I'm Preston Monk, a PhD economist. Applied Punk is where I think in public about the economics of AI and the increasingly strange question of what it means to be human.

One question sits underneath everything here: is this time different?

In 2016, Geoffrey Hinton, one of the godfathers of AI, said we should stop training radiologists, because within five years machines would read scans better than they could. The machines got good at reading scans. There are more radiologists today, not fewer. Two hundred years earlier, the power loom arrived, and the handloom weavers really were destroyed. Same experiment. Opposite endings.

The difference is not luck. A job is not one thing. It is a bundle of tasks. Machines take over some of the tasks, almost never all of them. What happens next depends on the tasks left over and the person still holding them. The radiologist's bundle got stronger. The weaver's bundle was the machine's whole target. Working out which ending is coming this time is the project.

We have been running this experiment since the Industrial Revolution, with a new machine each time. Every generation was sure its machine would replace everyone. Every generation was wrong about everyone, and right about someone. Real people lose real livelihoods. Some end up somewhere better. Others never quite find their footing again.

I don't know where this leads. I invite you to follow along. Subscribing is free, and posts land in your inbox.