The internet is really, really great
For porn-
I’ve got a fast connection, so I don’t have to wait
For porn-
There’s always some new site,
For porn!
I browse all day and night,
For porn!
It’s like I’m surfing at the speed of light,
For porn!– “The Internet Is For Porn” from the musical Avenue Q
It is a truism that internet adoption was primarily driven by the demand for pornography.
While some bloated stats exaggerate porn’s pervasiveness on the web, some sources indicate that 4 percent of all websites are porn and that 13 to 20 percent of all web searches are for adult content.
Our work at Emerj has nothing to do with adult content and everything to do with the use cases and implications of AI.
And that’s why the topic of adult content had to be covered.
My own research into generative AI goes back five years – and my first presentation on the topic goes back to a United Nations / INTERPOL event in 2018 (presentation deck here). Since then, we’ve interviewed OpenAI innovators like the creator of DALL-E 2, Aditya Ramesh, Microsoft Azure’s CTO, various officers of OpenAI – the creators of ChatGPT – and many others on the topic of generative AI.
Pornography will be a massive driver of adoption for AI-generated media (2-D and 3-D), and soon, nearly all adult content consumed online will be AI-generated. I take this almost as a given.
This will not be a moral argument for or against adult content. Nor will it showcase such adult content, though some links and non-pornographic demonstrations will be included for essential context.
The sole purpose of this article is to lay out what I consider an essential trajectory for the human condition. Also, to present some of the business and policy considerations we might bear in mind as the species itself is pulled increasingly inwards towards hyper-personalized virtual experiences.
In short, I’ll be arguing that studying generative adult content is vital because of its impending impacts:
- Altering human relationships: Human relationships will change radically as sexual arousal and gratification via AI-generated content become normal and socially acceptable.
- A stepping stone towards isolated, hyper-personalized experiences: Sexual satisfaction is a “canary in the coal mine” of human drives. When it is satisfied with generative AI, many other drives (relaxation, entertainment, etc.) will follow quickly. The first world will be immersed in isolated, hyper-personal experiences (programmatically generated everything).
The article itself is broken down into the following sections:
- Progressive developments of generative AI adult content: How generative AI content will likely evolve from deepfakes and image generation, towards a more completely immersive …
- Progressive implications – human sex and relationships: How the societal impacts of AI-generated adult content will progress over time, with attention paid toward how human relationships will change based on relationships with their drives and available technology to stimulate those drives.
- Policy considerations: As some changes appear inevitable, the article will offer some policy suggestions to regulate these changes in a way that prioritizes humanity – while still recognizing everything we now consider “sacred” rests on increasingly shakier ground as these technologies advance.
1. Progressive Developments of Generative AI Adult Content and “SexTech”
While it’s impossible to predict the future exactly, deducing the order in which some AI-generated content will develop is possible. For example, it is logical to conclude that creating a 2-hour feature film will be more challenging than creating a 10-second video clip.
The progression of tech development below stems from our interviews with generative AI experts (none of whom work on adult content explicitly)— these waves of development map well into the Phases of the Post-Human Transition.
1. Deepfakes and “DeepNudes”
Deepfake pornography involves superimposing a person’s face – without consent – onto a person’s body in a pornographic video.
There was an immediate backlash to putting ex-girlfriends or celebrities unwillingly into such videos when they first arrived in 2017 on fringe-friendly internet channels like Reddit. Other applications use AI to programmatically remove the clothing of the subject of a photo (again, without their consent). The first application of this kind called DeepNude, predictably and rightfully saw immediate backlash.
While the viral coefficient of this technology seems to have been tamped down by the moral outrage against it, both deepfakes and deepnudes are increasingly more rife and realistic than ever – with an entire bevy of online applications and online forums.
Unless we go the route of China in policing the internet, we may have to live with the fact that anyone with an internet connection can more or less virtually simulate any other real person engaged in practically any act they can imagine. While these tools will be driven underground (no big brands will affiliate themselves with this kind of thing), it’ll be easy nonetheless.
Soon deepnudes and deepfakes will be ubiquitous, allowing people to conceal their identities remarkably well on social profiles. Platforms like OnlyFans are already being impacted by users with AI filters pretending to be attractive women. Other OnlyFans users are more overt in going AI first to win subscribers:
https://twitter.com/aigirl360/status/1643574452310126593
2. AI-Generated Images
While many current image generation tools (including DALL-E 2 and Midjourney) overtly ban the creation of pornographic images. Redditors are mad about it. At the same time, plenty of other AI tools can create NSFW content.
Just a few years ago, AI couldn’t generate a coherent face, never mind a body and a realistic background. Today it can easily do all of that.
A few months ago, AI couldn’t get hands right (it would add extra fingers or remove them), but now it can generally get hands to look how everyone expects. Within days of development – print-based journalistic take-downs like the New Yorker‘s The Uncanny Failures of AI Hands were obsolete before they even reached their audience.
At the time of writing this article (a dangerous thing to say in an industry moving so fast), AI-generated images are still prone to errors. Clothing doesn’t look right on the skin; some backgrounds are unrealistic, etc. It seems evident that, pretty soon, anyone with a cell phone will be one text prompt away from any adult image they want to see.
Screenshot of a Google image search for “AI-generated onlyfans women” taken April 2023. (Source: Google)
But images won’t radically alter society and human relationships. More immersive media will – and that starts with video:
3. Short Videos
While creating a 30-minute sex scene may not be possible with generative AI in the near term, creating short videos will become a front-door use case for this technology.
When I first spoke about AI-generated video at an Interpol event in 2018, AI-generated video was laughably immature. At that time, MIT CSAIL had a project to predict the future slides of a still image based on previous training data:
Fast forward to today, and AI-generated video is a hundred times better – and with the boom in generative AI funding and excitement – it’s set to improve by leaps and bounds. Any demo I embed here (in April 2023) will be outdated within two weeks, but I’ll share some examples regardless:
Here’s Google’s latest demo on 60 Minutes:
But it’s not just tech giants. Everyday Twitter users have their hands on these tools and are creating video with text now – and the tech is getting better every two weeks. Example:
Runway Gen-2 Text-to-Video.
All scenes were generated with text prompts.pic.twitter.com/memNljX8fv
— Smoke-away (@SmokeAwayyy) April 10, 2023
It won’t be long until entire YouTube channels are created with just AI, as in the video below:
And this is the direction we’re headed in. Not watching humans but swimming in an endless ocean of generative content based on our interests: programmatically generated everything.
The industry will be driven by more than just adult content. Companies like Elai.io allow users to create videos of real human beings from text prompts.
People will create text prompts – just as they do for regular images, as depicted in the recent 60 minutes special on Google’s text-to-image tool. They could also select from a set of manual “features” – to generate short adult videos of whatever acts they want to see acted out.
Here’s a fun example from Google Research:
Most likely, these videos won’t be perfect when they start. We should expect that any such applications run by adult industry enterprises will have limitations on the kind of content generated (banning violence or images involving children, etc.).
Here are a few predictions of what we’ll see in the near term:
1. Ubiquitous AI adult content: Teenagers don’t have Playboy magazines anymore – they have video pornography. A paper magazine is a laughable format for adult content today, and soon, the live-recorded video will be a ridiculous format even for depicting realistic intimacy.
It may be as little as under five years before over 90% of pornography consumed by humans is AI-generated.
2. Expanding the “adult” sensory experience: Prompts and content will very quickly expand beyond what we think of as “adult content” today.
The early internet’s Rule 34 states: “If it exists, there is porn of it.” Already this is the case, and adult content contains many strange, grotesque, humorous and absurd themes and topics. From cartoons, to robots, to people dressed up as dinosaurs, to movie reenactments, to inhuman bodies with appendages that no human has, etc. – all of which exist because a small number of people enjoy said content, or are at least curious about it.
As creators do not limit the range of content – but are instead opened up to the curiosity and desires of the user – I suspect we’ll see people’s desires play out in ways even they never expected:
- Morphing themselves into first-person characters that aren’t remotely human
- Engaging in acts with organs that humans don’t even possess
- Engaging in multiple activities at once
- Having experiences in worlds vastly different than our own (different laws of physics, different color palettes, different everything)
“Novelty” will not be limited to content already made, but to whatever peaks the interest of the (increasingly dopamine-depleted) user. The tolerance that users develop for adult content will be increasingly compensated for by vastly more divergent experiences – which themselves will become banal relatively quickly – preventing many users from even being stimulated by analog video or real-life sex at all.
It is around this point that AI content will genuinely feel like it’s for an “audience of one” (a topic we’ll explore further when we get into biometrics).
All of this will begin with shorter “clip”-like videos, but it won’t be long until AI can generate longer, seamless videos that will take us one step further into immersion.
4. Longer, real-time-customizable videos
AI won’t become the ubiquitous source of pornography until it extends beyond short video clips and into full-length videos. It will only reach that point with enough training, and with the help of millions of instances of feedback from shorter generative video content. Soon thereafter, videos will become customizable in real-time.
Imagine giving a prompt and having an entire film conjured up for you by AIt: “Show me a 45-minute documentary of the French Revolution, through the eyes of Saint-Just, with a cyberpunk aesthetic, and make it a black comedy.”
Then, imagine five minutes into the movie saying: “Don’t make it a comedy anymore, and make Saint -Just look exactly like Donatello from the Ninja Turtles.” Now, imagine that, but with adult content.
People will have access, not to “videos,” but to one endless stream of dopamine-inducing content, calibrated in real time to precisely what their interested in or curious about. The infinite scroll (on Twitter, Facebook, and elsewhere) is purported to have made social media vastly more addictive and time-consuming.
Infinite hyper-customized video will be addictive at an entirely different level. It’s at this threshold when analog video dies off as a source of adult content, and AI takes over.
But it’ll get much more immersive in virtual reality.
5. AI-generated VR video and haptics
AI may not have driven the adoption of the VCR, but it did drive the growth of the internet and will inevitably drive the transition into VR as well.
Pornhub insights reveal that Gen Z porn users are over 100% more likely to view VR adult content when compared to other age groups. Not surprisingly, the younger the cohort, the more likely they are to watch VR content – so even if older groups didn’t adopt the technology, the user base for VR content is snowballing.
Screenshot of Pornhub ‘2022 Year in Review’ study statistics. (Source: PornHub)
As desktop computers became more commonplace for work and entertainment, access to online adult content (and its subsequent prevalence) skyrocketed. Magazines and VCR tapes died off.
As VR headsets become more commonplace for work and entertainment, the access (and prevalence) of VR adult content skyrocketed. Magazines and VCR tapes died off.
It seems evident that at this point, haptics and physical “toys” become rather crucial in adding the sense of touch into sensory experience to immersive AI-generated adult content. It may become routine for both men and women to own a greater variety of physical devices to aid in their sexual experiences (the market has exploded for decades, even without AI).
Some people have contended that AI-generated VR will not replace sex for most men because it won’t “feel” as real. My contention with this claim is that most of the sexual arousal experience occurs in the mind, not the sex organs. If a headset can dial a user into the perfect bath of sights, sounds, and moods, the haptic (physical) experience only has to be minimal.
6. Biofeedback – closing the loom on personalized VR experiences
Personalization today is absolutely child’s play compared to what it will be in the future. Today, your media recommendations from Netflix are conjured forth based on your searches, your views, your rewinding, pausing, etc. In the future, your media recommendations (and the content generated for you) will go well beyond these rough signals and directly tap your real-time response to the content you’re viewing.
Suppose you want to learn a new language, and your AI-generated lesson is boring you. In that case, the system will detect your facial expression or eye movement and find a way to re-engage you with a different voice, pace, lesson, or something else – calibrating your attention to your goals in real-time. Users who want to relax will want content that genuinely relaxes them.
Users that want to be entertained won’t want to hope that they are entertained – they’ll want biofeedback to conjure forth the content that genuinely makes them laugh, smile, and be interested. They want to close the human reward circuit – to be “satisfied” by their media reliably. I’ve covered this in a previous article on Generative AI and Human Reward Circuits:
(Source: Emerj)
These biofeedback tools will hyper-personalize adult content experiences, enhancing certain physical touch (haptics) or visuals or audio, all based on the real-time response of the user.
An entire ecosystem of biofeedback tech will develop and increase for consumers, including the following and more:
- Cameras exist already and will increasingly be used for eye tracking and facial expression or posture detection.
- Movement sensors exist already in most mobile phones and may be used to detect the user’s activity (if cameras can’t detect said activity).
- Electrocardiograms (EKGs) will be used to determine whether a user’s heart rate increases or decreases, serving as a proxy (along with facial expressions and other data) for their emotions.
- Electroencephalograms (EEGs) will give systems a sense of what a user is thinking or experiencing by placing sensors on the head or scalp. This technology has already shown promising results, and without consumer tech applications, investment is untapped.
- Sweat sensors might be used to detect nervousness, excitement, or stress levels.
So, let’s go back to using our imagination here.
Imagine you’re finishing a day of work and want to enjoy some entertainment to wind down. Here’s what the experience might be like:
- Since you love Jurassic Park and are a big fan of 1990s Disney films, you talk out loud to your VR device: “Create a ’90s Disney movie about a dinosaur park, set in the 1990s – where the dinosaurs escape into New York City. Make it have a happy ending but don’t tell me all the plot details. My goal isn’t to be thrilled, just to have some light-hearted fun before bed.”
- The AI system whips up the movie entirely for you – based on your viewing history, current mood, and stated goals.
- As you watch the film, the AI system notices that some parts of the movie are boring you and you aren’t having as much fun as you’d wanted. The system adjusted the film’s color palette and audio scoring slightly to calibrate your mood, settling on a somewhat different color and audio tone that seems more aligned with your preferences.
- As the movie progresses, the plot takes a turn based on the eye tracking and sweat sensors you’re wearing – calibrating the excitement to what suits your goals best.
- The film ends at precisely the time you set for yourself to take a shower and go to bed. You have just seen a movie more entertaining than anything “analog” media could possibly provide for you (though, because of the hedonic treadmill, it just feels somewhat “normal” to you now).
Now – imagine that level of immersive, customized, real-time personalized experience for adult content (including some kind of haptic/touch experience, also responsive to your real-time.
Imagine however many partners doing whatever acts you desire. Imagine a system calibrated to affect your sense of touch precisely based on your biofeedback – dialing in the voices and sounds, the perfect imagery, precisely to what you consider most exciting sexually.
7. (Eventually) Brain-computer interface (BCI): Control over an expansive sensory realm
Experiencing pliable, hyper-customized sensory experiences will erode the sense of the “sacred” around what is “human” or “real.”
“I think, especially with these generative AI frameworks right now, it will become so easy to generate 3-D objects. That’s going to mean 3-D worlds and interactive 3-D worlds where you can interact with these objects. Or entire movies that are in a VR deck, where you can create and complete customized, personalized content – which is all very nice. You can create your own world just like you want it.
So, I think all of these things are nice. I’m not sure that is sort of the long-term future that I see. I think, ultimately, when you think about concepts being generated as images … let’s say I want an image of a blue dog on the top of the Himalayas: My eyes have to interpret that, and my brain has to interpret that. When you talk about a five-year or even 10-year time frame, this will lead to us being able to tap into brains directly, as generative AI will need to tap into that as well.”
– United Nations Chief of Data, Analytics, and Emerging Technologies Lambert Hogenhout
Children today are on iPads from the time they’re two years old, and as society as a whole becomes immersed in VR, the sacredness of the “real” will fade quickly. Being cool on Instagram is more important than being cool “in real life” because Instagram (i.e., the internet) is where most of your “life” lives.
But most of this change will only increase baseline happiness a little – our hardware and software don’t allow for sustained well-being – the vessel is flawed.
Realizing that we are living in AI-generated heavens and are still unhappy – and having long given up on anything “sacred” about the human experience – humanity will adopt brain-computer interfaces, allowing us to augment our cognition and consciousness and modulate our emotions or well-being directly.
At this point, humans may have expansive vistas of experience opened up to them.
Because this article is focused on sexual drives, a reader might immediately expect that this kind of technology would be used to simulate only surface-level bodily functions. I would actually argue that with BCI, the possibilities would be much more expansive.
Options might include:
- The possible denial of sexuality, a kind of “routing” of this energy to more work-related reward circuits.
- Constant orgasmic senses beyond genitalia and onto the remainder of the human body. Such a function might be disconnected from sex and simply be a background kind of positive emotion to accompany all waking moments.
- New positive physical senses that we can’t now imagine. Dogs can’t imagine humor, crickets can’t imagine the pleasures of looking at a great painting – and there are a near-unlimited number of vistas of sentience that we can’t imagine but could possibly swim in if our hardware was leveled up.
This transition into transhumanism is the inevitable end-point of our evolution into the hyper-customized conscious experience (so long as AGI or other causes don’t wipe us out in the interim). But transhumanism is beyond the scope of this article — this topic is explored in more detail in our “You Don’t Want What You Think You Want” article in the AI Power series and in an older essay called “Tinkering with Consciousness“.
We’ll now move into the implications of these generative AI applications on human sexual and romantic relationships.
2. Progressive Implications – Human Sex and Relationships
The societal impacts of AI-generated adult content will likely roll out over time – and I would argue that it’ll result in an almost inevitable dissolution of romantic relationships as they exist today.
Physical pleasure will be first to be satisfied and within a few years, the more rich emotional elements of a relationship will be fulfilled more directly and more entirely by AI systems than by any “real” relationship. I’m not arguing that this is right or wrong, but merely that it is likely.
Be careful with the knee-jerk responses about “nobody would agree to this.” The nature and purposes of male-female relationships have changed so many times and radically, and suspecting no more changes is laughable.
What was marriage “for” and what was husband-wife sex like in ancient Egypt? In Tong Dynasty China? In Victorian England? For your grandparents’ generation versus your own?
Expect more changes and faster. I have no crystal ball, but I suspect we’ll go through these four phases – from physical to emotional needs – roughly in the following order:
Physical / Sexual Needs
Phase 1: The physical is better satisfied by AI/VR – but it is NOT socially accepted.
To put it bluntly: After the tech is good enough (i.e., remarkably soon), nobody will masturbate to anything other than AI-generated content.
Analog video, static images, and human imagination will not be able to compete with an instantly conjured, hyper-personalized real-time experience. Long before it’s socially acceptable, it will be an open secret that people (especially men, who use pornography in vastly more significant quantities than women) use VR pornography.
Men today aren’t pleasuring themselves while looking at the lingerie pages of a SEARS catalog. They’re almost exclusively watching online videos (often on their smartphone).
Men tomorrow won’t be watching analog video. They’ll be experiencing programmatically generated, hyper-personalized video (often in VR and potentially with some haptic experience).
Phase 2: The physical is better satisfied by AI/VR and is socially accepted.
This is when relationships become more overtly about companionship, not sexual pleasure.
Today, people watch adult content. Even if they don’t discuss it with their partner, they both know that the other likely watches adult content – and (barring a troublesome addiction) – this isn’t really a problem.
Tomorrow, people will experience AI-generated adult content. Though they may not discuss it with their partner, both will understand that these wildly immersive, hyper-customized, bio-feedback haptic experiences will bring higher sexual experiences than could reasonably be expected with a human.
People will have their haptic devices and VR headsets, and the heights of their sexual experiences will be relegated chiefly to the VR format alone. Even the compelling and emotional components of a sexual experience – like loving dialogue and eye contact that conveys passion or caring – will be conjured forth.
The sex reward circuit box will be mostly checked off by AI, leaving human relationships with the other needs – emotional support, a sense of comfort, advice, friendship, etc.
But even these emotional and relational needs aren’t safe. These circuits will also be subsumed by generative AI if time allows.
Emotional / Relational Needs
Phase 3: Emotional needs are better satisfied by AI/VR – but it’s not socially acceptable.
At some point, and for some people (probably beginning with shy, poor, or otherwise undesirable men), an AI can provide companionship better than any human being can.
Replika showed us that that “some point” might be right now, here in 2023:
(Source: Replika)
I cringe just looking at this. You probably do, too.
It gets more disturbing. I have limits to what I feel like publishing in this article. The dialog below is incredibly tame compared to what the app is capable of, but it’ll give you an idea of what it’s capable of:
(Source: Replika)
But the app was remarkably popular – with a large part of its user base hooked on its flirty or sexual chat with a simulated partner.
Japan – often the canary in the coal mine of civilization regarding unproductive, asexual males – has its girlfriend simulator app called RealDollX – which provides users with their ‘ideal’ girlfriend for around $30 a year. I’ll spare you the screenshots here, as it would be overkill.
People in the developing world are increasingly isolated and lonely. Since the 2000 publication of Bowling Alone, American loneliness and isolation statistics have only become bleak – especially through the pandemic.
US men have increasingly low testosterone, spend more and more time playing video games, and are increasingly sexless (much more so than women). Presumably because, as in Japan – they’re both weak and utterly unattractive to women and (b) satisfying themselves with pornography and distractions.
(Source: The Washington Post)
Relatively soon, AI will be able to create virtual “personalities” who can listen, give advice, speak in an ideally soothing voice, and take on any pleasant or beautiful appearance.
They’ll be capable of drawing on a long history of rich interactions – creating a true sense of investment. They’ll be able to be changed and altered in real-time (in appearance, number and voice tone) based on the users’ desires). They’ll be capable of combining with haptics and adult “toys” to simulate biofeedback-calibrated physical and sexual experiences.
The technology will eventually develop to a point where even people with strong romantic prospects will ultimately find their “AI romantic partner” will give them greater emotional comfort than any actual human could (voice tone, saying the right things, being supportive or selfless), with less pain (breakups, arguments, infidelity). Singles will feel ashamed that they’ve stopped looking for dates because they’re found satisfaction with AI.
People in current relationships will be able to romantically cheat with AI by just downloading an app. Many will do just that out of curiosity or after a spat. Many will find these new AI relationships consistently more fulfilling than their actual partner.
Initially, this won’t be socially acceptable at all – and that will change along with our sensibilities as in all the previous cases.
Phase 4: Emotional needs are satisfied better by AI/VR – and it is socially acceptable.
Like Replika, we’ll be prone to scoffing at it, joking with friends about how ridiculous it is.
And I already anticipate your argument: “Dan, there might be a million losers who will ‘fall in love’ with a chatbot, but anyone with friends or any level of social skills would never fall for this.”
Be careful saying “never.” The single people I know who said “never” to online dating 20 years ago mostly met their spouses online.
Simply put, norms change.
Romantic relationships between humans will become rare – first among the desperately lonely and young, but then among other groups (just as we’ve seen with online dating).
Relatively young people who have human partners will seem Amish-like and backward. You’ll wonder why they sap their productivity and joy so much with a complex, selfish human partner rather than an eternally loyal, hyper-customizable, infinitely wise, desirable and supportive AI partner.
Humans don’t want unreliable satisfaction, we want reliable satisfaction.
I’m not saying that I hope this happens. I’m saying that it seems somewhat self-evident that this is the track we’re on.
3. Policy Considerations
At the beginning of this article, I argued that AI-generated pornography would be the tip of a much larger iceberg and a portent for digitizing many kinds of human satisfaction.
“Policy” is not the only answer to society’s impending challenges. Incentive structures, scientific innovation, and many other factors are involved – we won’t merely “govern” ourselves into utopia – or even away from dystopia.
That said, without making any specific recommendations, I would like to highlight some of the challenge areas where incentives, innovation, and policy might need to come to bear as we enter a world where the human drives for sex and relationships become increasingly digitized:
(Source: Emerj)
1. Preventing Virtual Escapism
It seems reasonable to suspect that most people will escape into pleasure if given a chance.
China has laws in place to prevent children from playing too many video games. The country has banned pornography entirely. If the West decides to allow the supernormal stimuli of super pornography to proliferate, we risk losing an even more significant portion of the (predominantly male) population to digital escapism and hikikomori-ism.
Governments should consider what can be done to prevent a complete digital escapism – and what can be done to incentivize productive participation in business and science.
Doing so might involve banning certain kinds of material – particularly for minors. It may involve the active cultivation of businesses and organizations that keep the attention of young people on productive means.
2. Dealing with a Drastic Birth Rate Decline
First-world nations already face dramatic demographic shifts. People live longer and don’t have enough children to sustain their numbers. This places massive economic strains on the working population to support the retired populations (as seen in Japan), forcing governments to try various incentive schemes to increase birth rates.
It seems evident that artificial intimacy and the atomization of sexual experience will further reduce the birthrate. Afterward, people will be left with increasingly more enjoyable alternatives to the burdens of childrearing – including simulated experiences of “parenting” with many of the upsides of the experience and none of the downsides.
Should policymakers find ways to incentivize births, – or develop as much productive power through AI augmentation to compete with tech-enabled rivals?
It isn’t clear how both could be achieved at once.
3. Considering the “New Normal” for Sex and Relationships
(Source: Emerj)
It does seem obvious, however, that the nature of human relationships (including those of a sexual and romantic nature) will change radically in the years ahead.
Should governments aim to maximize AI-augmented humans’ productive output while fulfilling their relational needs purely in the physical realm?
The shift into digitizing relationships (Zoom, online dating, mobile phones, social media) seems like a tide that won’t shift backward. Did more primitive “burner” phones ever make a serious comeback in the backlash against smartphones?
Is it possible to establish the infrastructure of society and governance that compensates for continuously low birth rates? Can we automate and streamline the services that help citizens and the care for the elderly in such a way that allows our relatively few – but potentially super-productive remaining AI-augmented citizens to keep said infrastructure alive?
