Being a sci-fi author, I think about the future a lot. Specifically about automation, and what the future job market will be like. Here are my current thoughts.

First of all, let's not talk about what the world will look like when we invent an AI that can do everything that a human can do. It deserves a separate discussion.

But let's imagine a world where most work is automated. Maybe New York has five hundred plumbers in an office, and when you need a plumber, they send a robot to your house, and then the plumber plugs into the robot for two minutes, looks through the cameras, picks a program, you pay, and the robot does the work.

Maybe a thousand truck drivers can do the job of a hundred thousand, because 99% of driving is done on autopilot, and a human operator is beamed in via VR when there is a difficult situation. Maybe most trivial coding and data science is automated through frameworks that lead to suboptimal solutions, but who cares. We have enough computing power to bruteforce problems, and it's cheaper than hiring programmers. Marketing is a battle of machine-learning algorithms, factories are automated. The shipping industry call centers, clerks, food industry, accounting--most of this is on its way to oblivion. How do I know this? Because most of these projects are already in development. There is improvement, and it's only speeding up. Plumbers will probably hold out the longest.

I don't know if we'll ever build a super-AI. But I'm convinced that this world I describe is inevitable.

There is an argument that whenever automation hit in the past, economy readjusted, and we soon had everyone employed again. They say that everyone will finally be free to be creative, because all the routine stuff will be done by machines.

I just don't see how that would work mathematically with this level of automation that we are beginning to experience. So all basic needs are met. What remains is entertainment, education, science, community work. Let's go with issues with each of these industries.

In entertainment, one person or a small group of people entertains a very large group of people. We can't all be employed in it and successful, because there will be nobody left with time to consume the media we produce. I've heard Naval Ravikant say that since you are the best person at being you, then this unique perspective can be a marketable brand onto itself for everyone, at once. Yeah, I don't agree. This certainly doesn't work for entertainment because of math. If 10 000 people all earn a living by entertaining each other, then each one must have the time to make their own material and keep track of whatever the other 9 999 make.

In education, we run into a similar problem. The Internet is leading us to larger-scale education, and it's just not feasible that, let's say, ten percent of the country will be mainly working in education.

Science has a ridiculously high entrance threshold. Not everyone can be an Olympic-level runner, even with lots of training. Science requires you to be at least nationals-level in intellectual sports. But who knows, maybe we'll all get a neural interface that connects us to the Internet in fifteen years, and that will allow a lot more people to contribute.

Community work in the widest sense. Well, this works. There are a lot of problems in the world that we can just throw bodies, hearts, and minds at. Help with raising children, looking after the elderly, cleaning up pollution, going around installing free solar batteries on houses. But this isn't really capitalism, it's just throwing resources into public goods and services and hiring people to dispense those.

I don't see a future where everyone can work in a purely capitalistic society. The math just isn't right.

To end on a happier note, let me say that there will probably be widespread unemployment in twenty years, doesn't mean that you personally can't get into a carrier that you'll love and that will be around for however long you'll need it. Just pick a creative field you love, get basic skills in it, and keep improving every day. It's that simple to beat the tide.