> Once robots and their AI software are good enough, once self-driving is good enough, it's game over for most manual jobs Why would those getting good enough

> Once robots and their AI software are good enough, once self-driving is good enough, it's game over for most manual jobs Why would those getting good enough happen sooner than people expected? I'd assume those will get good enough at roughly the rate pessimists expected, at least if we look historically that seems like a good bet, these things move really really slowly once you try to apply it to the real world in practice.