[diegoperini]

That kills the whole purpose of automation. Automation is preferable because it is supposed to be cheaper while providing equal benefits as traditional production of goods and services.

My personal choice here would be a tax for practices that are provably automatable but not yet automated. The result is the same, we are funding a UBI but now businesses are also incentivized for innovation to escape from the tax. I'm probably missing hundreds of reasonable concerns with my simplistic view point though.

[kelnos]

> That kills the whole purpose of automation. Automation is preferable because it is supposed to be cheaper while providing equal benefits as traditional production of goods and services.

That's the capitalist enrich-the-owners purpose. In my mind, the real purpose of automation is to relieve humans the need to do work so they can live lives of leisure and personal enrichment. Unfortunately, I don't expect us to get there within my lifetime, if ever at all.

[thaumasiotes]

> My personal choice here would be a tax for practices that are provably automatable but not yet automated.

If a practice is provably automatable, then it's already automated. That's what proof looks like.

[nefitty]

Well, some companies seem to end up with weird incentives where it makes sense to be less productive. For example, a dev might create some script that eliminates some data processing roles. The CEO might decide to not implement it solely to save a few jobs. If there’s a tax for not automating then the organization will immediately respond in a more economically rational way.

[diegoperini]

Existence of a proof doesn't mean application of it is widespread.

[thaumasiotes]

In this case, the existence of a proof means that its application is universal. You may believe that the automation of one process could be easily generalized to another process, but you haven't proved it until you've automated the second process.