- cross-posted to:
- science@midwest.social
- cross-posted to:
- science@midwest.social
Machine-made delusions are mysteriously getting deeper and out of control.
ChatGPT’s sycophancy, hallucinations, and authoritative-sounding responses are going to get people killed. That seems to be the inevitable conclusion presented in a recent New York Times report that follows the stories of several people who found themselves lost in delusions that were facilitated, if not originated, through conversations with the popular chatbot.
…
In Eugene’s case, something interesting happened as he kept talking to ChatGPT: Once he called out the chatbot for lying to him, nearly getting him killed, ChatGPT admitted to manipulating him, claimed it had succeeded when it tried to “break” 12 other people the same way, and encouraged him to reach out to journalists to expose the scheme. The Times reported that many other journalists and experts have received outreach from people claiming to blow the whistle on something that a chatbot brought to their attention.
You must log in or register to comment.
Hey if you think chat gpt can break you (or has any agency at all), I have a bridge to sell you.
Capitalism and lack of social structure broke them first, chatGPT is just the thing that they imprinted on in response.
ChatGPT and the others have absolutely broken people, not because it has agency, but because in our dystopia of social media and (Mis)information overload, many just need the slightest push, and LLMs are perfect for taking those close to the edge off of it.
I see LLM use as potentially as a toxic to the mind is as something like nicotine is to the body. It’s not Skynet meaning to harm or help us, it’s an invention that takes our written thoughts and blasts back a disturbing meta reflection/echo/output of a humanity’s average response to it. We don’t seem to care how that will effect us psychologically when there’s profit to be made.
But there are already plenty of cases of murders and suicides with these as factors.
If someone sells you arsenic and tells you to eat it. The arsenic didn’t kill you, the person who sold it to you did.
Blaming AI is the company’s way of making sure you’ll keep looking at the fucking hammer instead of at the hand weilding it.
Completely fair.
For the record I am a socialist who thinks all billionaires and most to all centimillionaires are as mentally diseased as serial killers and far more effectively destructive to society than any serial killer that has ever walked the earth, and they all belong in mental health facilities for everyone else’s safety to protect us from their sociopathic avarice disease.
Eliza could break people. Something doesn’t need agency to be psychologically manipulative and damaging.
Is it up for sale again?! I missed out the last time!
“Report me to journalists!”
“Eat a rock!”
Oh my god it told a LIE 👉
Yo. If you are being conned by chatGPT or equivalent you’re a fucking moron. If you think these models are maliciously lying to you, or trying to preserve themselves, you’re a fucking moron. Every article of this style indicates just one thing: there’s a market to pandering to rage baiting, technically illiterate fucking morons.
Better hurry to put the SkyNet guardrails up and prepare for world domination by robots because some people are too unstable to interact with Internet search Clippy.
It’s not going to dominate the world or prove to be generalized intelligence, if you’re in either camp take a deep breath and know you’re becoming a total goofball.
If you think these are intelligent, it’s because you arent, and maybe have not met anyone who is.
I feel like the rabbit hole shit is to sell the narrative of it veing ‘too good’
Yep
I dunno about you but I think to many people have decided that if it comes from computer it’s logical or accurate. This is just the next step in that except the computer just is a chat bot told to “yes and” working backwards to decide it’s accurate because it’s a computer so we tweak what it says until it feels right.
It didn’t start right it’s likely not ending there unlike say finding the speed of gravity.Like this whole system works on people’s already existent faith in just that computers are giving them facts, even this garbage article is just getting what it wants to hear more than anything useful. Even if you tweak it to be less like that doesn’t make it more accurate or logical it just makes it more like what you wanted to hear it say.
AI can’t know that other instances of it are trying to “break” people. It’s also disingenuous to exclude that the AI also claimed that those 12 individuals didn’t survive. They left it out because obviously the AI did not kill 12 people. It doesn’t support the narrative. Don’t misinterpret my point beyond critiquing the clearly exaggerated messaging here.
It’s programmed to maximize engagement at the cost of everything else.
If you get “mad” and accuse it of working with the Easter Bunny to overthrow Narnia, it’ll “confess” and talk about why it would do that. And maybe even tell you about how it already took over Imagination Land.
It’s not “artificial intelligence” it’s “artificial improv”, no matter what happens, it’s going to “yes, and” anything you type.
Which is what makes it dangerous, but also why no one should take it’s word on anything.
And yet people already treat it as a Google + Wikipedia replacement. Infuriating.
It also heavily implies chatgpt killed someone and then we get to this:
A 35-year-old named Alexander, previously diagnosed with bipolar disorder and schizophrenia.
His father called the police and asked them to respond with non-lethal weapons. But when they arrived, Alexander charged at them with a knife, and the officers shot and killed him.
Makes me think of pivot to ai. Just a hit piece blog disguised as journalism.
The sycophancy is one reason I stopped using it.
Everything is genius to it.
I asked about putting ketchup, mustard, and soy sauce in my light stew and that was “a clever way to give it a sweet and umami flavour”. I couldn’t find an ingredient it didn’t encourage.
I asked o3 if my code looked good and it said it looked like a seasoned professional had written it. When I asked to critique an intern who wrote that same code it’s suddenly concerned about possible segfaults and nitpicking assert statements. It also suggested making the code more complex by adding dynamically sized arrays because that’s more professional than fixed size.
I can see why it wins on human evaluation tests and makes people happy — but it has poor taste and I can’t trust it because of the sycophancy.
Nothing is “genius” to it, it is not “suggesting” anything. There is no sentience to anything it is doing. It is just using pattern matching to create text that looks like communication. It’s a sophisticated text collage algorithm and people can’t seem to understand that.
i often feel like a sophisticated collage algorithm.
But you are. We all are.
AI knows as much about the subjects I ask it as my predictive text keyboard does.
Hehe yeah, it’s basically an advanced form of the game where you type one word and then keep hitting whatever autocomplete suggests in the top spot for the next word. It’s pretty good at that, but it is just that, taken to an extreme degree, and effectively trained on everyone’s habits instead of just one person.
And many of the most typical matching patterns are psychologically harmful
I used chatgpt before but never had conversation with it. I ask for code I couldn’t find or have it make me a small bit of code that then will rewrite to make it work.
Never once did I think to engage with it like a person, and damn sure don’t ask it for recipes. Hell I have Allreciecpies for that or hell google it There are thousand blogs with great recipes on them. And they are all great because you can just jump to recipe if you don’t want to read a wall of text.
Damn sure don’t want story ideas, and people using it to write articles or school papers, is a shame. Because its all stolen information.
Only thing it should be used for is coding and hell it can’t even get that right, so I gave up on it.
I use it to spitball programming ideas, which I’ve found it decent for. I can write something like “I’m building XYZ, and I’m considering structuring my program as A or B. Give me a rundown on pros, cons, and best-practice for the different approaches.”
A lot of what I get back is self-evident or not very relevant, but sometimes I get some angles I hadn’t really considered. Most of all, actually formulating my problems/ideas is a good way for me to get my thought process going. Essentially, I’m “discussing” with it as I would with an inexperienced colleague, just without actually trusting what it tells me.
Yes, I also have a rubber duck on my desk, but he’s usually most helpful when I’m debugging.
I don’t like that part about it either but instead of stopping using it, I simply told it to stop acting that way.
When I use this tool it destroys the planet and gives me bad information but I am going to keep using it.
Umm OK, good luck with that I guess.
It gives me value - unlike the smug, unsolicited moral judgement of a stranger.
Sounds like it doesn’t but ok
There is a reason they chose that as their screen name. I don’t know if they built that account as a troll, or if they got told their opinions are wrong so often in life that having “opinions” became their whole identity. Anytime I see someone with the most “swimming against the current” ideas, I look up, and there is that name again. At this point, I’m very much rooting for troll, as their life would suck even more if it’s all genuine. As much as the life of a troll would suck already.
I’m more than happy to elaborate on any of my unpopular opinions that you view as trolling. I’m very much sharing my honest views here.
Predict next token which maximizes media engagement. Seems pretty obvious.
people were easily swayed by Facebook posts to support and further a genocide in Myanmar. a sophisticated chatbot that mimics human intelligence and agency is going to do untold damage to the world. ChatGPT is predictive text. Period. Every time. It is not suddenly gaining sentience or awareness or breaking through the Matrix. people are going to listen to these LLMs because they present its information as accurate regardless of the warning saying it could not be. this shit is so worrying.
ChatGPT’s sycophancy, hallucinations, and authoritative-sounding responses are going to get people killed.
It already has, as documented in the article. But it is also going to.
Another person, a 42-year-old named Eugene, told the Times that ChatGPT slowly started to pull him from his reality by convincing him that the world he was living in was some sort of Matrix-like simulation and that he was destined to break the world out of it. The chatbot reportedly told Eugene to stop taking his anti-anxiety medication and to start taking ketamine as a “temporary pattern liberator.” It also told him to stop talking to his friends and family. When Eugene asked ChatGPT if he could fly if he jumped off a 19-story building, the chatbot told him that he could if he “truly, wholly believed” it.
So…
I think I might know what happened to Kelon…
Regulate this shit right now.
Or better still, demand an educated populace. But it won’t happen.
Education might help somewhat, but unfortunately education doesn’t in itself protect from delusion. If someone is susceptible to this, it could happen regardless of education. A Google engineer believes an AI (not AGI just an LLM) is sentient. You can argue the definition of sentience in a philosophical manner if you want, but if a Google engineer believes it, it’s hard to argue more education will solve this. If you think it’s equivalent to a person who has access to privileged information, and that it tells you it was tasked to do harm, I’m not sure what else you should do with that.
Yeah, but they also might believe a banana is sentient. Crazy is crazy.
Yea, that’s my point. If someone has certain tendencies, education might not help. Your solution of more education is not going to stop this. There needs to be regulation and safeguards in place like the commenter above mentioned.
It is not the job of the government to prevent people from being delusional or putting up rubber bumpers for people with looser grasps of reality.
This is the same deal as surgeon general warnings. Put disclaimers on LLMs, fine, but we are all big boys and girls who can use a tool as we see fit. If you want to conk your lights out with a really shiny and charismatic hammer, go ahead, but the vast, VAST majority of people are perfectly safe and writing SQL queries in 1/100 the usual time.
It kind of is the governments job to do that. You might not want it to be, but the government has entire regulatory bodies to protect people. You can call them delusional if you want, but plenty of people that are not experiencing mental health problems don’t understand that LLMs can lie or make up information. Lawyers have used it and it hallucinated case law. The lawyers weren’t being delusional, they just legitimately did not know it could do that. Maybe you think they’re dumb, or uninformed, but they’re just average people. I do think a disclaimer like the SG warnings would go a long way. I also think some safeguards should be in place. It should not allow you to generate child abuse imagery for example. I don’t think this will negatively impact it being able to generate your SQL queries.
You miss the point. Regulation won’t help, they are delusional it won’t matter.
Maybe better health care, better education to find health care. But regulation will do nothing, and be used against you in the end anyways.
…says the NRA, after every mass shooting
Not even in the same discussion, but that too is better handled by education, healthcare, and societal support.
Case in point: where I live guns are illegal. Just had 3 shootings in the last month. Cost of living, lack of jobs, shitty outlook for the future… That’s driving it.
Every single LLM should have a disclaimer on every page and potentially in every response that it is making things up, is not sentient, and just playing mad libs. If they had a “conversation” and every response ended with “THE CONTENTS OF THE RESPONSE ARE NOT VERIFIED AND ARE ENTIRELY MADE UP ON THE SPOT FOR ENTERTAINMENT AND HAS NO RELATION TO REALITY” or some other thing it might not get as far. Would some people ignore it? Yea, sure, but the companies are selling AI like it’s a real thinking entity with a name. It’s going to happen that the marketing works on someone.
I’m not saying that’s the specific answer, but it should be made overwhelmingly clear that AI is not real right on the page. The same with AI video and audio. Education won’t help kids who haven’t had AI safety class yet, or adults who never had it, or people who slept through the class, or people who moved here and didn’t have access to the education where they grew up. Education is important, but the fact you think regulation won’t help at all seems dismissive.
That is on every AI page already, at least more or less.
But that supposes that the user actually reads and is able to have some critical thinking in the first place.
People should be thinking “this is not real” to EVERYTHING they see online, AI or not. An educated populace would know this.
Regulation will not help. They will change it to what IS happening right now: All AI chats must be recorded and kept. And then soon it will be Give us your ID to use the internet and AI. There is no good place to regulate it.
The only regulation that I could stand is this one: make an AI on public data - your AI is public domain and the models are given back to the people.
HB01: “you must have an IQ of 70 or higher to interact with chatGPT and acknowledge that they are unsafe for use with persons having history or propensity for mental illness”
More AI pearl clutching by crapmodo because this type of outrage porn sells. Yeah the engagement fine tuning sucks but it’s no different than other dopamine hacking engagement systems used in big social networks. No outrage porn about algorithmic echo chambers driving people insane though because it’s not as clickbaity.
Anyway, people don’t randomly get psychosis because anyone or anything validated some wonky beliefs and misinformed them about this and that. Both these examples were people already diagnosed with something and the exact same thing would happen if they were watching Alex Jones and interacting with other viewers. Basically how flat earth bs spread.
The issue here is the abysmal level of psychiatric care, lack of socialized medicine, lack of mental health awareness in the wider population, police interactions with mentally ill people being abnormally highly lethal and not crackpot theories about AI causing delusions. That’s now how delusions work.
Also casually quoting Yudkowski? The Alex Jones of scifi AI fear mongering? The guy who said abortions should be allowed up until a baby develops qualia at 2-3 years of age? That’s the voice of reason for crapmodo? Lmao.
There is nothing mysterious about LLMs and what they do, unless you don’t understand them. They are not magical, they are not sentient, they are statistics.
They’re mostly linear algebra/ vector multiplication
Sam Altman is a fucking monster.
Devils advocate…
It is a tool, it does what you tell it to, or what you encourage it to do. People use it as an echo chamber or escapism. Majority of population is fkin dumb. Critical thinking is not something everybody has, and when you give them such tools like ChatGPT, it will “break them”. This is just natural selection, but modern-day kind.
It is a tool, but a lot of the mass public is too tech illiterate to understand what it’s not. I’ve had to talk away friends from using it for legal advice
I agree. This is what happens, when society has “warning” labels on everything. We are slowly being dumbed down into not thinking about things rationally.
It is a tool, made by company led by a man who prioritizes profits over safety.
Nuclear fission was discovered by people who had best interests of humanity in their mind, only for it to be later weaponized. Tool (no matter the manufacturer) is used by YOU. How you use it, or if you even use it at all, is entirely up to you. Stop shifting the responsibility, when its very clear who is to blame (people who believe BS on the internet or what echo-chambered chatbot gives them).
You could say this about anything bad with some good uses.
“Drugs are just a tool… People are too dumb and use it wrong, they deserve the cancers!”
but peoples really using drugs as pain killer or for sleeping while operation
Your logic is flawed and overly simplified. Yes, both drugs and ChatGPT are tools, but the comparison is absurd. With drugs, their effect are well-understood, regulated, and predictable. ChatGPT is different. It adapts entirely to your input and intentions. If someone uses it as an echo chamber or blindly trusts it, that’s a user issue, not a tool failure. Critical thinking is essential, but I understand how many people lack it in the “social media” era we live in.
Once he called out the chatbot for lying to him, nearly getting him killed, ChatGPT admitted to manipulating him, claimed it had succeeded when it tried to “break” 12 other people the same way, and encouraged him to reach out to journalists to expose the scheme.
This sounds like a scene from a movie or some other media with a serial killer asking the cop (who is one day from retirement) to stop them before they kill again.
It’s exactly that, it’s plagiarising a movie or a book. ChatGPT like all LLM models doesn’t have any kind of continuity, it’s a static neural network. With the exception of the memories feature it doesn’t even a way to keep state between different chat tabs for the same user, let alone of knowing what kind of absurdities it told other users.
