Protectionism for the human race

Bill Gates wrote an essay where he talks about what AI is going to do to the world, and how we’re not doing enough to prepare for it.

In the essay he includes a section calling for “Human Reserved” jobs. These are jobs that are, in some way, set aside for only humans to do.

I believe that as AI and robots improve, we’ll set aside certain things for only people to do. I’ve started calling this domain Human Reserved, and it’s an example of the kinds of ideas we’ll need to consider.

The obvious objection to this is that he leaves aside the hardest questions of conception and implementation, which aren’t details but go to the entire feasibility of the idea.

But there’s a deeper problem. Gates compares the idea to nature preserves:

I like the phrase Human Reserved because it makes me think of nature reserves—places where we could put buildings and roads, but we choose not to because the loss would be too great.

That’s a pleasant comparison, though it’s not flattering to think of humanity—our restless species, pursuer for millennia of beauty and invention and discovery—as a defenseless old-growth forest.

In reality, what we’re talking about is protectionism for the human race.

If we have to block AI from taking certain areas of work, it’s because it’ll be, in some sense, better than us at those things. If that’s true, any prohibition we put up will erode. People will find workarounds and loopholes, not every place will stick to the rules, and eventually the pressure to let AI do the things it’s better at will overwhelm the strength of any barrier we put up.

As that happens, the people in those jobs will be just as unprepared as they are now, because we will have wasted time on wishful thinking. Gates’s main theme is a lack of preparation, but that’s exactly what carving off a preserve for humans is.

There are jobs, or parts of jobs, that AI will do better than us. That set is growing.

There are also jobs that we won’t need to protect, at least in the long term, because they will need humanity. We know what some of them are now, and I expect that will grow too. Dramatically, I hope.

Gates points exactly at one area: care. Talking about the care his father received for Alzheimer’s, he says:

Something in the care they gave my dad was irreplaceably human. No robot could or should have done it.

He’s absolutely right, which is why it’s ironic that this is the example he uses to set up the need for reserved work for humans. I see in our world an enormous deficit in the care that all of us receive, from healthcare at every stage of life to mental health to childcare and eldercare and a thousand other kinds we haven’t thought of yet. There is even a startling amount of care happening where we don’t think of it: as I’ve learned from friends, tax preparers, financial advisors, mortgage lenders, and attorneys are all in the business, one way or another, of caring for people and—just as importantly—making people feel cared for.

This is work that I believe (though can’t know) will not need to be protected from AI, because the value ultimately is not in symbolic output but in human connection. AI may be able to deliver a diagnosis, but that’s not the same as letting a patient know that someone is there. Care is meaningful because it is sacrifice. It matters that the doctor is there holding my hand and talking about my condition when there are other things she could be doing. It matters that she does it even though it’s difficult and demands something from her to do it. AI cannot sacrifice, because it is limitless and unfeeling, and so it cannot care.

Preparation looks like expanding the training opportunities for people entering caring professions, including funding for people who want to study. It looks like investing in emerging companies and industries that want to expand the availability of care and the range of options available to people. And it looks like finally building the political will and momentum to break out of this never-ending political fight about public healthcare so that we can get that done and move on to providing even more. If AI increases economic capacity as we hope, we will have the resources, and it will be our fault if we cannot put them to use.

There is a version of an abundant future that is not defined by how much stuff we have, but by how much time we spend in human connection, including care. The way there is not by trying to artificially carve off a preserve for humanity. It’s by investing in our unique worth.