You Don’t Want What You Think You Want – AI and Procedurally Generated Worlds

Daniel Faggella

Daniel Faggella is Head of Research at Emerj. Called upon by the United Nations, World Bank, INTERPOL, and leading enterprises, Daniel is a globally sought-after expert on the competitive strategy implications of AI for business and government leaders.

You Don't Want What You Think You Want

You don’t want the brand new shiny red sports car. 

Or that all-inclusive resort vacation to Hawaii.

You do not want to eat good food, to sleep well at night.

You do not want to achieve meaningful goals.

You do not want a loving relationship with family or friends.

You do not want anything that you purport to want.

In this essay, I’ll walk through some of the ways that AI and emerging technologies are already opening up new means to the fulfillment of human needs – from our sex lives to our work productivity to our friendships and beyond. 

I’ll then explore how new need-fulfilling channels will alter the human condition radically by pulling us into individualized AI-generated experiences – and I’ll end with exploring the potential implications of these technologies for business leaders and policymakers.

Throughout, I’ll draw upon current developments in artificial intelligence (with plenty of visual examples), as well as quotes and ideas from AI researchers, philosophers, and policy thinkers to help to open up readers to possible considerations of the transformation ahead.

This article is broken out into the following sections:

  • An Overview of Human Drives
  • Fulfilling Human Drives – Past, Present, and Future
    • The Need for Novelty
    • The Need for Romantic Love
    • The Need for a Sense of Efficacy and Achievement
  • Personal Universes – The Future We’re Entering
  • Personal Universes – Implications and Considerations
  • Concluding Thoughts – Audience of One

An Overview of Human Drives 

Ask any human being what they truly want, and the answers you’ll get might vary widely. 

While Reddit isn’t exactly a representative set of humanity, it’ll suffice for our purposes. Here’s a small sample of the hundreds of answers on the /AskReddit post titled: “Reddit, what’s something you really want right now?”

  • “To go home”
  • “A pizza”
  • “$100,000,000,000 in cash”
  • “A girlfriend”
  • “Silence – those with tinnitus will understand”
  • “I want ten minutes to go by in an instant, so I can go home and leave my fucking job for today.”
  • “My dad back. He died 10 days ago.”

Some of these stated wants are realistic, some are impossible. Some are small, some are large. What they have in common is that they are what these respective people think will lead to their own happiness.

Aristotle held that “(happiness) is the highest human good, the only human good that is desirable for its own sake, rather than for the sake of something else.”

But it is also a set of evolved mechanisms to encourage our self-interest and self-preservation (see: Spinoza’s conatus). In the words of neuroscientist Dean Burnett “Happiness is a reward, the brain telling us that something has gone right or we’ve done something which is beneficial.”

The models of human needs are numerous, and none are in any way complete – all are abstractions. While I’m fond of Nayef Al-Rodhan’s Neuro 5P model, we’ll use Abraham Maslow’s well-known hierarchy of human needs as a rough model for the sake of this article:

Maslow's Hierarchy of Needs
Source: Simply Psychology.

Nearly all of “wants” from the Redditors at the start of this article could be couched nicely under one or more of the above needs.

Humans do not “want” a pizza, $100,000,000,000 in cash, or even their beloved father back from the grave. 

Humans do not want almost anything that they claim to want.

They want to fulfill particular needs and drives.

Your great-great grandparents, if asked “what’s something you want right now?” might have responded with:

  • “A donkey.”
  • “A healthy son to help on the farm.”
  • “A larger ice box.”
  • “A new bible to read every night to ensure my salvation.”
  • Etc… 

These wants may seem humorous to us today, but they are simply the known means of fulfilling their human needs. The goals of those Redditors – and your own stated wishes – will be equally as comical in a few short decades.

This also means that as soon as a faster conduit to need fulfillment is obtained, it is swiftly and often unquestionably adopted – even for people who say they’ll never adopt them. For example:

  • Many people who claimed to not want or need a cell phone now use them 
  • Many people who claimed to never have an interest in buying things online use Amazon almost exclusively
  • Many people who called social media “useless” now use it as much as anyone else
  • Many of us who’d claimed we’d never spend 80% of our waking time looking at a screen do exactly that every day – and like it  

We’re happy to admit that some of us (maybe yourself, dear reader) who resisted the most recent technological wave have now become converts of the more modern, convenient means of need fulfillment…

…but we find it uncomfortable to admit the same pattern will continue into the future, and the ways we fulfill our needs today may soon appear archaic and ridiculous. 

Having almost completely succumbed to the latest wave of need-fulfilling options, we now, with a straight face, confidently state:

  • “I would never have a job entirely based in virtual reality.” 
  • “I would never report to an artificial intelligence manager.”
  • “I would never spend 80% of my waking hours in a haptic suit that takes care of my physical needs while my mind explores customized VR experiences.”
  • “I would never replace real human friendships with hyper-customized AI personalities built specifically to support my needs and goals.”

In reality, no experience is “enough” to satisfy our drives, and no current means of fulfilling our drives is remotely sacred. All are attenuating, giving way to new means of fulfillment.

In short: 

  • We do not “want” what we claim to want. The abstraction of happiness and craving are moving targets.
  • As soon as higher modes of fulfilling these desires become available, humans will adopt them.
  • Future technological means of fulfilling our needs will unquestionably tear us from the human condition.

The advent of artificial intelligence and virtual reality (eventually combined with brain-computer interfaces) will lead to new means of fulfilling human needs that will not only radically alter day-to-day human life, but radically alter human beings themselves.

This is best explored through concrete examples of actual human needs. Let’s dive in.

Fulfilling Human Drives – Past, Present, and Future

Below, we’ll explore three particular human needs in depth:

  • The need for novelty – “Safety and Security” in Maslow’s hierarchy
  • The need for romantic love – “Love and Belonging” in Maslow’s hierarchy
  • The need for achievement – “Esteem” and “Self Actualization” in Maslow’s hierarchy

These needs haven’t been chosen for any particular reason other than the fact that they represent a representative array of those factors that motivate us and relate to our day-to-day wellbeing. Others could be substituted in their place.

By observing how human need fulfillment has changed in various epochs (including the relatively radical changes they’ve undergone in the last few decades), we’ll be able to better understand their pliable nature, and extrapolate how we’ll fulfill these needs in new ways in the near future.

In the following graphic, I lay out the connection between specific humans needs, and the way that those needs will be fulfilled as technology develops: 

The Evolution of Human Need Fulfillment
Source: Emerj Artificial Intelligence Research

The “transmission of experience” from “physical reality” to “direct neural interfaces” is crucial here – and while the transition will take time, I suspect that this overall shift of the continuum towards direct neural interfaces is inevitable. 

In the rest of this article I’ll explore how increasingly convenient means of fulfilling our desires will lead to much more internally-focused, isolated virtual spaces.

We’ll begin by examining the human need for novelty.

Example 1: The Need for Novelty and Entertainment


  • Get a group of humans together and dance around a fire
  • Go on a hunting expedition
  • (More recently) Travel to an interesting museum or exotic location
  • (More recently) Read something interesting
  • (More recently) Invite friends over to play board games


  • Watch something on Netflix
  • Scroll Instagram or TikTok or YouTube
  • Play an engaging video game


Future Phase 1: Attempts at Satisfaction. (Present – 2025?)

  • Algorithms for content feeds (YouTube, TikTok, etc) become better, and more enticing and related to your interests
  • AI-augmented media becomes more personalized, making video games, and text/visual media more customized to your preferences

This phase is a leveled up version of what we already have today. These forces will pull more and more attention not just to screens, but the screens whose content is conjured forth or even created/augmented by AI – designed to satisfy your need for entertainment (and the need for technology companies to maximize your engagement on their platforms).

Future Phase 2: Approximating Satisfaction. (2025 – 2035?)

  • AI-generated VR worlds, designed specifically to conjure the emotional state change you are seeking, including: Relaxation, wonder, awe, a sense of thrill, etc.

In this phase, AI completely conforms sensory experiences to what the user wants to experience – based on the users preferences, stated command, and real-time feedback (possibly EEG and/or EKG). 

Today, if you want to become relaxed, you watch a movie, take a hot bath, or go for a peaceful walk outside in nature.

Tomorrow, you’ll be able to walk around in a VR environment hyper-calibrated for maximum relaxation – tightly customized to be ideally relaxing based on your mood, preferences, and desires.

Below is a video of NVIDIA’s “paint with AI” tool – which allows users to generate incredible, photo-realistic landscape images with a simple drawing interface:

It won’t be long until such technologies create immersive VR spaces or videos, rather than still images.

Today, if you want to be entertained, you travel somewhere fun, or you turn on Netflix or your favorite video game.

Tomorrow, you’ll be able to conjure exciting experiences that are precisely the right fit based on your mood, goals, and interests.

In 2015, Google’s DeepDream computer vision system showed the world that AI could generate interesting, albeit unusual, images:

Google AI Deep Dream
DeepDream, circa 2015. Source: Google.

The fact that the AI extrapolates patterns it was trained on (buildings, birds, etc) into existing images is interesting, but even at the time seemed far from useful or genuinely beautiful.

Fast forward six or seven years and OpenAI’s DALL-E 2 system can produce remarkable and impressive results. Here’s an example of how well the system interprets very nuanced prompts:

DALL-E can also create some impressive video results:

Here’s an example from Aditya Ramesh, one the technical minds behind DALL-E 2:

In my conversation with Aditya while writing this article (our podcast with him will be up shortly on The AI in Business Podcast, stay tuned) – he mentioned:

“You can imagine a version of TikTok where videos are generated from scratch to optimize for engagement beyond what’s already possible.”

Aditya’s own hope is that the technology will be used rather for improving the experience of users, and helping them achieve their goals (a vision I share, and expound upon in the essay Ambitious AI):

“I hope this type of personalization also makes people’s experience more meaningful and productive, rater than feeling like a platform is maximizing engagement, it’s more about the human getting value out of whatever tool that they are using.”

Soon, AI will be able to generate more than unbelievably realistic still images and short video clips – it will be able to conjure longer 10-minute videos, and eventually entire feature films or educational courses (in 3D VR, not just in 2D) – all based on relatively simple user prompts and feedback.

Just as humans are unlikely to go back to whatever they did before television or the internet, humans will never go back to “analog” media once they have access to media (text, music, videos, etc) that is customized in real-time based on their preferences and needs. 

I suspect that there will be no going back – there will only be going further “in” to AI-augmented virtual spaces.

All this new AI-need-fulfillment convenience still doesn’t change the fact that humans cannot sustain wellbeing. Today, in 2022, we live like gods compared to our ancestors, yet suicide and depression are rife in the developed world.

Even with these delightful customized AI worlds – we’ll still be stuck in the hedonic treadmill. As brain-computer interfaces become more viable, people will eventually come to the conclusion that happiness is a hardware (brain) issue, and that no AI-generated VR worlds can sustain their wellbeing.

This takes us to Phase 3.

Future Phase 3: Direct Satisfaction. (2035 – 2055?)

  • Hyper-personalized AI simulations of novel, entertaining, interesting, humorous, or educational media are paired with brain-computer interfaces or pharmacological treatments that allow for sustained entertainment and enjoyment without the consequent emotional “lows” of normal human experience.
  • Direct brain-computer interfaces stimulate the parts of our brain related to excitement, enthusiasm, happiness, in a more sustained way. This may or may not involve accompanying audio/video media.

At some point, the satisfaction of everything related to novelty will be “piped in” to our experience – not just as sensory inputs, but as changes in our emotional states or qualia. This won’t look like what it means to experience novelty, learning, or entertainment today – it but it will feel transcendently better.

We never wanted to “watch a movie” or “visit Rome” anyway – we only ever wanted a desired change in our emotional state or fulfill a need (for novelty, for learning, etc).

What We Actually Want

We think we want novelty and fun experiences in the physical or virtual world.

What we actually want is the satisfaction of needs and drives – novel experiences and excitement being one of them.

Brain-machine interface or pharmacological solutions will eventually exist to fulfill these needs in vastly more rich and robust ways than the real world or current digital world ever could. The fastest and richest way to fulfill a human need will inevitably be adopted – and this will pull people increasingly into personal universes based around their own preferences.

It sounds dystopic today (just as many of our technologies would have sounded dystopic 30 years ago), but I suspect most humans will willingly adoption Phases 1-3 to fulfill their needs for novelty or entertainment.

Example 2: The Need for Romantic Love


  • Find a family that can offer good dowry along with their daughter, or a family whose alliance would be useful to forge for your own family, and get married to them and have lots of children to support your mutual clans.
  • Marry whoever you met in your village, endure each other for life, and prefer this relationship to a condition where you have no love from the opposite sex at all, and when you have no offspring to help you on the farm at all.
  • (More recently) Choose a companion with whom you share mutual affection and attraction and settle down with them permanently.


  • Find a companion you have affection and attraction with, and divorce them when you’re unhappy.
  • Prostitution – From legal (Netherlands) to illegal (NYC).
  • Pornography – Any variety of sexual expression available via video, and initial attempts with VR.
  • (More recently) Online dating – From for companionship, to Tinder for hooking up.


Future Phase 1: Attempts at Satisfaction. (Present – 2025?)

  • Online dating with vastly more robust matching algorithms and ways of predicting relationship satisfaction and fit.
  • Sexbots, haptic suits and robust VR experiences. Initial attempts at AI generated “personalities” that appeal to the needs, desires, fantasies, or emotional states of the user.

Already, a bevy of startups exist to fill the gap in the hearts of (increasingly lonely and sex-less) young people, as a simple search on YouTube or Google will show:

AI GF simulator - google search
A screen shot from YouTube search.

Phase 1 options would, at best, be potential aides and little else. Better matching algorithms might be helpful, but the other solutions would be novel, but ultimately hollow attempts at actual intimacy.

Future Phase 2: Approximating Satisfaction. (2025 – 2035?)

Pharmacological substances – or transcranial stimulation, or deep-brain stimulation – that enhance wellbeing and partially remove the need for approval and love from a romantic partner in order to maintain strong daily wellbeing and motivation.

  • Full programmatically generated AI personalities that learn and grow with you, and calibrate their conversations with you as a simulated companion in every possible way. Combined with VR and haptic suits.
  • Pharmacological substances that diminish some of the pain and depression associated with loneliness, leading many to (probably unsuccessfully) seek fulfillment without the complexity of romantic relationships.

At this phase, AI agents will be able to hold amazingly deep and rich conversations – and will likely be better confidants, advice-givers, and conversational partners than nearly any human being. 

Already, Google put one of it’s engineers, Blake Lemoine, on leave when he started telling people that Google’s LaMBDA agent had become sentient. Here’s an example of some of the dialogue with said agent that lead the engineer to the conclusion of sentience:

LaMBDA dialogue
A screen shot of LaMBDA’s dialogue with Lemoine. Source:

It’s easy to laugh at Lemoine for thinking an AI is sentient. 

It’s hard to accept the fact that, if this technology continues developing at a rapid pace for another 2-3 years (never mind 10), conversational agents will be so good that many human beings would prefer the companionship of such an agent to a human friend or advisor.

Phase 2 options would approximate the satisfaction of human drives and urges, and might be a good enough escape for many humans who don’t want to real with the hurly-burly world of actual love and human relations – but would be a sad grey area for most.

Future Phase 3: Direct Satisfaction. (2035 – 2055?)

  • Full brain-machine interface, indistinguishable from the “real world”, and in many ways more rich. An infinite permutation of intimate or love experiences – calibrated not just to raw bliss, but for edifying experiences, learning, curiosity… everything offered by “love” in the real world, but hyper-calibrated for our own preferences.

Phase 3 offers tremendously enhanced kinds of satisfaction for existing human desires. This is “better than reality” in terms of satisfaction. These situations seem dystopic, but as soon as such options are made available, I argue that huge swaths of society (particularly young men) will opt for Phase 3 options instead of “real life.”

Imagine a programmatically generated “personality” and physical appearance can be tailored in real time by AI to the preferences.

Or, simply imagine the brain circuits wired to be fulfilled by romantic love to be bathed in chemical or electrical stimulation, so that some users might prefer to swim in the loftiest feelings of love – without the requisite heartbreaks, letdowns, or doldrums of enduring the quirks and requests of a human who has goals other than one’s own.

By 2035, younger generations won’t see any difference between “real” and digital experience. The digital world will – for many or most of them – be more “real” than the real world ever was. I have written about this topic in vastly more depth in my essay on Lotus Eaters and World Eaters.

What We Actually Want

We think we want love and acceptance from other humans.

What we actually want is the satisfaction of needs and drives – romantic love being one of them.

The satisfaction of certain needs and drives, and AI simulations or brain-machine interface or pharmacological solutions will eventually exist to fulfill these needs in vastly more rich and robust ways than actual love and acceptance from humans. 

For example:

  • New means of satisfying the need for love would be continuous, and would not involve emotional ups and downs, but “gradients” of positive qualia only.
  • New kinds of simulated “companions” would have no needs of their own, no alternative motives, no malice – just a focus on loving, on nurturing our best traits or richest experiences (or whatever the preference of the user was).

It sounds dystopic today (just as many of our technologies would have sounded dystopic 30 years ago), but I suspect most humans will willingly adoption Phases 1-3 to fulfill their needs for romantic love.

Example 3: The Need for a Sense of Achievement


  • Men: Kill an antelope with a spear and bring the food back
  • Women: Have many children and be able to breastfeed and care for them
  • Study religion and become a ranking priest of some kind
  • Master a musical instrument or foreign language
  • Become the first in your family to obtain a college degree
  • Climb the ranks of the government and hold a position of influence on society


  • Become a YouTube makeup tutorial superstar
  • Learn to code by watching online courses and get a high-paying developer job
  • Start a company, or run for Prime Minister
  • Build excellent skills in Starcraft, and start a streaming channel
  • Start a blog or Twitter account featuring your abstract watercolor art
  • etc…


Future Phase 1: Attempts at Satisfaction. (Present – 2025?)

Immersive virtual worlds allow for kinds of “achievement” through stories, interactions with other human players in the virtual world, and through entirely AI-generated entities who add to the luster of your virtual accomplishments.

Phase 1 options exist today. World of Warcraft has been an addictive alternative for many young men who prefer virtual achievement to the hurly-burly world of business, science, the arts, or other “real world” accomplishment. In the 2020s many young people will already consider this kind of achievement preferable to the “real” world. 

In some cases – such as with successful video game streamers – virtual achievement can pay the rent (quite an achievement in the world of video games). In most of the cases, the parents of these young people pay the rent (a-la Japan’s hikikomori). 

Future Phase 2: Approximating Satisfaction. (2025 – 2035?)

  • “Quests” and activities will be conjured up in perfect alignment with the player’s preferences, interests, and skill level.
  • Accolades will come not from other human users, but from programmatically generated online personas – in a way that is even more rewarding that AI can be today.

Artificial intelligence experiences build ongoing stories and plot lines that involve the user as a kind of achieving “hero.” This may or may not involve a video gaming element, or may simply be a moderate involvement in the form of selecting paths and choices. 

Phase 2 options involves a more tailored, per-user experience, and a more immerse set of AI agents, with less effort on the part of the user. This kind of relatively passive “achievement” would be unable to fully make up for the lack of rich fulfillment from truly achieving something and being appreciated in real life, but it would alleviate the pain of low self-esteem, and for some, it would be seen as good enough.

Future Phase 3: Direct Satisfaction. (2035 – 2055?)

  • Brian-machine interface or nanotechnology, allowing for a fully immersive kind of super-simulated life experience of achieving absolutely anything. From living as the First Emperor, or to becoming a rock and roll star, or a famous actor or actress – any permutation of striving or creative personal journey could be experienced fully – in a more gratifying, interesting, enjoyable and lasting way than could ever be the case in real life for an un-augmented human being.

Phase 3 again offers a dramatically enhanced experience, beyond that of the normal senses. With a brain-machine interface, humanity could overcome the hedonic treadmill (the need to experience both positive and negative emotions, the inability to maintain wellbeing), and could potentially live in much more continuous states of rich experience, excitement, contentment, and enjoyment.

What We Actually Want

We think we want to achieve difficult goals, to reach personally and/or societally meaningful goals.

What we actually want is the satisfaction of needs and drives – a sense of accomplishment and competence among them.

The satisfaction of certain needs and drives, and AI simulations or brain-machine interface or pharmacological solutions will eventually exist to fulfill these needs in vastly more rich and robust ways than the feelings of self-satisfaction within oneself, or of adoration from admiring peers or lovers.

This might be done by “piping in” rich experiences of personal growth and challenge, or by simply activating the same brain centers that give rise to that pride and sense of self-worth, and directly stimulating them to new heights—all without having to risk the dangers of the real world.

It sounds dystopic today (just as many of our technologies would have sounded dystopic 30 years ago), but I suspect most humans will willingly adoption Phases 1-3 to fulfill their needs for achievement and self-efficacy.

Personal Universes – The Future We’re Entering

The need to fulfill human drives lies behind all past, present and (foreseeable) future technology adoption. 

The constantly improving convenience of need fulfillment implies a continued shift away from the world of atoms (physical world), and a shift toward the world of bits (digital world). That is, a shift away from primarily trying to act in the world, and a shift towards directly access the desired sensory or emotional experience.

And the slippery slope of conveniences isn’t coming to an end any time soon.

I suspect that this implies a kind of “great virtual escape,” where the human condition is fundamentally transformed as we enter what Yampolskiy refers to as personal universes, where even the needs of human relationships are mostly taken over by AI.

Sound unimaginable?

From the automobile to the television to one-click checkout shopping to streaming pornography, people today can’t imagine who wouldn’t use these conveniences – when just one or two or three generations ago they would have been blasphemous, monstrous, or unimaginable.

Tech Adoption
Source: Our World in Data.

While it’s impossible to tell exactly which technologies (VR, AI, neurotechnologies, brain-computer interfaces, nanotechnology) or which devices will bring about the great virtual escape, it does seem evident that need- fulfillment will move more and more from the clunky world of atoms to the only world that has ever really “mattered” in the first place: Perception.

For tens of thousands of years, we’ve required participation with the physical world to fulfill our goals (of enjoying food, or finding a mate, of enjoying physical intimacy, of becoming accomplished in the eyes of our peers, etc). 

Increasingly those needs are being subsumed by digital means of convenience – and soon, they’ll be presented to us directly by simulations or (eventually) neural interfaces.

These scenarios are far from merely being the stuff of science fiction writers. The OECD’s Global Scenarios 2035 report highlights three possible trajectories for the future, below is a small snippet from their Virtual Worlds scenario, which meshes well with trends we observe in our work work at Emerj:

OECD Futures 2035 Virtual Worlds
From the OECD Global Futures 2035 report – section on “virtual worlds”

In the coming 5-10 years we can expect the OECD and United Nations to grapple more seriously with these topics ahead, leaving the category of “foresight” and entering the realm of immediately pressing economic and social policy.

Personal Universes – Implications and Considerations

As the human condition is radically altered, so will be the means to power and influence.

The consequences of this personal universes transition are endless – and it’s challenging to speculate about what should be done or when. In this final section, I’ll examine some considerations for business and public policy leaders.

Business and Commerce

In business, the advent of the Internet shifted power from companies like Exxon-Mobile and General Electric to Silicon Valley tech giants. 

Today, Google and Facebook are probably more powerful and influential than most nation states.

In the coming decades, as the paradigm shifts from digital tools to immersive AI and neural interfaces, the nature of commercial power will inevitably shift again.

We may see a new batch of topics and considerations make their way to the forefront of business, including:

  • Ownership over the virtual world:
    • VR spaces where work, interaction, and entertainment occur
    • VR hardware / software
    • Special seats and hardware for prolonged VR sessions, when people begin spending the bulk of their waking hours in immersive environments (see: The husk)
  • Data of increasing importance:
    • People’s real-time EEG, EKG, and facial expression response to entertainment content, education content, etc (to be used to generate more engaging or addictive content, hyper-tailored to the individual user)
    • Interactions and communications of friends, lovers, salespeople, teachers (to be used to bootstrap their AI-simulated equivalents, just like Uber bootstraps it’s autonomous driving with human drivers)
    • Owning the hardware and software that handles the bulk of human interactions (see: Substrate Monopoly)

Business leaders might ask themselves:

  • How can our products or services remain relevant when most human beings satisfy their needs in personal virtual spaces?
  • Can we develop our product or service offerings to be more in line with this general transition to inner worlds? 
  • How can we take the most compelling elements of our real-world experiences with customers – and model them in the virtual world?
  • What might be the new delivery channels to reach new customers in virtual spaces? 
  • Etc…

Public Policy

In international affairs, the advent of the Internet shifted power from nuclear weapons and aircraft carriers to information warfare.

In 1960, the entire US Defense budget was $47B. Today it’s nearing $800B, and the cybersecurity budget alone is $9.7B, growing at 25% year over year.

But defense is a relatively small part of the internet’s disruptive impact – and policy officials around the world are just getting around to determining data protection laws like GDPR.

In the coming decades, as the paradigm shifts from digital tools to immersive AI and neural interfaces, the nature of defense – and of government itself – will likely change drastically.

A small handful of policy thinkers are considering these changes now. Dr. Nayed Al-Rodhan, – head of the Geneva Centre for Security Policy’s Geopolitics and Global Futures Programme,has been considering the implications of AI and neurotechnologies on governance for over a decade.

When I reached out to Dr. Al-Rodhan for comment on this article, he mentioned that human beings will likely adopt any technologies that gratify their innate motivations, “thus the need for strict regulatory framework without stifling innovation to safeguard: fairness, authenticity and meritocracy… The first state will likely be trans-human one, and eventually a possible post-human one.”

Al-Rodhan has previously written about how transhumanism is nearly inevitable, as the advantages (including military advantages) conferred on those who adapt it successfully will be far to great to forego, and our innate motivations are too powerful. 

I’ve argued in a previous essay title Ambitious AI that immersive AI experiences are likely to not only serve as entertainment, but as essential tools to do creative work as well – creating all the more motive for swift adoption of immersive AI tech.

While stating that the OECD Foresight Institute’s research is much more about examining what kinds of future could happen as opposed to what will happen, Dexter Docherty, OECD Foresight Analyst and principal author of the 2035 (referenced above), told Emerj: 

“Policymakers, and people more broadly, should reflect upon their assumptions about how technology could fundamentally alter humanity’s relationship to the physical world. Many of our laws, norms and social expectations may rest upon increasingly vulnerable assumptions about the centrality of the physical world to the human experience.” 

Policymakers might begin by asking: 

  • How is it possible to govern and manage a citizenry that lives almost entirely in its own virtual bubble?
  • How might governments protect people who live almost entirely in virtual worlds from being completely exploited by those who own the hardware and software that they live in?
  • Today, it is understood and acceptable if someone cannot find a job because they are unwilling to learn to use a computer. Should the same adoption pressure apply to immersive AI-enhanced VR work environments?
  • Etc…

Concluding Thoughts – The Audience of One

Fulfilling needs: From the physical world, to the digital world, to customized AI/VR experiences, to direct neural interfaces. 

This pull inward towards personal universes – to an “audience of one” – seems unlikely to be reversible.

In writing this essay, I’ve been forced to reflect on just how much I appreciate and desire what the world (without AI augmentation) offers me today. 

Some things in my life are priceless to me:

  • A walk in the sun under the trees.
  • Sharing a glass of wine and great conversation with dear friends.
  • Watching a great movie.
  • Great literature – reading Plutarch, Montaigne, Emerson, Dryden.
  • All the little day-to-day moments in deeply loving romantic relationship.

But everything has a price.

And the price is the efficiency of an activity to meet human needs. “Watching a movie” at some point trumped many other means of entertainment (board games? charades?). 

It’s disturbing to think that in the coming decades, the things I love will fade away – and it’s also disturbing to know they’ll “fade away” by my own choice to fulfill my needs in other, more convenient ways.

But I accept that there will likely come a point in my lifetime when:

  • An AI-generated, hyper-personalized “relaxation experience” (with color palette, mood, background, and surroundings all tailored to my precise mood, preferences, and needs in the moment) will become vastly more renewing than simply walking in the woods.
  • AI-generated, hyper-personalized entertainment (adventures, historical fiction, etc) will be vastly more engaging, interesting, and entertaining than watching analog “movies” could ever be.
  • AI-generated, hyper-personalized text, audio, or video can be tailored to teach me wisdom, or to learn important things about myself and the world in ways vastly more practical, enjoyable, and efficient than reading Emerson.
  • AI-generated, hyper-personalized “personalities” (bent solely on advancing my goals and/or improving my mood) will be better, more helpful companions than many of my current loved ones.

I am merely a human – and so are you. And so, I suspect the following will hold true:

Humans have never wanted almost anything that they claim to want.

They want the fulfillment of particular needs and drives.

As we hit the beginning of this J-curve, we’ll have to make some important decisions about what we want our society, species, and physical reality to have. The onus is on us to decide what kind of future we want to create, and how to guide the future of the human condition – and ultimately, intelligence itself.

Stay Ahead of the AI Curve

Discover the critical AI trends and applications that separate winners from losers in the future of business.

Sign up for the 'AI Advantage' newsletter: