What are we going to do now that AI will take all our jobs?
In 1930, John Maynard Keynes predicted1 that, in 2030, society would be so productive that we would barely need to work. Presumably, the more technology we create, the more productive we become, the less work we need to do.
Of course, this isn’t what has happened. We do have the most technology ever, yet we work more than ever2. Indeed, we are more productive than ever—and the richest we have ever been—overall—yet inequality is so high that everybody’s experience is that we are poorer than ever.
Similar, from the Chicago school of economics, the theory of supply-side economics3 implies: the more productive businesses are, the more money they will pay their employees. The opposite is what happens in the real world: the more productive a company is (ie. the more money a company makes), the more powerful it becomes, and thus, the more able it is to get away with paying less.
Humanity’s economics are—these days—based on a model of transactions. For example, we expect agriculture workers and companies to cultivate food because they will be rewarded. This reward will be in money, which is a universally accepted representation of (a) wealth and (b) appreciation (ie. receiving money means they do something that matters).
What happens in practice is that people work in a number of different jobs and earn this universally accepted reward which they can exchange with other materials and/or services that they do not have and desire. For example, they can exchange it with the food that agriculture workers produce.
Now, the problem is that if people develop highly efficient AIs, all jobs will gradually disappear because these AIs will execute more efficiently than any human could ever imagine.
In that case, all these people whose skills have become trivial will become poor. They will have no way to find rewards, which means they won’t have money to live. Gradually, all money will converge to the AI people (either because they create AIs, or because they operate them, or because they own the metal that AIs run on). The non-AI people’s future will be dire.
We might feel there is no other way than the transaction model but I think that’s only because we are too deep into it to see something else. If we look around we might find some disturbing absence of transactions, given that market-based rewarding is so important to us.
To start with a simple and almost silly4 example: we never properly rewarded the people who first found out how we can farm the land and produce a ton of food. This we might consider an advancement made not by one individual but by humanity as a whole. Presumably, though, there were some people or groups of people who figured it out first. Were they fairly rewarded, given the humanity-changing impact and cumulative benefit of their work?
I don’t think they did. More examples: we never properly rewarded the first people who figured out how we can sew clothes and shoes and coats and make tools and houses. We never properly rewarded the people who discovered how electricity works or the people who designed our cities and the streets we walk everyday.
This list is truly unending and impossible to complete. People who figured out how the human body works and how to cure diseases, people who built ships and airplanes and an insanely complicated shipping network, people who figured out all technologies required before some other groups were able to put them together into a smartphone or a computer or the internet.
We haven’t even talked about the people who supported these pioneers. People who produced their food and built their houses so they can have time to think of new ideas. And what about the people who supported these inventors—not materially but—mentally? Whether these were partners or religious entities or anyone or any-thing else.
All these were actually gifts.
People like Archimedes, Leonardo Da Vinci, Leonhard Euler, Nikola Tesla and so many more were not motivated by money rewards. They thought and created all they did for reasons that were not transactional. And today all their work is free for us. All their work is a gift.
For this reason, by looking a bit farther into time, I claim that humanity doesn’t operate on a transactional model but on a gift model.
There exists an example in today’s real-world economy that shows part of my argument. Open source software5 is inherently based on an non-transactional gift economy since its inception.
It is partly a factor that programmers are weird people who intrinsically enjoy the process of writing code so much that they do it just because. But beyond that, as a programmer, I can truly say that the fact that somebody makes use of code I’ve written makes me proud. The fact that I made something that helped someone—in addition to the fact that a programmer would rather use my code than theirs—is simply cause for celebration. It’s even more than that: motivation to do more of the same.
It’s the same feeling I have when I gift someone something that they really like.
I really think the open source software movement’s model is the future even though some people think it should be the past.
There are many arguments as to how open source is fundamentally broken because it’s free work with no reward. Of course, people working tirelessly two jobs and people getting burned out is without question terrible. But that’s only because there is no space for leisure time in our lives anymore. Everything is becoming extremely efficient. We can only use whatever time we have awake to achieve and produce—or else we become homeless or losers.
There is no time for experiments anymore, no time for failures, no time for gifts.
But I think that’s the solution. To have time only for experiments, only for failures, only for gifts. To realise the advent of a so-called age of leisure.
The famous quote6 that “it’s easier to imagine an end to the world than an end to capitalism” still echoes in my mind every time I encounter economic dead ends and unsustainabilities.
I really do think there must be a lot of alternatives that current ideologies are too strong to allow us to see. Probably because through our current lenses they appear silly. Or nonsensical, or pointless, or insane, or irresponsible, or utopic, or dystopic.
Whatever the case, it seems we might be forced to choose between the end of capitalism and the end of the world. Maybe we don’t even have to imagine this dilemma because it’s already here. The decision for this choice is humanity’s current challenge. I’m not saying we’ll figure it out—it might even destroy us. That’s why it’s a challenge. It’s a riddle: can we imagine beyond our collective bubble?—and riddles are things that beg for solutions.
There are two perspectives here.
Perspective one is: humanity is a bunch of units who compete with each other over time, money, resources, et alia. Units who have power over AIs are more capable of enslaving units who don’t. If the units with AI powers are peaceful they can leave the rest to live without AIs. Or maybe units with AI would rather take all resources for themselves and leave nothing to the rest. Whatever the case, it seems likely that a society whose imaginary7 is based on competition will engage in some kind of violent resolution for such power imbalance.
Perspective two is: humanity is a bunch of units who cooperate with each other to maximize wellness for everyone, without keeping ledgers as to who did what. No ledgers is the essence of lack of transactions. Maybe they can all coordinate to enable AIs to build whatever all units want without zero-sum systems and power over others.
We are definitely—currently—seeing the world through the first perspective. It seems to me that unless we consciously decide to change to the second kind, we’re highly likely to fight with each other. I’m pretty sure we can change. How can we change? That I will not reveal. Everybody can8 figure it out for themselves. I’m not implying there is a different subjective response for each person. There is one specific solution in my mind that we can all arrive at. I just don’t want to give it away. But I can give away a hint.
Addendum: I’m writing a book on democracy, which explores ideas like this one. If interested, read more here.
In his text Economic Possibilities for our Grandchildren. ↩
Living in humanity’s most work-intensive era can be a controversial statement. Before dismissing it let’s consider that a lot of people work today because they want to and not because they have to. The number of people who engage in such a pursuit of achievement, I feel, is more than ever. ↩
There is another term: trickle-down economics, popularised more recently, it refers to policies favouring the wealthy bracket of the population with the hope that their wealth will trickle down to the less wealthy. ↩
Silly ideas can be very powerful. It used to be very silly to want a computer in your house. To want it in your pocket wasn’t even silly, though, because it was simply unimaginable. ↩
I’ve written more about the open source movement and what impact it can have across society in Open source movement as societal theory ↩
Mark Fisher mentions this quotation in his book Capitalist Realism, which he attributes to both Fredric Jameson and Slavoj Žižek. Interestingly, Mark Fisher hanged himself as a result of his struggling with depression. He said: “the pandemic of mental anguish that afflicts our time cannot be properly understood, or healed, if viewed as a private problem suffered by damaged individuals”. Extreme productivity and efficiency, as part of capitalism, not only can lead to destruction through (possibly inevitable) highy advanced/powerful technology but also through the absense of space for doing otherwise (ie. doing inefficient actions) which leads to depression (exhaustion). ↩
An imaginary of a society is a set of core, underlying values and beliefs which constitute its foundation. They define how people perceive what’s worth doing and what matters in the society. Wikipedia says an imaginary is “the set of values, institutions, laws, and symbols through which people imagine their social whole”. ↩
Can, here, refers to ability, not possibility. ↩