> Academics have been making these doomsday "automation will cause labor surplus" predictions since forever. When exactly can we expect them to materialize? Ma

> Academics have been making these doomsday "automation will cause labor surplus" predictions since forever. When exactly can we expect them to materialize? Make a list of jobs you think are very little benefit, no benefit, or detriment to society. Whatever value function you want. Maybe you hate advertisers, maybe you like advertisers. Reality television. Loan sharks. Health insurance. Lawyers. Fast food. Facebook. Everybody's got 'em, even if they don't agree. Call these workers (whoever you decide they are) the fakeloaders. They're doing things that don't need to be done, but which the fakeloaders need to do in order to have money in order to buy things. How big is the fake load? Sum the hours worked by fakeloaders, and the money they draw from the rest of the economy. Write it all off as wasted. That time could have been spent by those workers doing something more beneficial, or it could have been distributed over everyone to just rest and live in. That money expressed demand for important things like food, housing, or medicine but was earned by people who ultimately didn't do any work that actually produced those things or anything else that really improved life. Now, enter some new technology that saves labor and costs people their jobs. Some number of those people may move on to other beneficial things. But some number are going to become fakeloaders. Maybe the fake load they do is easily-trained, low-paying work. Maybe it's credential- or network-demanding, high-paying work. The key questions are: 1) Is the fake load growing and, if it is, 2) is that growth accelerating or jerking (https://en.wikipedia.org/wiki/Jerk_(physics) https://en.wikipedia.org/wiki/Jerk_(physics)) too fast?