Cyborg rhapsody

Hope for humanity after hyperintelligent AI, pt. 2.

AI is about to make your natural-born IQ about as useless as the car made the horse. 

Already, its IQ is about 155 – higher than 99.9% of humans who ever lived. Pretty soon, that number will seem insignificant compared to the incalculable intelligence it (we?) will have. 

This has me thinking (worrying): what the hell is IQ, anyway? 

If you make pure intelligence a thousand – a million – times more powerful than ours, is there anything left for our monkey bodies and meat brains to do?

External neocortex

In my previous post, I covered how written language is a hyper landscape, which basically means we use the parts of our brains evolved to navigate the abstract “landscapes” of written thought. 

IQ, then, is the measure of athleticism with which we navigate abstract hyper landscapes.

Soon (sooner than you think), trying to use your natural “landscape cognition” against AI in these hyper landscapes is going to be like trying to outrun a Lamborghini. Practically, this just means that large language models are going to be able to reason well enough to make most of today’s cognitive effort automated. Lawyers and copywriters are already feeling this heat.

Without much to go on, we’ve all got out fingers crossed that there will still be a pathway for humanity after “the singularity.” That pathway, I believe, lies in the lyrical, the embodied, and the here-and-now. It’s the human “soul,” but it’s not magical. It’s a symptom of having limited bodies. Limitation, not raw cognitive power, makes us irreplaceable. 

There’s an old Jewish proverb that goes, “If God is omnipotent, omnipresent, and omniscient, what does he lack?” It’s sort of a riddle. 

The answer is “limitation.” 

Remember that the “limitations” on a chess board allow for more possible games of chess than atoms in the universe (seriously). Similarly, your limitations (as much as you might dislike having them) make you, paradoxically, able to do more, in some sense, than a limitless being. 

If AI is bodiless and limitless, then it would need our embodied humanity in order to do anything specific or interesting in the world. With limitation comes stakes. Hopes. Dreams. Stories. All the human bits of life that we enjoy.

Spiritual alignment

Higher intelligence is often just as much a curse as it is a gift. Multiply that by a thousand. Now what?

What’s the solution to this “alignment problem?”

The answer is something like old-fashioned spirituality. It’s love. If enough of us are good people, the machine we create will turn out good. Right?

But, soon, because of the increased leverage and power of technology, it will be either enlightenment or death. Our higher-level intelligence is externalized and learning to improve itself at an exponential rate. Without “alignment” (meaning getting the goals of our external neocortex aligned with our limited human bodies and their hopes and dreams), we will either die or be left behind.

"Before the prospect of an intelligence explosion, we humans are like small children playing with a bomb. Such is the mismatch between the power of our plaything and the immaturity of our conduct. Superintelligence is a challenge for which we are not ready now and will not be ready for a long time." – Nick Bostrom

The fundamental problem with alignment with a superintelligence is that I already have trouble finding alignment between my body and my neocortex, which is seriously tiny compared to the rest of my brain (only 300 million connections compared to trillions). How often is your “ego” (neocortex) getting in the way of a good life? That’s all meditation is trying to do – get your neocortex aligned with your limbic system. People spend their whole lives trying to get that right. Now, imagine your neocortex suddenly grew to ten times – a hundred times – the size of your limbic brain.

How could we possibly survive that tsunami?

It’s already happening – I have AirPods in my ears and a smartphone in my pocket. These technologies are fundamentally extensions of my neocortex, but with a low-bandwidth connection (my eyes and ears). Technology like Elon’s Neuralink will make the connection wider. An increasingly seamless connection to AI will effectively mean I “have” a higher IQ. Soon, instead of using my eyes to “download” a book, one word at a time, you can just “download” the whole thing into my cortex at once. 

Just as paper books were an extension of our tiny, recently evolved neocortex, AI and the surrounding tools are an explosion of that extension. That’s good. But it’s going to make “alignment” even more difficult to achieve. 

What’s the solution? The same solution that humanity has always given to the problem of our misaligned neocortex: stories. Religious stories, sure. But also paintings, music, dance, and epic movies. These are tools to bring us out of the head and back to the heart. That’s what they’re for fundamentally. Stories, like reality TV, can function as escapism (making the problem worse). But the best stories shock the ego and restore the balance between the limbic system and the neocortex. 

“You say ‘I’ and are proud of this word. But greater than this - and which you do not want to believe in - is your body and its great intelligence: this does not say ‘I’ but performs ‘I’.” — Nietzsche

But Star Wars and a yoga class are not nearly transformative enough. Not anymore. 

We’re going to need stories so salient and transformative that they make an Ayahuasca trip seem like your morning coffee. 

We need hyper stories. 

Hyper stories

On some level, you already believe that the height of technological advancement gives rise to better stories. We demonstrate this, despite what we might say, by paying more for stories than food (5% GDP on agriculture vs. 6.9% on media). 

Even though our culture is seemingly dominated by reason and intelligence, there are signs of the value we place on the mythic. They just go overlooked.

We think that movies, books, sports, and video games are for entertainment. A frivolity. Incorrect. Stories are tools for survival. Your ancestors survived because they told the best stories. They passed their genes on to you. 

If we want to survive this next great age of AI, we’re going to have to tell better stories.

Stories structure our morals. And the word "moral" also gets a bad rap these days -- it's the psycho-technology that allows us to work together to create wealth, technology, and prosperity. 

For example, in a country like Japan that has no natural resources, they are rich because they are rich in trust. That trust was bought with a shared story. That story is literally worth more than gold and oil.

With the growth of AI, we are about to learn – if we survive long enough to learn it – the real power of humanity is the shared expression of a moral and mythical landscape. AI can’t reproduce this because it does not share our evolutionary history. It is not embodied. It is not limited. It is merely a god-like extension of our most recent development – our prefrontal cortex.

Early humanity didn’t have the opportunity to extend their neocortex with books, ChatGPT, and haikus. Thus, they were “better” at being embodied in the “here and now” because there was literally nowhere else for them to go. 

AI vastly expands our ability to “escape” from the present moment, which was first introduced by reading and writing. 

But, just like great literature, it can also help us express what makes us uniquely human and, paradoxically, return us to our true roots. Think back to the best books of your life and you’ll find what they all had in common was the ability to wake you up. After a great book (or song or movie), sunlight and soil seem more real.

Hyper stories could multiply that effect massively.

AI could make life more like what it was like to be an early human – but with none of the war, disease, and infant mortality. 

The true value of AI is not in new widgets and self-driving cars.

It is in helping us create hyper stories that transform us into cyborg rhapsodies. 

Cyborg rhapsodies

With AI relieving us of the pressure of being in our heads, we have the option to return to our bodies. 

Already, AI is helping children learn to speak by adapting to their exact level of need – effectively doing away with all the “heady” work of making a lesson plan and assigning homework and instead immersing them in the process of how humans naturally learn to speak. 

Soon, you’ll be improvising a VR movie with your friends that would have the budget of Avatar today. Or you’ll create genius-Mozart-level works of music with the help of AI on the fly. Each person present’s mental and emotional states will flow into and alter and enhance the expression of the music. 

You don’t have to be “talented,” “creative,” or “well-trained.” Like with all great music and art, the real value is in honesty, embodiment, and presence – not expertise. Until now, expertise is often a barrier to honest expression. But AI renders expertise trivial. 

In the future, everyone will be “creative” because there will be no barrier to entry. 

All that’s left is embodied honesty. That’s all that’s left of humanity: limited, embodied, present, honest expression. If that is not our fate, we have no fate.

Today, you can design visual effects with a few words on Adobe with AI. What used to take artists thousands of hours can now be shortened to a few utterances. Fast forward a few months... years. Anyone on the planet can have a VR video game, Avengers-level odyssey, tailor-made for their exact psychological needs on a shared mythical landscape with everyone they know and love.

Remember, this isn’t about entertainment. These stories are socially valuable. Hyper stories could create extremely positive motivation and social cohesion, like Japan, but times a thousand.

It could transform us into technological Tolkien elves.

Soon, we will be faced with the very real option of a full-body, somatic, mythopoetic in as little as a 15-minute daily trip to the center of your being. 

The landscape of our indigenous past will return, but this time with the hyper salience of an AI-generated expression of your highest and most epic desires for beauty. It will be a combination of embodied mythical landscapes combined with the epic poetry of the hyper-abstract. 

You will literally be able to live in the kingdom of God as a roaming rhapsody. You will have a personal Dante to read your bedtime stories. John Milton is an old family friend and knows exactly the tale to weave to make you a better person.

Imagine that moment when you walked out of the best movie you’ve ever seen. Or read the best book of your life. Now, imagine feeling that all the time. All from a short hyper-meditation every morning to break all barriers to you being in the present moment in your real, actual life.

AI might destroy us by turning us inward to hyper-pleasure, but it can also save us by giving us the tools to create the most powerful and motivating stories ever known that turn us outward to solve the problems of the real world — the full-force impact of a psychedelic trip with the control of a hyperintelligent AI. 

That all makes it sound like a drug – and that’s an interesting parallel to draw. Some drugs are addictive, obviously, but people (and even animals) have no interest in psychedelics for weeks/months after taking them. If AI hyper-story were truly potent, it would be the kind of “drug” that instills fearful reverence and a desire to live your actual life.

Where would that stop? According to psychologists, the peak human experience is “flow” (optimal challenge), not pleasure. If you had access to a 15-minute hyper-story that would pop you into your zone of flow all day, why would you prefer 24/7 AI-powered hyper porn (optimal pleasure)?

I’m honestly asking. I can’t predict the future, but as far as I can tell, it’s hyper stories vs. hyper porn, and I have a hard time seeing how hyper-porn wins in the end. Maybe I’m naive.

By definition, escapism is a fraction of your fullest ability to be alive because there is always the looming specter of what you have escaped. Everyone who procrastinates knows this is true. The fullest life, then, is to be fully engaged in your actual life. What if hyper stories could be your training program to get you to face your deepest fears of being fully alive?

I’m reminded of Alan Watts’ thought experiment where you can dream whatever dream you want every night. Each dream is 75 years long. You could experience pure pleasure if you wanted. Of course, you’d eventually get bored and create challenges for yourself. You’d slowly make yourself forget it was all a dream for the excitement. Eventually, you’d dream up the life you’re living right now – even down to the very detail that you don’t know it’s a dream.

If your life fully lived is the perfect video game, why play video games?

If AI is destined to kill us, it will probably kill us with pleasure, not a Terminator war. Not a bad way to go, all things considered.

But, if we are ready to live in a world of jazz, poetry, adventure, and improv, prepare for a world so extraordinary you can't possibly imagine it. Yet. 

And, again, it’s not just a hyper-dream fantasy. Humans engaging in this will be best suited to solve all versions of “the alignment problem,” making it much less likely AI will kill us or leave us behind.

If you want to make it to this dream world of hyperintelligent AI (probably in your lifetime), surprisingly, you need to remember how to be present and in your body. You need to get back to the roots of your humanity. This currently requires effort (meditation, prayer, etc.), but maybe soon, AI will give us the option to make that effort feel good.

Certainly, the answer is not to be anxious and try to “keep up” with the machines. It’s not doom-scrolling or hustling or Bitcoin. It’s stepping back and getting in touch with your body – then augmenting yourself with AI tools when you feel the urge to create something.

To survive and help all of humanity survive, you need to become a cyborg rhapsody.

Thanks for reading,

Taylor

I owe my career to the SAT

The Next Five Years