People in 2024 aren’t just swiping right and left on online dating apps — some are crafting their perfect AI match and entering relationships with chatbots.
Eric Schmidt, Google’s former CEO, recently shared his concerns about young men creating AI romantic partners and said he believes that AI dating will actually increase loneliness.
“This is a good example of an unexpected problem of existing technology,” Schmidt said in a conversation about AI dangers and regulation on “The Prof G Show” with Scott Galloway released Sunday.
Schmidt said an emotionally and physically “perfect” AI girlfriendcould create a scenario in which a younger male becomes obsessed and allows the AI to take over their thinking.
“That kind of obsession is possible,” Schmidt said in the interview. “Especially for people who are not fully formed.”
I have been saying for a while, I think hyperealistic vr and believable AI personalities are going to be the ‘great filter’ that limits advanced civilization.
Given the chance to have your sexual and emotional fantasies fulfilled in a satisfying way, many many people will take it. Especially when ‘real life’ is getting harder with everything from the cost of living making the dream of ‘married with home and children’ less obtainable to hyper competitive online dating disenfranchising increasing proportions of both men and women.
Having a believable relationship with AI is far closer than we think. It doesn’t have to perfectly replicate real life, it just has to be satisfying enough in a few key ways that people begin to prefer it.
Obviously. But I haven’t yet seen anything from an AI that could fulfill either.
The problem is that a lot of men find real women scary.
Their idealised partner is, more or less, already a sexbot. A woman who dwmurely laughs at their jokes, listens intently when they talk, and showers them with praise.
They’re not looking for intelligence, humour, independent thought, creative expression; those things terrify them. They’re looking for a sexy lamp that asks them how their day went. If it can cook, clean the house, and do all the other things mommy used to do, so much the better.
Sounds like femcel commentary, no offence.
You’re looking at the AI equivalent of where stop frame animation was in 1930. It’s a few years old and rudimentary. Give it 80 years. That’s the time frame I was talking about.
50 years ago people saw the advancement of cars and assumed we’d have flying cars by now. Nobody predicted computers would be as ubiquitous as they are. The best method we have of drying clothes is still blowing warm air through them thousands of years later. Advancement isn’t guaranteed.
True. But I’m not talking about fooling people. Just lowering the amount of “willing suspension of disbelief”.
Depends on how low your standards are. I mean, there were [a small number of] people who convinced themselves that the 1960s chatbot ELIZA was a person with feelings, and the bots have only become more convincing since then. I can certainly see the modern ones fulfilling the emotional needs of someone who really, really wants to believe they’re speaking to a sapient being who cares about them, and as for the other, well, some people have pretty low sex drives or find phone sex fulfilling enough (at least for a time).
I personally agree that this is a kind of a regulator.
Like what Tao Te Ching says. Humans shouldn’t have too much of what they desire. Such a reality in some area fails them as an opportunity to learn. It’s also a dead end - you can’t have children with a robot. You can’t grow children with a robot, or maybe you can, but it will not be sufficiently complex and it will have different criteria of success.
But this doesn’t limit advanced civilization.
It’s just that in these things we are trying to cheat. Advancement to AGI, if it ever happens, should be done in its own turn. We have means to solve a lot of purely technical problems, but we haven’t yet. There’s no reason to hurry with AGI.
The reason Europe has conquered the world was that European cultures had this respect to simplicity, born from Christian morality, but also respect to choice and logic. Remove any one of these three, and you lose that power.
Europeans had sometimes less sophisticated technologies in any particular area than, say, China or Safavid Persia or Ottomans or Southeast Asian nations or even at some point some African nations. But what they had was complete and comprehensive system, civilization as a whole. They always had the lower level of the building finished before going to the next one.
This was due to that Christian modesty combined with antique philosophy.
I also think we’ve diverted from that relatively recently - around the dotcom bubble crash, maybe. I think it had deeper implications than what people think, because the trust into said philosophy started eroding at that very point. Which created imbalance in favor of forces less affected, like Microsoft and others, who have eroded it further, and the “good” forces, like Sun or Compaq or Motorola or what not, have contributed more into it with their attempts to survive after the 90s than they would if they’d die with the crash immediately, because they showed the public something that looked like a loss in honest competition.
And there’s also the climate factor. The world is going to get even more hellish in the next decades, not just hotter, but more extreme weather is near. Thanks, in parts, to the older generations (boomers), it won’t be easier for the current generations, and it’ll be even harder for the next generations (considering that humanity has not yet become extinct in the next few decades). It’s just unfathomable to bring children to this future hellish world.
Spending all time dating AI partners means that we have achieved labor post scarcity. If labor post scarcity isn’t achieved, then it means you have to do a job to survive (like now), thus not spending entire time with AI partners.
Achieving labor post scarcity means that scientific progress too would stop being connected with the economic productivity of individuals. Basically, AI scientists. Scientific progress means expansion of humanity through space.
Therefore, your great filter idea doesn’t really hold imo.
Bit of a weird non sequitur
What I’m saying has nothing to do “labour post scarcity”
I’m referring to immersive VR and AI overall contributing to a falling birthrate. If immersive realities become truly immersive, it’s reasonable to believe they will occupy leisure time. This has nothing to do with people’s relationship to work. They’ll still need to be economically active, whether or not this takes place in the VR is neither here nor there.
It’s a point about what people will do with their time when they are lonely, want connection, or pleasure. And if VR / AI (whatever other technologies) becomes believable and more satisfying then there’s little reason to believe people will continue the “unreliable” tradition of dating. And even less to engage in the mucky and very biological habit of reproducing.
Witness Japanese culture. And then just add 100 years of immersive believable AI personality and sexual fantasy. Do you think that will make people get married and have babies or do you think it will help them being content being single and childless?
Your conclusion is based upon an assumption that we need more humans to progress as society. If AI develops to the point where it is better as a partner than a human being, it likely means that we have achieved, or are very close to achieving labor post scarcity (the assumption being that an AI capable of achieving this is capable enough to do most/all human work).
When we achieve labor post scarcity, the number of humans has nothing to do with progress. Therefore, falling birthrates won’t have any negative effect on progress.
When we achieve labor post scarcity in the medical field, life expectancy would increase, with us achieving biological immortality at a certain point. This means, that death rates also go down.
Considering the above, I thought you were referring to “dating and fucking AI partners” as the end of human progress (presumably because of a lack of any motivation to cause any more development). That’s what my reply was talking about.
I understand where you’re coming from now. Yes I agree. Though I’ll add that I’m pretty confident that the sex industry is going to be at the bleeding edge of AI / VR as things progress. At least, I think the bar for making people interested in what an attractive AI has to say rather than another disappointing night on tinder is far lower than automating all human labour. Even if we’re talking physical “sexbots” I think, practically speaking, that more likely to be rudimentary ‘equipment’ greatly enhanced by augmented VR. Again, far closer to reality than Boston dynamics + son of chatGPT replacing the workforce. My point being that the bar at which young people become disinterested in physical reproduction is far, far lower than a post scarcity society in which all labour is automated. And that’s the risk. That we start to have a shortfall in workforce replacement long before we can manage without it.
There are AI girlfriends in japan long before chatGPT… It does not seem to bad to me. Sad, but not bad.
I think the part that feels ‘sad’ to you is what’s going to change socially over the next 50 years. I think it’s going to become extremely normal to at least have a “mental health AI friend” who knows you really well and keeps you going through the day, is someone to talk to, someone who’s always there, someone who’s the first to detect that you may be in danger. Overall I think society’s going to receive that as a good thing. And it will, I think, be normal because it will be so believable, and so useful, and for a large number of people, keep them well and feeling good about themselves. In that context some of those attachments turning romantic, or people just being sexually into whatever that assistant can say or do will be increasingly normal. It will also feel really good, let’s not forget that. We’re really only at the very start of what immersive VR is going to be. Once AI becomes not a little better but 50-100 years of innovation better I don’t think we can really underestimate how much it’s going to feel like you’re actually interacting with [insert fantasy here]. Once tactile feedback sees similar improvements we’re about 75% of the way to what people would use an actual holodeck for anyway. I can’t see how that doesn’t have a dramatic effect on how people view human-human romantic relationships. Over time the proportion of people who can have a believable experience of their absolute sexual fantasy is only going to grow over time. With how ubiquitous that will be I can’t see how in most relationships people know they’re “second best”. I think that has a profound effect on how people make attachments to one another. I think once “having a real girlfriend” is seen as the secondary way to get your sexual needs met, that that will have a terminal effect on how many young men even want to be in relationships let alone stay around to be a father.
That’s you. You need the skills and attention to reassure yourself and introspect.
You just gave me a new reason to dislike LLMs, they allow people to refrain from maturing.
Yea, I guess Apple is aiming somewhere in this direction by giving new AI Siri all the context information to make you feel that it cares about you 🤔
Microsoft’s Recall as well
At some point they’re going to have to depart from the “single name” branding. Because the selling point is going to be how unique the AI is for you, how different from all the others. I wonder how they’ll handle branding at that point? Maybe if you create “Marla”, Apple are still going to say it’s a type of “Siri” if that’s the base technology. But I think we may start seeing weirder more organic things like “your AI Marla - a daughter of Siri - is ready”
😂I totally see them saying that at a keynote