3 Challenges for Humanity in the Exponential Age

We seem to be living in an age of accelerating exponential change much in line with Ray Kurzweil and the Singularitarians claim. Elon Musk and others are worried about superintelligence, several billionaires including Jeff Bezos and Bill Gates have put together a 1 billion dollar fund to change energy as we know it, Softbank has raised a 100 billion dollar fund to capitalize on exponential technology and Bill Gates has transitioned from ending malaria to taxes for robots. The world around us is indeed shifting rapidly.

This shift makes us have to change completely our world views and our list of key problems. This has happened with every major transition. The biggest problem in London before the advent of the automobile was horse manure. Agricultural age problems became quaint and meaningless in the industrial age. Whatever hunter-gatherers used to worry about before the neolithic suddenly became absolutely irrelevant.

We have our own favorite set of 20th Century worries: global warming and ecology in general, cancer, work-life balance, poverty in the developing world, the threat of nuclear war, and many others. What if all of this problems are quickly becoming irrelevant and will be solved by themselves? 

When you look at trends in renewable energy and storage, biotechnology, AI, additive manufacturing and nanotech it does indeed seem plausible that all of our twentieth-century problems will become obsolete over the next couple of decades. This might sound shocking and absurd, but there is no need to debate we can just watch history unfold short term. The real question is: What should we be worried about then?

My thesis is that there are three main life or death problems for humanity in the next 20-50 years. These two first problems will come one after another as bridges to cross, the third one has been and will continue with us forever:

• Bridge 1: Transitioning to abundance. How do we transform our economic and social system to transition from capitalism to abundance? Here we need to avoid a permanent two-tier society (Elysium scenario), civilization destroying revolution and conflict (Mad Max scenario) and technological apocalypse (World War Z scenario). 
• Bridge 2: Passing the torch. How do we let machine-based organisms (which might or might not contain our consciousnesses) take over our role at the apex of the complexity pyramid? Here we need to avoid triggering either a Luddite revolt (Butlerian Jihad scenario) or an AI Sovereign that exterminates humans (Terminator scenario) or enslaves them (Matrix scenario).
• Meaning. How do we keep humans engaged and productive? We will not be at the apex anymore, robots will do most things for us and evolution will have peaked creating the functional equivalent of God in our solar system. The dangers here are virtual escapism into the simulation (Wall-E scenario) or losing the will to live (Childhood's End scenario).

Most of our potential problems have been imagined by some of the best science fiction writers out there. The scenario names are just a short list, those scenarios have been replayed countless times in the last 50 years of sci-fi books and movies. Imagination surely still fails us and we will imagine new problems as the next decades unfold.

If the problems and the scenario names make sense for you, no need to read any further. The article is longer than usual. Otherwise, you can find a more detailed explanation of each of the problems in the rest of the article.

Transitioning to abundance (2015-2030)

Even if the capitalist system for organizing human action has its flaws, it has served us well to advance human prosperity across the world. All indicators are moving in the right direction and the amount of progress since the start of the industrial revolution has been staggering. However, we have lost sight that money and markets are just useful shared fictions to organize human action. As the context changes, these shared fictions stop being useful. If you read Mises's monumental Human Action, you realize that we have confused economics (or praxiology as Mises called it) with the study of capitalism. Capitalism is only an economic system that has proven very useful while labor and capital were the bottleneck resources. 

Exponential technology is making capital infinite through abundance and is making labor infinite through robots and AI. These two changes make money, jobs, and marketplaces as we know them obsolete. We cannot longer use a system prepared to optimize for two bottlenecks when those bottlenecks are removed. If we do, we run the risk of giving disproportionate economic control to those who own capital while needlessly pauperizing everyone else. Price signals lose meaning, and thus the economy loses efficiency.

This is not only unfair but also extremely dangerous. It leads to polarization with an incredibly rich and small super class and a broad pauperized serf class lacking opportunity. This has happened often throughout history (Roman oligarchs, Russian nobles, French nobles, Robber Barons, Russian oligarchs etc..) always with similar consequences. There was plenty of human suffering and lost potential while the top classes managed to keep their privileged position. Then there was some sort of explosion that used physical violence (e.g. French revolution, Russian revolution, Chinese revolution, etc...) or democratic institutions (e.g. New Deal) to shift to a new system.

In this round of the wheel of history, polarization is already happening at an accelerated rate in developed countries over the last 30 years as documented by Thomas Piketty in Capital in the 21st Century. Additionally, for the first time in history violence doesn't require large numbers of humans and can be mechanized almost wholly. At the same time, human labor won't be particularly needed. So the combination of robot armies and nuclear weapons could easily end up with the rich winning (Elysium or even worse extermination of the poor), everyone losing and being reset to a pre-technological state (Mad Max) or everyone losing to an uncontrollable nuclear, biological or robotic threat (World War Z, grey goo, nuclear armageddon, etc...).

Now that labor and capital are infinite it is not clear what is the bottleneck. Some commentators consider it is human attention, others talk about ideas and creativity. So it is not clear what the solution should be. There are many ideas out there like empowering entrepreneurship and the Missing Middle, Universal Basic Income, Robot Taxes, the Digital Dividend and anti-hoarding mechanisms. The only way out is a Global New Deal in which the economic system is restructured, inequality is reversed, opportunity for everyone is created and abundance spread.

Passing the torch (2025-2040)

Human society is predicated on humans being the apex predators and the apex everything. During the last century, we have gotten used to it, as we tamed nature in all its manifestations. However, this is going to change. It is very clear that our biological limitations around neural processing speed will allow other substrates to surpass us. As vacuum tubes were substituted by transistors, human brains will be substituted by processing on artificial substrates. 

There is still a long way to go, machines already win at Go and Chess, and will have equivalent processing capacity in the short term. They are still much less energy efficient so we still have some time to go until we are totally obsolete. However, exponential progress in computing makes it a matter of time. As good and ingenious as natural selection's processing architecture is, our 120Hz brain can't compete with 2GHz server farms.

This doesn't mean that we might not be somehow part of what comes. Some humans might transfer their consciousness to artificial substrates or human brains might be a part of whatever system that emerges, much like the reptilian brain is even part of our overall brain. In any case, humans won't be the most complex thing in the universe anymore. We will have passed the torch to something else.

The trick is how do we pass the torch with dignity and safety:

• With dignity because we might follow Ned Ludd's steps or Frank Herbert's imagination and try to destroy our descendants in a Butlerian Jihad. Trying to freeze development of AI is one of the recurring themes that great science fiction writers like Frank Herbert or Isaac Asimov have employed to create a comprehensible future. The consequences of actually trying to do it in the real world are scary, but this will probably be a viable political option that is considered in Bridge 2.
• With safety because, as Nick Bostrom covers in Superintelligence, the Sovereign AI we engender might be inclined to be nasty to us and will be beyond our control. On the one hand, it might decide we are useless and sweep us away like ants in a Terminator scenario. On the other hand, it might decide we are useful and enslave for some obscure purpose in a Matrix scenario. Finally, somehow the sovereign might end up with some really stupid measure of success, like maximizing human happiness and store trillions of humans in self-contained vats while constantly hyperstimulating our happiness centers. This would be a Hitchhiker's Guide to the Galaxy scenario for me.

How to manage this still requires a lot of thought and thankfully we have some time. For my perspective, we will need to follow three principles: give meaning to our passing of the torch, give integrity and morals to our creation and find them a different place to live so we don't interfere. Imagine that we are indeed contributing to the creation of the Functional Equivalent of God at the end of the infinite series of intelligence transitions. We will be passing the torch to something better (Angels? Devas?), we would teach them by the example of our selflessness and we would set them up somewhere outside Earth (e.g. the asteroid belt) where they can find a physical substrate and solar energy to evolve. Hopefully, they will leave Earth and the other planets as a kind of primitive reservation to celebrate their origins.

A poetic justice ending to an Elysium scenario to Bridge 1 is that those lucky few ultra rich would probably end up either punished or exterminated by their AI descendants. If the Sovereign AI was perfectly moral it would punish them, if it was immoral it would follow their example.

Meaning (always)

If we are skilled and lucky enough to get through bridges 1 and 2 we will have to deal with problem 3. Actually, we have been dealing with problem 3 since consciousness appeared, so maybe we should call it problem 0. Why are we here? Why should we act? What should be our goals? What is the meaning of life? These are eternal questions humans have been grappling with for as long as we have records of human thought.

However, the problem of meaning acquires a new even more urgent quality when humans have perfect physical abundance, may have created the functional equivalent of God and you are no longer at the apex of complexity. Basically, you've finished the game and are now mostly irrelevant. It is a great place to be from the perspective of human need and physical comfort. However, it might prove challenging psychologically.

The human brain is wired to find meaning. Religious meaning, meaning in adversity, meaning in accomplishment... All these types of meaning might be denied to us. There would be no adversity anymore, accomplishment and religion might seem meaningless when compared to more complex intelligences.

Two plausible scenarios could the simulation or oblivion:

• The simulation means getting lost in a virtual reality that we gamify to satisfy our mental urges. Perennial World of Warcraft. There is even a hypothesis that posits that we are indeed living in the simulation already. The Indian concept of Maya would also be compatible with this position as it claims reality is an illusion. It is depicted in a funny way in the film Wall-E in which fat humans fly around in their VR chairs while a lone robot tries to clean up Earth.
• Oblivion is losing the will to live and gradually disappearing. No more children and people committing suicide as they get bored. This is exactly what happens in Arthur C. Clarke's Childhood's End. Severe depression at species level due to lack of meaning.

Meaning is difficult to solve even today. At a tactical level, we need to understand the human mind better to see what drives us. We need to gamify our day to day to make it more appealing to our brain as proposed in Reality is Broken. There is a lot we can do even today as depression is ever more prevalent in developed societies and work is increasingly meaningless. At an existential level, we face the ultimate question. What should our goal be? Filling the Hubble volume with computronium to be able to compute as much as possible as proposed in the Singularity is Near seems very bleak to me. I am sure we can find something better to do. Maybe helping other worlds get through their own superintelligence transitions, we could sure use some help on our own.

What should I do now

A kind of Matrix pill question comes to mind: Do you want to change your worldview or are you happily living in the post-industrial world? This is not a rhetorical question. The above reasoning might be flawed and we could continue in a more linear change trajectory, with the problems stated above being relevant in 100 years instead of 10. Even assuming all of the above reasoning is correct the blue pill (going back to normal) might be the right answer for most people while the red pill (shifting worldviews) might be a recipe for unhappiness. After all, the world is going to take some time to change and we have to continue to operate in a system governed by the old rules. 90% blue and 10% red might be the right mix for most.

The 10% red will help you start paying attention to the developments that point in this direction (e.g. income inequality, AI arms-race with GAFA/BAT, etc...), open your mind to new types of solutions (e.g. universal income, robot taxes, digital dividends, entrepreneurship for everyone) and maybe even vote for them. Additionally, the main adjustment I would recommend is to increase your discount rate substantially. Sacrificing now for something that happens 10 years down the road might be a losing proposition, 10 years down the road might be wildly different from today. So carpe diem and enjoy life. After all, people in their deathbed wish they had done more of that anyway.

Some might find they are in a position to take the red pill (change worldview completely), as they have managed to place themselves beyond the day to day demands of the current system. For those bridge 1 seems the right initial focus as it is the one that brings most danger short term. Help reduce inequality both at the micro and macro level. Give opportunities to those who have less, support entrepreneurship and the "missing middle" everywhere, support politically and otherwise those actions that might make a difference (e.g. taxes on capital, donating fortunes, universal incomes, helicoptering money, taxes on robots,...). We will all appreciate your work. It is good to have an even smaller minority focused on bridge 2, the Asilomar conference on AI and space exploration with Space X and Blue Origin are great ways to start preparing for AI.

Finally, you can always search for meaning. This is our day to day as humans. We are meaning-seeking animals and the Oracle of Delphos motto, know thyself, always applies.