The Future Belongs To The Weird

or the gospel for those out of distribution

If you were born in pre-modern times, the last thing you wanted to be was weird. Because the most likely outcome of being weird was dying an untimely and painful death. Unusually weak? They will not want to feed you. Unusually strong? They will send you to fight an impossible war. Unusually clever? They will accuse you of witchcraft and burn you at the stake. Yes, there were some exceptions (for example, "freaks" kept as curiosities by the kings), but the general rule for the vast majority of the population was: don't be weird, or get killed.

The European pre-modernity was not unique in this way. Every age and culture seemed to find its own way of punishing those who had an uncontrolled tendency to look or behave out of distribution. And so trying your best to blend in, concealing your spiky sides, behaving in predictably average ways and staying "on rails" used to be the most rational life strategy for the vast majority of people who have ever lived.

To be fair, most cultures also developed deliberate valves and institutional escape hatches for excess weirdness of various kinds. These valves could be broadly categorized into 3 types: temporal, spatial and role-based. Temporal ones were restricted to certain times of the year: for example, during the Dionysian orgies the usual societal rails and expectations were temporarily suspended. Then the spatial valves are about the designated areas where you could escape the system: bugger off into the wilderness, sail to the New World and do what you want. Temporal and spatial can often be combined for maximum containment: Burning Man is both a time and a place. And then there were role-based escape hatches: the village idiot, the king's jester, untouchable castes or certain kinds of monks. If you needed to only escape the system occasionally - go knock yourself out at a festival. If you were not ready to comply with the societal rails on a permanent basis - become a hermit. Or a pirate. Or run away with the circus. Or go found a new colony (because in those wild and beautiful days the frontier was still endless and order-abiding societal enclaves were still an exception in the vast, scary, free-for-all dance of mother nature). Staying normal was expected, and rational. Escaping the system was rare and dangerous and irrational, but possible and somewhat accepted.

But slowly, and then suddenly, the frontier shrank, the industrial models of production and education marched across the Earth, and the situation changed for the weird ones. In some ways it got better, obviously: these days almost nobody gets literally burned at the stake for being too clever or contrarian (though recent outbursts of cancel culture got suspiciously close to the Inquisition). Overall, it's fair to say that the majority of the population is not facing the "don't be weird or die" dilemma.

Yet in other ways, the rails also got stronger and the escape hatches fewer. Skip college and your chances of getting past the automated CV filter are vanishingly small. Ruin your score with Meta and good luck ever really escaping the shadowban, no matter which country you try to reconnect from. Global institutions, global databases, global algorithms, global castaways. It's a lot harder these days to just "set sail to America and start a new life".

But let us not talk about the "evil system" here, for that has been talked about enough. Let us instead talk about how and why "the system" used to work and why its dominance may end in the near future.

Before we speculate how the system may end, we have to try and understand how it began and what keeps it going. So why was it good to be normal in the first place?

Humans, like all life, are engaged in a constant (and ultimately hopeless) battle against the forces of chaos and confusion (a.k.a. the second law of thermodynamics). Evolution equipped her warrior children with various weapons. The hydra can regenerate. The virus can use other bodies to propagate its patterns. Our weapons of choice have been intellect and institutions. Intellect allows us to identify existing patterns in the environment and use them to our advantage. Institutions allow us to effectively pool our individual agencies to propagate our own patterns across space and time. The main thing that you build institutions from is other humans. And so for other humans to be an effective building material, they need to be predictable and interchangeable. Because what you can predict - you can control. What you can predict - you can rely on and build from. The more reliable a component a human is - the easier it is to use them as a cog, a building block in an organisation (an army, a factory, a company, a church, a country). If someone is trying to build and scale an effective institution, a system that is complex but reliable - they don't want their components to be too weird. The complexity of the system should emerge from its overall design, but its reliability can only come from the simplicity and predictability of the components. At least this is what you believe, if your intuition about building comes from building houses out of clay bricks, or if you conceive the world as a clock.

Imagine building a clock - would you want your cogs to be spiky? Or changing over time? Clearly not. And so, as the clock mechanism has been the metaphor of the world and the dominant model for our institutions for at least three centuries, it's no surprise that the predictability of components has increasingly been the baseline measure of their value. You may object that science departed from the clockwork-like world model over a hundred years ago in favour of the weird world of probabilities. And you would be right. The problem is the quantum world is too strange, too counterintuitive for our culture to adopt. It doesn't match our lego-like macro intuitions. You may object that modern biology and agriculture could also provide a compelling alternative to the "clockwork" metaphor for the builders of societies. For example, bioelectric experiments are showing that the best way to create reliable and resilient systems is not to have predictable and uniform components, but to have components capable of learning and adapting and communicating with each other freely, while the larger blueprints and environmental constraints and incentives provide the overall developmental direction. And yet, even in the face of the overwhelming evidence to the contrary, this all seems too subtle or too early or too far from our everyday intuitions. Narrowly Newtonian worldview persists and most builders of societal systems still think in clockwork terms. And in clockwork land, every single one of your cogs better be exactly the way you expect it to. And if one breaks or changes - you better be able to order a replacement that would slot right in.

To summarize: from the beginnings of history to the present day being weird has been a risky and mostly unprofitable business. In the early stages of our fight against chaos this preference for predictability was necessary. Maximizing weirdness would not be an advisable life strategy for anyone born for most of history, up to the present day. The exact way in which being weird was punished changed over time, the escape hatches available differed. And the mechanical clockwork obsession of the past few centuries has not made being weird any more profitable or rational than before. Life used to favour the average and it continues to favour the average in the present day.

Now on to the good news.

Call me a dreamer, but I don't think the "normal == good" status quo will be the case for much longer. Yes, this is my gospel for the misfits, the weirdos and the out-of-distribution ones, for theirs is the unimaginable future we are about to enter.

My argument at its core is simple:

1. Being normal (or at least suppressing your inner weirdness enough to appear normal to others) used to be a good strategy for most people because being normal == being predictable == being useful in the context of larger structures of society. The value of being predictable and therefore the incentives to be predictable were much higher for the majority of people than the value of being strange. The risk / reward calculus of being strange was not worth it. It only made sense to be strange for those who simply couldn't help being strange, no matter the cost.

2. But with AI (disembodied in software and embodied in robots), the value of being normal and predictable is rapidly evaporating. If you can be predicted, you can be modeled and automated away. You can't out-cog a robot any more than you can outwork one. If your value to others is conditioned on the fact that you can achieve a certain measurable outcome, and do it well and do it over and over, then you are now competing with the price of electricity, which is not a good thing to compete with.

3. In building, maintaining and expanding institutions you need more than predictability. Occasionally you need to leapfrog the current pattern, to escape the local maximum, to unreasonably jump into the abyss or set sail to the west, so that you have a tiny but non-zero chance of discovering a new and better land. So weirdos, the people in charge of the injection of randomness, have always had a role to play. It's just that being one was an unprofitable business.

4. But now that normalcy is being rapidly automated, the relative value of weirdness is shooting upwards. The most valuable thing you can contribute to the world is no longer the reliability of the safe pair of hands but the sensitivity, strangeness, courage and adaptability of the artist. And so, for the first time in history, the rational strategy for the majority of people is to embrace their inner weirdness rather than to suppress it.

Obviously there is nuance. You can take this argument to the limit and say that ultimately the world is made of only two things: Chaos (a.k.a. the vast seas of Wolfram's computational irreducibility, a.k.a. pure noise, a.k.a. quantum randomness) and Logos (the small islands of compressible patterns). And AI will eventually be better at hunting all sorts of patterns in all sorts of spaces, so in the limit - the only two paths for humans are to:

a) accept our fate of being very ineffective and limited pattern hunters for pure sport and joy, or: b) accept our fate of being very energy-inefficient randomness generators

This may indeed be true in the limit. But this limit is far away. And surrendering agency in the face of the ultimate limit is not, as far as I'm concerned, an interesting or productive line of thinking. It's the same thing as giving up all human activity because of the imminent heat death of the universe. Intellectually sensible (why bother fighting the second law of thermodynamics if it always wins in the end?) but pragmatically meaningless (the fact that the game will end and the winner is known doesn't automatically render the process of playing categorically uninteresting or devoid of joy or value).

So let us not worry about the inevitable limit. Death awaits us all, and yet today we are miraculously alive and we have the agency to choose to live interestingly and with gusto. We can focus on the vast and valuable future between today and the ultimate limit. The future that is upon us is the future that belongs to the weird.

What will have to change in the world where a person's weirdness is the key component of their value? Pretty much everything. But for now let me highlight just three foundational elements we will need to rethink as a society:

1. Education. Our systems of education have been built to bang the weirdness out of our younglings - to get them into cog-shapes. To make them uniform and predictable and therefore useful. But now that predictable and uniform is the enemy of usefulness - what would education look like? How do you bring up children in a way that maximizes their in-built strangeness and spikiness without turning them into monsters? This second half is important. Because maximizing spikiness in an unwise way can easily lead to disaster. Weirdness without the adaptability, weirdness without tolerance, without tools to manage it, without a sufficient amount of common ground to make communication and empathy and communion possible - this kind of weirdness is clearly as disastrous as the totalitarian dictate of normalcy. Another danger of the education redesigned to maximize self-determination and spikiness is that it can lead to early explosions of the intellect and the character. Future freedom is clearly dependent on developmental constraints. Vygotsky's scaffolding is necessary. It just should be constructed with the recognition of the individual's potentialities and should be gradually removed so that it doesn't become a crutch or a cage. The path forward for the radically reimagined education is narrow and full of danger. And yet travel it we must, if we care about the future in which our biological descendants are neither robots nor monsters.

2. Community. Spiky people will by definition have less common ground. Which could lead to further worsening of the loneliness epidemic and ultimately to extinction. There are a few ways we can solve for this. We could try to put more effort into bringing similarly spiky people together - better infrastructure and norms for subcultures, interest groups etc. This is what the early internet did for a lot of geeks and weirdos and boy, was it beautiful to finally find "your tribe" for the first time. But restricting community to in-group connections is not enough. As long as we share one planet and one set of resources, more spikiness will need to be counterbalanced by more cross-pollination. We will need to find ways to grow people who are less alike, and at the same time more interested in each other.

3. Economy. For the majority of people, our economy is not very good at rewarding failure. And yet weird people will need to do more weird things. And weird things are less likely to succeed. UBI may or may not be the answer. Academia provides an interesting lesson here. The whole idea of tenured professorship is that guaranteeing employment without the pressure to teach or publish or hunt for grants should encourage scientists to pursue more risky research paths and more long-term projects that may or may not pay off. And yet the evidence on whether this actually works is not as strong as you'd hope for. Simply removing the pressure may not be enough, even for the very selected group of intellectually spiky people. So we will likely need to find ways of incentivising the more risky, blue-sky and long-term human endeavours in the post-normal economy. Otherwise a simple UBI could lead to mostly "dead wood".

The world of the weird is not destined to be wonderful. And the transition to it will undoubtedly be painful, and will likely take a few generations. But a few generations is a very short time in the grand picture of history.

For thousands of years leaning into your weirdness only made sense if you couldn't help it. For thousands of years our societies and economies were only possible if the overwhelming majority of their members suppressed their inner strangeness most of the time. Yet, thanks to AI, the days of dominant normalcy are numbered. We are now entering the times of the strange, where weird-maxxing is set to become the rational strategy for most humans. So I humbly suggest you get ready for the always-on and ever-present Burning Man because, whether you are ready or not, it's coming.

On the off chance you'd like to subscribe to my writing, please feel free to subscribe on twitter or use this RSS feed. Thank you.