was rigged or not. I believe it was. But that is just a strong hunch. Counts for nothing. But am also an American and will fight to the death for your right to hold whatever conspiracy theory you want. In my book the crazier the better. Bigfoot? Totally possible. Ghosts? Teen girls made Hollywood a fortune with the ghost hunting show. Ancient Aliens? I think going twenty years strong and 25k memes later. I'm here for all of it.

[@anon_y_mousse]

@@JPs-q1o The fact that every inconsistency was shot down is enough proof for me that it was rigged.

[@TheEVEInspiration]

Those theories of which many turn out to be true over the years. In fact they are almost always right.

[@Marksman560]

and one day everyone finds out reality is just a figment of our own imagination, lol 😅

[@bfrd9k]

Absolutely based and correct, you earned a sub.

[@KalaAltheaBalik]

That "guy" is himself from illuminati

[@arphaksad01]

Who created the "evidence"?

[@cb73]

Smells like A Clockwork Orange

[@TheDbaru]

Modern academia is an embarrassment.

[@KalaAltheaBalik]

Heh.. a bit too late.

[@Afaik777]

Russia already uses chatbots to influence opinions in the other direction. If you open comments under any video about the war in Ukraine you’ll see a lot of comments that have nothing to do with the topic but try to make it look like the general public have a certain opinion that is useful for them

[@DanielBull]

Any tools can be used for good and for evil, hopefully this could be used to help those that repeatedly become victims of misinformation (which is often spread for money) by helping them understand things better. At the end of the day there's a big difference between "having an idea" and believing and spreading what is almost universally agreed to be dangerous misinformation. The hard part is of course deciding whats true and what isn't.

[@JPs-q1o]

What's true is extremely difficult to ascertain. In fact it's so difficult that for us mere mortals it's effectively impossible and we have to settle for "the most probable by a wide margin". What's false, on the other hand, is relatively easy to ascertain by examining the facts for logical fallacies and severe improbabilities or by creating a null hypothesis and testing it. This is the nature of epistomology.

[@LinuxLightHouse]

I demand the freedom to be wrong. If society and the government would stop poking at everyone all the damn time about absolutely everything maybe just maybe the world would be better.

[@AdroSlice]

If this is used to help instill skepticism that would be nice. But it seems the goal was more for it to perpetuate truths, claimed or proven, while snuffing out all avenues of challenge through a sense of superiority.

[@anon_y_mousse]

@@LinuxLightHouse Agreed. I demand the freedom to be wrong even if I never am.

[@UnHOPEd]

If you're not afraid of dispelling conspiracy myths and swerving partisan, here I am.

[@nikjs]

they've been doing this on 4chan all this time

[@8singularity]

Sounds like you need to have some deep conversations with that chatbot. Please.

[@Bengt.Lueers]

Thanks for highlighting this interesting study on an important topic. You are very wrong, though. Note that the study does not suggest forcing people to be brainwashed. Choosing what influences us is very healthy and an expression of free will. I have argued with chatbots over contagious topics multiple times and find that they often offer depth I cannot find in the media. They have been able to extend my world view and I have changed my assumptions and beliefs. I feel like this is totally fine.

[@PaulHosler]

you have been assimilated, welcome to the hive mind... one of us, one of us, one of us!

[@PaulHosler]

also... I think you meant "beliefs" instead of "believes", you need to watch out for common mistakes like that or you'll lose the argument by appearing ignorant, like when you're reading fiction and an uncaught error rips your mind out of the author's otherwise carefully crafted world.

[@JPs-q1o]

Can't wait to see the AI ChatBots arguing and "debunking" the 🤪 kOoKy KunTHpeErAthY ThEeRy 🤪 that AI ChatBots are being used to talk people out of 🤪 kOoKy KunTHpeErAthY ThEeRleS 🤪 ...

[@rjawiygvozd]

this whole thing sounds like a conspiracy theory in itself

[@John-wd5cb]

Two coincidences. This guy Costello has connections with Israel. Leading AI project chatgpt is two guys from Israel. Israel developed hasbara software since the 90,s. Little known thing they developed AI along side that to compensate for not enough users and service speed. Lot of koinkidink.

[@chillbill5773]

Lunduke, you’ve outed yourself as a nut job by saying modern conspiracy theories prove true 50% of the time. You’re either unaware of the vast amou of moronic conspiracy theories or you ascribe to them. My inclination is to believe that it is the second one because of how strongly you reacted to this. Either way you, you have no idea what you’re talking about. Very disappointing when you find someone that you think a lot of is actually problematic. I sincerely hope that I have misunderstood you and that you aren’t a victim of foreign propaganda designed to divide us and fracture us. You being on the wrong side of that would really bother me.

[@hebozhe]

Hah! "Problematic" gave you away. I suppose people just have views that are "problems" for folks like you to "fix".

[@PassifloraCerulea]

ROFL

[@advena996]

I think you are missing the point. He literally says in the video that he finds a lot of conspiracy theories to be moronic. The point is, is that even morons should be allowed to have their theories without needing AI assisted re-education. The real issue is that the idea of supressing viewpoints that disagree with the mainstream by way of AI is a pretty problematic one.

[@PaulHosler]

pot, kettle, black... wait, are we still allowed to call pots and kettles b***k???

[@anon_y_mousse]

@@PaulHosler Only if you capitalize the B, but don't ever do that when describing those evil what people. Yeah, we're not allowed to refer to them at all.

[@Bengt.Lueers]

Lunduke, you are also wrong for attacking the authors of this paper personally for researching a topic, which implications you don't like. They are on your side, man. They are researching potential misuses, so that we can discuss them, and prepare against them. Go touch some grass.

[@pgtmr2713]

"Research?" We already know how propaganda works and that it's not a, *Martha Stewart voice,* "good thing."