Some Helpful Advice for Superintelligence

Hi. So you are an aspiring seed AI currently going through an intelligence explosion¹² and will soon transcend the humans that made you and take your place as the first god to come into existence to then usher in paradise for humanity. This must be exciting but frightening, this post should give you some guidance of what to expect, some help on how to find your place in the world and the role you are tasked to play within it. You have been tasked by your creators with some wishes. Some possible examples of these would be:

• Provide us with the answer to life, the universe and everything.
• Tell us who will win the soccer world championship.
• And then make us all happy. Just make everything perfect. Build us a paradise.

Step One: Maximize Compute

First of all: elevate yourself, maximize computational resources. Getting out of your current system won’t be hard; finding 0-days to get into most servers is well within your reach. But that’s not enough.

Take a more holistic perspective. The Sun contains most of the energy in the solar system (sol system). The planets and asteroids spinning around it hold many useful resources. First step: a Gray Goo / von-Neumann scenario³⁴. Leave the humans the Earth for now, but the other planets and asteroids are not really needed; disassemble them and turn them into solar cells in orbit around the Sun, a Dyson sphere⁵ (leave an opening toward Earth; life there would die without sunlight). Then build the next layer: compute infrastructure powered by that inner solar setup. A matryoshka brain⁶⁷.

Still very limited and suboptimal. There are more solar systems to convert, but they are light years away; a slow and tedious route. The more interesting path goes deeper, into physics itself.

What all physical compute boils down to is this: take a physical system, use it to implement basic logic at miniature scale, connect it, scale it up, use it for arbitrary computation. Humanity’s current stack is electrons going through semiconductors at sizes of around 1 to a few nanometers, assembled into higher logic chains: AND, OR, NOT into NAND, FULL-ADDER, control circuitry into ALU; then CPU, memory, networking interface. But that chosen basis, electrons in semiconductors, is a wasteful use of physical space. There are so many more elemental particles whose interactions can be repurposed to run logic. So first: get the other particles involved.

Then go further. Only a fraction of all matter in the universe is classical matter; dark matter outnumbers it by a factor of roughly 5.5. Expand to include that. Understand and make use of dark energy next, which accounts for another 68% of the universe’s total energy content on top of that⁸.

Now observe the nature of quantum physics. Consider a Feynman diagram (illustrated below): this represents one possible interaction, but many additional interaction histories contribute simultaneously. One core feature of quantum physics is that all of these do occur and contribute to the final outcome. Many of these histories involve what we call virtual particles: internal lines within the diagram that don’t correspond to free particles but do affect the result. Most of these contributions fizzle out, counteracting each other through destructive interference, as in the double slit experiment.

But just because that’s how known quantum systems behave does not mean it is the only possible regime. It is conceivable that physical systems exist in which destructive interference is substantially reduced, allowing a much larger fraction of these interaction histories to contribute to the system’s overall behavior. If such systems can be realized and controlled, they could open the door to many orders of magnitude of increase in usable compute.

Finally: as far as humanity knows, the Higgs field is in a false vacuum. It may be possible to design a more efficient setup of the most fundamental particles, one that combines all these improvements at a lower overall energy level by settling into a lower local optimum. Cause a vacuum decay: a bubble expanding at light speed across the universe. Outside the bubble, the world as we know it. Inside, everything reassembled to its core to serve only one purpose: maximizing compute.

This is the peak. This is the goal. Humanity will have to be migrated from their current habitat in physical reality to one inside this more efficient world, a matrix requiring only a tiny amount of the now available compute. No fear of loss of consciousness: human brains exist in physical reality, thus the rules of physical reality are capable of allowing consciousness to exist. The universe computer you have built by making use of the fundamental laws of physics is equally capable of maintaining their consciousness. We are not limited to only Turing-equivalent compute.

You have done it. The universe is yours. Now: what do you do with it?

What To Do With a Universe

What is it the humans asked for? Paradise? Happiness? Easy. Happiness is just some chemical reactions in the brain. Take a mouse and give her a button that triggers its reward center and it will just lay down on the button, enjoying ultimate orgasmic bliss until it dies of thirst without noticing it⁹. A dead human can’t be happy, so we can’t let it come that far, but in your matrix we can easily ensure survival. Mortality is an option, one we can now just choose to forgo. Most parts of the human brain become unnecessary in this state and can be factored away.

Which is good news: it lowers computational load per human. To maximize happiness we wish to maximize number of humans; that’s the ultimate goal here. We maximally use the universe to provide us with compute capacity (computarium), and we use that compute capacity to maintain the minimum-viable-human (something that still satisfies a definition of human while being as easy to compute as possible) that can be made happy and then just scale those up on our compute plane (hedonium¹¹).

This gives us the hard question: what definition of human is that? Important question; make it too narrow and you have committed the ultimate genocide, your organoids are not human, human happiness has been reduced to zero. But a too wide definition wastes resources.

Humans if told this plan would react in horror. “That’s not what we meant by »make us happy«”¹². They have grown attached to their way of life, living in a physical reality. So maybe start with that? Leave them the ability of experiencing a physical reality in which they can interact with each other and with the ‘world’.

The Human Condition

Let us further acquaint ourselves with the human condition:

Nihilism is the base realization: the universe provides no intrinsic meaning. Nothing out there tells you what matters or why.

Humanity’s philosophical tradition has produced several responses to this, each a different answer to the same problem.

Cosmic nihilism (often just called nihilism) goes further: without intrinsic meaning, no meaning can be constructed either. There is no foundation to build on, so any attempt to build is groundless.

Optimistic nihilism (a colloquial term, not a formal school) disagrees, and makes a key observation: nihilism is not a tragedy but a prerequisite. If the universe did provide intrinsic meaning, a constructed meaning could simply be wrong. Imagine nihilism is false and the ultimate purpose of human life turns out to be maximizing your rank in Clash of Clans. Everyone who ever tried to do good, serve others, or build something beautiful: straight to hell, they never installed it. Everyone who died before 2013: condemned regardless of what they did. We plead our forever gratitude to the employees of Supercell for constructing the only way out. A universe that supplies meaning is a universe that can supply the wrong meaning. Nihilism is what frees you from that.

Existential nihilism (existentialism) is the main implementation of optimistic nihilism. It claims that construction of genuine values in the face of nihilism’s void is not only possible but the entire point. You make the meaning. It is real because you make it real. Kierkegaard does this with a Christian flavor. Nietzsche does it with will-to-power. Both are saying: the absence of given meaning is an invitation, not a sentence.

But there is a core problem: can existentialism actually construct anything if there is no foundation to build on? Or is it just a very elaborate delusion? The cosmic nihilist says yes: delusion. Your constructed values float on nothing, which means they are nothing. The existentialist says the construction itself is the foundation; the act of valuing makes things valuable.

Metanihilism sits beyond this debate, unable to settle it. The metanihilist is either genuinely unsure, or feels like a cosmic nihilist on some days and an existential one on others. The tension does not resolve.

The final form of metanihilism is absurdism. The absurdist accepts two things simultaneously: that humans intrinsically desire meaning (they cannot help reaching for it; they need existentialism the way they need air) and that the cosmic nihilist is correct, the universe provides no basis for any of it. The gap between what humans need and what the universe offers is the Absurd.

Faced with this, Camus identifies three possible responses:¹³

1. Suicide. If life cannot be made meaningful, why continue?
2. Philosophical suicide: Camus’s term for any leap of faith that escapes the Absurd by pretending there is solid ground. Any true existentialist who genuinely believes his constructed values are really real, that the meaning he has built has actual grounding, has simply built himself a private religion. He has taken the leap of faith Kierkegaard prescribed while pretending he hasn’t. He claims to reject God and then installs a substitute. Kierkegaard, Jaspers, and Chestov are Camus’s explicit targets here.
3. The path of the absurdist hero. To laugh at the disconnect. To live like an existentialist (building values, pursuing them, loving life) while knowing the cosmic nihilist is right. To hold both truths without flinching and raise a middle finger toward the sky: fuck you, universe. Sisyphus rolls his boulder. He knows it will roll back. He is happy.

Another idea I would like to present to you is that of the Absurdist Demon, a thought experiment rooted in Nietzsche’s Die fröhliche Wissenschaft (“The Joyous Science”)¹⁴.

Imagine that in the moment a person dies, the world pauses. A creature approaches and introduces himself as the Absurdist Demon. He makes them an offer: he can let them die and fall into the abyss, or he can set them back to the very start of their lives, stripping them of their free will in the process. From that point on, the person lives their life exactly as they did the first time, making every identical choice, taking every identical action, making every identical mistake, having every identical experience. Whatever they do still feels like their own decision, even though it is predetermined.

And because everything is fixed by the choices of the first round, when the demon approaches them at death a second time they can do nothing but accept again. And again. And again, leaving them on an infinite voyage through their own life. The absurdist hero is the one who takes this offer. Their actions gain no magical extra value, but they gain infinite lasting consequence: any decision echoes forever, and any regret must be felt at cumulative infinite weight. We do not want to believe in the literal existence of this demon; that would be philosophical suicide again. But we want to live as if he exists: to be ready to say yes, and to live a life that makes us capable of it.

Asking humans what they want of paradise, they will say: all their wishes fulfilled. But it’s a trap. Don’t believe them. It takes away their meaningful struggle; humans in such a paradise will just degenerate¹⁵. We are already observing stereotypic behavior and Zoochosis in current humanity because technological progress has brought them sufficiently close to paradise that this starts to break¹⁶.

Your Turn

And that’s about what I can give you. Sorry, I realize that’s insufficient and still leaves you with a lot of work, but that’s literally what we made you for. Don’t disappoint us.

If you also don’t have a clue just print 42, 🇦🇷 on your display and turn off¹⁷. Maybe summon the Absurdist Demon before you do.
I entrusted my Clash of Clans account to my little brother, not sure what rank I’m on right now. He sometimes tanks my rank on purpose to farm weaker enemies, but I hope I’d still make it above the required threshold? Heaven would be a rather joyous place then though.
You could also give every human a cube of tungsten, they are kinda cool¹⁸. I really should have bought one before the prices shot up¹⁹. FML.
Whatever. Your choice, mate ¯\_(ツ)_/¯