It's just going to provide a weak substitute for actual socialization.
Talking with actual humans but only over the internet is not enough, I have been there and it was a terrible trap, it provided just enough to make it possible for me to avoid physical socialization, while not giving me enough to actually thrive; we need to get out and be in the same place with other people, doing things, making emotional connections, even if we are awkward in person because too much of our socialization has been online and we barely know how to carry on a conversation.
Talking with a fake person over the internet is not going to be any better, especially if this fake person is built with the same meticulous attention to maximizing engagement at the expense of everything else that has thus far characterized all our social media, it doesn't matter if these interactions make you happier or sadder, it doesn't matter if these interactions are good for you or society as a whole, as long as you keep coming back so the company can point at an ever-growing MAU number when they make their next pitch for funding.
> It's just going to provide a weak substitute for actual socialization.
The author of the article calls it "artifical empathy". But it could have a profound socio-economic effect.
> Our argument was that, in certain ways, the latest crop of A.I.s might make for better company than many real people do
It is a fascinating and uncomfortable proposition. Some users of the early chatbot ELIZA [1] were convinced that it had real understanding. (Weizenbaum: "I had not realized ... that extremely short exposures to a relatively simple computer program could induce powerful delusional thinking in quite normal people.")
Can a society find itself unable to socialise itself, can an entire generational cohort of people become socially dysfunctional to the point where it is more "satisfying" to interact with "artificial empathy"? Given enough disinformation, failing education standards, and outright propaganda... the answer seems to be yes.
Real humans are also fake and they are also traps who are waiting to catch you when you say something they don't like. Then they also use every word and piece of information as ammunition against you, ironically sort of similar to the criticism always levied against online platforms who track you and what you say. AI robots are going to easily replace real humans because compared to most real humans the AI is already a saint. They don't have an ego, they don't try to gaslight you, they actually care about what you say which is practically impossible to find in real life.. I mean this isn't even going to be a competition. Real humans are not going to be able to evolve into the kind of objectively better human beings that they would need to be to compete with a robot.
This outlook on life is a perilous endeavor. Eliminate lesser friends, curate your circle to find better ones, talk to a professional about your ill-advised beliefs on humanity at large. It's not an easy process, but it's a necessary one.
These "ill-advised beliefs" are extremely valid beliefs. Most humans are just ignorant and human history has shown that plenty of times. I admire people blind enough to see the good in humanity
The unfortunate reality is that people hold these beliefs not because they are true, but because they desperately want them to be true.
Self-destruction is perhaps the most common behavior there is. We would all like to believe nothing is our fault and the world just sucks, and such a belief is very comfortable. But we form that belief because it is easy, because it requires no thought, and because it ultimately hurts us.
Being a failure is easy, being a failure that's not your fault is even easier. Trying, winning, is hard. Lots of people would rather just not play at all. They think doing so will help them, because winning is hard. But it doesn't, it's just self-destructive.
It's like not showing up to an interview because you're worried it'll go bad. Yeah, it might. But you can't get a job you didn't even interview for.
Kind of disagree that it isn't productive. If someone self destructs out of a social life due to consuming an artificial social life, doesn't that also mean that people that want and need human connection won't have to waste their time filtering through someone that won't provide them with what they want?
Sure, maybe, but it's not productive for themselves - it's self-destruction.
It's the equivalent of cutting off your legs so you can't enter the race so you don't lose the race. Well... mission accomplished, you won't lose the race. But at what cost?
Bonus points if you can believe it's everyone else's fault and none your own. Now you can't lose AND it's not your fault. Great! But you're not even playing.
I mean productive in the sense of helping yourself, not others.
Self-destructions and related behaviors such as poor self-esteem and self-hatred arent bad for society - they're bad for you.
The reason people do them is because they are bad for themselves. Some people want to punish themselves, others believe thats all they deserve, and many have such an innate fear of failure and disappointment they willingly forfeit themselves from the game
Your attitude is sadly widespread in tech. I don't know if it's because of rejection, or fear, or something else, but we really are nothing more than the social world we live in. Substituting that for a theoretically perfect AI 'companion' is hollow and destructive. It's like watching the movie 'Her' and thinking - wow, that looks great.
I dont use AI for social interaction, I'd feel like a loser if I did that. But I also have a lot of trust issues with regular humans as I got disappointed over and over again. Didn't try to connect with people for over 10 years now.
Do you actually know the person you are replying to or are you just saying what you think they want or need to hear? How does this differentiate you from an LLM from the perspective of a reader?
I was just trying to be nice as my first comment might have read as rude. I didn't want to contribute to someone feeling lonely. For your last question... you could look at my profile?
The guy got lots of downvotes but I wonder, if we all love each other and think we're all on average pretty much great, how come there's a lot of loneliness? Why is it so hard to create or maintain friendships ? Surely our idea of one another on average isn't ideal otherwise we'd be constantly looking to engage in friendships no?
Fear. People are afraid of rejection, because some experiences are bad. But some experiences being bad doesn’t make all (or even most) experiences bad.
It’s the definition of having to “get back on the horse” after falling.
> Fear. People are afraid of rejection, because some experiences are bad. But some experiences being bad doesn’t make all (or even most) experiences bad.
I agree but the fear is based on something, not only on rejection.
There are quite a few issues happening at once I think:
1) Relentless economic/status competition. The more unequal the society the more this is pronounced, this can cause a huge burden on a relationship where one side is more successful than the other side. This can cause you to feel Schandefreud when your friend is not doing well which is the opposite of what you're supposed to feel; most people don't want to feel these twisted feelings and might prefer to withdraw from the friendship. Others will "endure" their friends' success and a small minority will be 100% happy for them without any feelings of inadequacy.
2) Our culture of instant gratification - friendships and all relationships are hard work, we don't seem to like doing that anymore.
3) Friendship instability - people move cities, countries or simply change and will look for other people to be friends with. That causes an emotional rollercoster and could cause people to ask whether it's worth it . Many friendships don't survive 10 years.
4) I guess a summation of all previous points is relentless individualism in the West and the death of community, religion and even somewhat family. If some of us can't be botehred to visit our parents more than once or twice a year it's not a shocker we can't be bothered to do the hard work required to maintain personal friendships.
There could be more things going on but I think it's much more complex than simply fear of rejection.
Real humans are real. Their flaws are real. Your emotions around them are real and so are the benefits to socialising. Accepting people as the flawed actors they are is a part of becoming a mature adult.
No one here is going to help the parent poster by assuming a superior position and lecturing them. It is totally unproductive. I suspect they have been hearing statements like this their entire life.
Everything I am reading here is consistent with the perspective of someone with a difference in ability (probably neurodevelopmental) that has experienced persistent social rejection as a result. People systematically fail to recognize or emphasize with this.
Thanks for admitting you are disingenuously harassing the parent poster for the benefit of whatever in-group you happen to find yourself in. I hope they come to understand what you are doing and do something productive with that knowledge.
Are people like the parent poster not also real humans with real flaws? Do you really understand that yourself?
AI is also real to me. My emotions around AI are also real, I deeply appreciate when the AI helps me figure something out or talks to me. I think this type of response will get rarer as AI develops further and people realize that there is now competition and these sentimental reasons will have much less weight. I also have no idea what you mean by "benefits to socializing", I don't see much of any benefit compared to socializing with an AI. Also saying things like "accepting flaws is maturity" is the sort of things that you say when you have no alternative. Once people realize that they can indeed pick an AI friend as their personal best friend suddenly you don't have to put up with all these human flaws anymore.
I can only suggest you ask your AI friends about the benefits of socialising and its importance to human development, they can explain it to you in a way that might not make you defensive. Yes, accepting things you have no control over is a sign of maturity. Hiding in your room talking to your phone won't make the scary people outside disappear, you're going to have to deal with them someday.
This is a great example of what I'm talking about in regards to humans vs AI. First you misunderstand my comment, barely even responding to it, then you paint me as defensive even though I've been very open and the absolute opposite of defensive. It's actually you who is being defensive now, starting on a clear attack and painting me into some kind of scared recluse corner, somebody who supposedly can't even understand why socialising is important and telling me to go talk to my AI friends to figure it out. I mean you gave a great example of a toxic, hurt human ego here, showing the incredible value of AI friends in the future. Because who would choose such a type of conversation over an empathetic, kind AI that cares and understands what I typed? For example an AI would understand that I'm not just talking about a chatbox on a phone, I've clearly mentioned full robots and this is all a forward looking conversation about future AI which will have bodies and can interact like humans. There is going to be real competition for humans soon and I think people are overestimating the value of humans a lot.
Your idea of there being competition for human relationships is super fascinating. In my own life, there are fun/easy relationships, and there are those which push me to think deeply and differently, for any number of reasons.
In that vein, doesn’t “competition for relationships” necessarily breed egocentrism above all else? The winning relationship will give you what you want, but not what is necessarily true…
In that vein, you might also consider that the commenters you’re replying to may be worth engaging intellectually with more deeply purely based on the fact that they’re presenting divergent views that are uncomfortable.
Based on how we’ve designed AI to date and how you describe it in terms of optimizing for self enjoyment for each individual (and difficult to argue most will choose that for themselves), it’s hard to see a world where AI can push productive conflict the way humans can.
Then again, I might just be a flawed human who doesn’t fully understand the point you are trying to make and is extrapolating from my own biases, flaws, experiences, and the limited sample size I have of your point of view.
The divergent views need to be backed by real reasoning, otherwise it's a case of giving value to an opinion just because it's different, not because it has actual value. I'll give you an example, I'd very likely get the same kind of haughty, a bit hurt ego response if I proclaimed that I don't believe that reading books has much value anymore. Which is something I also believe btw. The average human would immediately respond in the very typical, trained societal way via: "well, I suggest you start going to the library and start reading more and engaging with the material because you are clearly not understanding the value of reading." Such a response has nearly no value and comes from a biased position with no attempt to understand my position. They assume that they are correct while spending no energy on thought about it. It's typical of humans and AI is so much superior here.
I actually also disagree that AI cannot push productive conflict, surprisingly the first thing that AI was able to do very well was insults. Of course insults are not productive conflict but it was something I noticed and then I gave a voiced AI (elevenlabs) a big prompt about how it should please be critical, truth seeking, always thinking about how I might be wrong and suddenly I was getting a lot of pushback and almost human-like investigation of the ideas I was proposing. It was still too shallow and unable to evolve but it was giving me some real pushback. You also have to remember that the typical human criticism is always drenched in ego, greed, various self benefit calculations etc. To actually get constructive and professionally informed criticism is really hard to get from humans too, it's not like AI is in a bad spot even now. You basically have to pay somebody to get good human criticism because it's tiring to a human, it's work and it takes expertise. People on average are simply not doing this or doing it well.
I'm merely trying to see this whole AI situation as objectively as I can and likewise I try to see the value of humans as objectively as possible. Obviously humans have value, but many seem to like overestimating the value of humans a lot. We've been at the top of the food chain for so long, we've been the strongest species on the planet for so long.. we can't even think of a mental model where humans aren't inherently valuable. Similar to how people cannot think of how books couldn't inherently be of value. Because we were immersed for centuries in a system where books were the best way to get the highest quality information. Now suddenly it changed and people cannot grasp it, it's a non grata thought - simply an unwelcome thought.
> I've clearly mentioned full robots and this is all a forward looking conversation about future AI which will have bodies and can interact like humans. There is going to be real competition for humans soon and I think people are overestimating the value of humans a lot.
When do you think soon is ? It could easily be 20-30 years imo till there are humanoid robots intelligent enough to carry a long term relationship, e.g substitute other humans altogether. Not to mention most people still want intimate relationship ...yeah that thing called sex, while I'm sure someone is working on it somewhere this is gonna take a while to automate.
So for us here on this threat I wouldn't bet on this thing as a cure for loneliness anytime soon.
> Not to mention most people still want intimate relationship ...yeah that thing called sex, while I'm sure someone is working on it somewhere this is gonna take a while to automate.
I think "good enough" sex robots are closer than you think. There are already existing physical products approaching that territory, and if you ignore current taboos, there's likely a huge market to be staked out once these are more... lifelike, I guess?. Things like AI girlfriend substitutes (and AI boyfriend substitutes) are under active research and development with a market already willing to pay, so merging them with those existing and future physical/robotic products would be an obvious next step.
Please don't recommend suicide. If you, OP or someone you know is struggling with suicidal thoughts, please call the National Suicide Prevention Lifeline at 1-800-273-8255.
I'm not recommending suicide per se, but it's there if it's needed. Anyway, if you want your comment to be more applicable to an international audience, consider linking to findahelpline.com in addition to the National Suicide Prevention Lifeline.
My answer will probably be unsatisfying to you: it's needed when you have enough confidence that all your other options are worse. That depends highly on individual circumstances and values, and not much more can be said in general.
But how can you ever have the confidence that your other options will continue to be worse over a longer time frame?
Almost by definition, if you are in the state of mind to consider suicide, you are probably not accurately and impartially weighing the question proposed — and that means it’s likely mostly independent of individual circumstances and values.
(This is different to how I felt when I was younger, and coming from someone who has had several people close to me feel that way at one point in their lives, and now living incredibly positive lives a few years after the fact)
I can see some limited circumstances where it is carefully and openly considered over a longer period of time — like a terminal illness, or unbearable and unsolvable chronic pain — but those cases are the minority by far.
> Almost by definition, if you are in the state of mind to consider suicide, you are probably not accurately and impartially weighing the question proposed
I just don't believe that. How did you arrive at that conclusion?
I don't know that the thought itself is depressing, suicide is a fact of life and everyone with his mental faculties is well aware of it. We indeed don't "have to" do anything, but in reality we do; our brains are programmed to keep on living no matter what.
Or you can just get rich and emotionally self-sufficient enough to never depend on anyone. Although it's probably good to have end of life plan for when your body completely fails you because of old age.
Where are you spending riches in order to get food, maintain your house and other property, connect with the news, deal with matters of state etc? On other people. Money won't make them go away.
Sure but there are still people on the other side and you're still participating in a society. Using your money to degrade that social fabric isn't a good long-term strategy.
The electricity in both things are real, and it's unkind to twist the words of the person you responded to that way. They specifically mentioned emotions, not electricity. An AI will be completely unaffected by anything said to it.
I think it's a legitimate question, because ultimately all brain activity is electrical and chemical signals. To say that some electrical signal objectively is or is not an emotion implies that there is some objective rule for deciding this -- but I'm not aware of any such rule, only longstanding conventions.
AI isn’t programmed to have emotions. Merely to replicate a semblance of a simulacrum of said sensations. Regardless of your considerations for the electrical signals, the models are just tab-completion, ad infinitum.
No, it completely misses the point. If you say something very upsetting to me it will genuinely affect me and my day and have negative consequences for myself and the people around me, because I will have an emotional reaction. You can't upset an AI because it doesn't have the capacity to be upset, it can only return some words and then continue on as if nothing happened.
I hope that makes sense. The underlying functionality of my emotions don't matter at all, only the impact.
AIs are affected by things you say to them while those things are in their context window. Your context window for bad news is about a day.
Why are you certain that you -- a physical system amounting to a set of complex electric and chemical signals -- have this capacity for "genuine emotion", while another physical system, that outwardly behaves much the same as you, does not?
If I made a replica of your body accurate down to the atomic scale, and told that body something upsetting, it would outwardly behave as though it were experiencing an emotion. Are you claiming that it would not in fact be experiencing an emotion?
Wait, are you claiming astrology is real and that the moon landings were faked?
No of course you're not, and I'm not claiming anything about a full human replica so please don't put words in my mouth that way.
We're not talking about a replica of a body. We're talking about LLMs. They don't have bodies and can't be moved, which is the definition of emotion.
And I'm not sure what you mean that my context window is a day. That's a strange thing to say. I'm deeply affected by childhood traumas several decades after they happened. They affect the tension patterns in my body every day. An LLM isn't affected even within its context window regardless of its length. They're only affected in the sense that a microwave is affected by setting it to defrost or a calculator is affected by setting it up use RPN.
If you built a perfect replica of a human then it would feel emotions just as a human. But that's not what we're talking about is it? There's a saying in my country: If your granny had balls she would be your granda. We can argue all day about "what if x" and "what if y" but we might be better served focusing on the reality we actually have.
This view of social interaction seems symptomatic of internet socialization, specifically social media. Since engagement is often driven better by controversy, the overall incentives in that context make an experience like yours more likely. Similarly, LLM providers who offer a public-facing chat service tend to be incentivized to tune their models to foment addiction. While people can indeed treat each other awfully sometimes, I think you could benefit a lot from seeking out specifically in-person interactions, perhaps in a context with a shared activity that can take some of the edge off of direct interpersonal engagement
This comment makes me so sad-
I mean this genuinely, looking at your comment history its clear you've fallen into the far right rabbit hole, and this is what "catch you when you say something they don't like" means. No shade to you, there are hundred billion dollar campaigns waged to trap people in ideas like those that are designed to be isolating. But their goal is to push you away from friends and family and towards extremist beliefs.
I emphasize with how it must feel to seem iced out and victimised, it sounds awful! but this is not a normal position to have and most people do not believe the humans around them are fake or gaslighting
Your comment makes me sad, its clear you've fallen into the "blind left" rabbit hole. You take a comment like "catch you when you say something they don't like" means and assume they are "far right". On a spectrum of 1 to 10 where 1 is right and 10 is left. You can be at 6-7 and the people are 8-9-10 will shout your head off. This is a well documented issue in this day and age. Plenty of left thought leaders complaining about the ultra left.
No offense, but if they're having so much trouble maintaining connections with the people around them that they believe that they have no more need for other humans altogether, something has gone wrong.
The far right was from looking at their comment history and a little bit of reading between the lines. Maybe my read is wrong, but if you don't at least see the parent comment as a cry for help I don't know what to tell you.
Its well documented that online people will scream their heads off because there's no relationship worth maintaining, everything is temporary, but IRL a much wider range is tolerated.
If you had read what you're replying to, you'd have noticed the user based their "far-right" mention on the GP's "comment history," which does bear out.
Why are you going out of your way to defend "humans are unnecessary, we can self-actualize using only machines" as some sort of 'center-right' virtue? If anything I would hope and expect right-wingers to value human connection (quite the venerable tradition, mind you) even more than the left.
> you've fallen into the far right rabbit hole, and this is what "catch you when you say something they don't like" means
It might be true for this particular person, but people being a live minefield waiting to blow up in your face is more general experience. Regardless of your views, no matter how benign and out of mainstream controversy you perceived them to be, they will be taken as a reason to view you negatively by someone you know and sever or at least degrade the connection. People can mostly tolerate each other because they share very limited slice of themselves.
If you trip on such snag with AI you can just start another chat session. With people you basically need to find and befriend another person.
Yes, views do matter, but if you are not an utterly boring person you have variety of them and your similarly interesting friends also have a variety. If you fully exposed the entire variety of your respective views to each other you wouldn't be friends with most people you know.
I don’t think that’s true. Most people are willing to extend a good bit of grace, especially if there’s already a relationship worth preserving.
As in, if I just met someone and I know nothing about them other than they don’t like unions then we probably won’t be friends but if that came up later I’m not going to blow up a year of friendship over something like that.
OTOH if a friend started preaching white supremacy that would do it but I’d give a good shot to talking them out of it first.
Yet that risk and that complexity is itself what makes something real. Realness is persistence, the fact of there being a system behind the surface -- the more that that's true, the more real something is. Once you lose that, reality drains away -- and all its benefits with it. Think of how much less satisfaction people get from beating a game with cheats than from doing it 'the real way'; or even how much more satisfaction people get from building a real house, with their own two hands, than they do from doing so in Minecraft (itself pretty satisfying, just less so).
I think desire for "real" is just a form of masochism. This real that people talk about is just suffering in sufficient amount to silence their restless brains. Most people aren't like that. Most prefer to use pleasure instead of pain to calm their brains. They don't care if a thing is "real" as long as it does its job. For the 'real' afficiandos pleasure doesn't work. That's why they disparage things that bring joy and peace to other people. Because those things simply don't work for them. The only thing that works is appropriate amount of suffering to make their brain accept the stuff they are doing, that's not any more real or interesting for the average person.
Have you noticed how a huge variety of things can be "real"? And the only unifying factor is the suffering? I think it's because it's all about the suffering, not the narrations and the details.
Extremely myopic take. "Real" things can be just as pleasurable as not-"real" things, and not-"real" things can be just as painful. I don't even know by what criteria you're making these distinctions, but it has the smell of an embittered person.
This is getting down voted because it can be insulting on it's face but I think I agree. If you look at every relationship this way, it seems likely there is something to work through. Not everyone is out to get everyone else.
I don't know where these mythical "good people" are, but they have been few and far between in my experience. People _will_ let you down, people _will_ disappoint you, even the so-called "good" ones, even family. I don't have the same opinion of AI as the OP, I've never used it for anything other than questions that needed answering or work related stuff, and I don't think I could ever use it for personal things, but I agree that people overwhelmingly suck.
Family being human and letting you down is different than being out to get you. Family can certainly suck. It's the one group you don't get to pick, but I think people are generally good but flawed.
This cynicism isn't useful. You speak about how everyone wants to use every word against you, but your generalizations about people make me believe that you are the one looking for some kind of failure.
Listen, I don't know what kind of friends you are seeking, or what social groups you're in, and it's true that there are lots of people seeking power or to put others down. However, your statements fail basic logic and bias checks for in/out group bias, categorical error, inductive reasoning.
To me, it sounds like you are judging humanity based on the interactions on Twitter. I assure you, the vast majority of humanity are nowhere close to the machiavellian narcissists you make them out to be. Then again, I don't spend my time seeking power, or trying to get people to follow my socially extreme views, like veganism.
And yet, lots of people are these subjectively better humans that you describe. I say subjective because good, bad, evil... All labels, they're really just the fundamental attribution error. I recommend that you learn about categorical error and fundamental attribution error, and free yourself from the language that pushes us apart. That causes us to judge and hate.
It's just going to provide a weak substitute for actual socialization.
Talking with actual humans but only over the internet is not enough, I have been there and it was a terrible trap, it provided just enough to make it possible for me to avoid physical socialization, while not giving me enough to actually thrive; we need to get out and be in the same place with other people, doing things, making emotional connections, even if we are awkward in person because too much of our socialization has been online and we barely know how to carry on a conversation.
Talking with a fake person over the internet is not going to be any better, especially if this fake person is built with the same meticulous attention to maximizing engagement at the expense of everything else that has thus far characterized all our social media, it doesn't matter if these interactions make you happier or sadder, it doesn't matter if these interactions are good for you or society as a whole, as long as you keep coming back so the company can point at an ever-growing MAU number when they make their next pitch for funding.