> We don't see amateur engineers, doctors, lawyers... Not yet! > amateurization is correlated with benign consequences for being wrong Not that simple I thin

> We don't see amateur engineers, doctors, lawyers... Not yet! > amateurization is correlated with benign consequences for being wrong Not that simple I think. it's correlated with either (1) as you say, not so bad consequences for being wrong, (2) being able to quickly check/measure both an individual's actual skill and the possible consequence of their decision before implementing thee, or (3) "rollback"/undo-ability - if something wrong happens, there's a relatively cheap way to undo the consequences and try again a different thing. You can't easily measure skills in medicine or law (you can't create simple and fast tests/exams yet for such areas). You can't really simulate either by running a mental experiment or by running some software the consequences of decisions in medicine or law. You can try but it will fail horribly most of the time. And there's no "undo" after someone dies or stays wrongly in jail for X years. I think that, paradoxically, as fields get more technical and precise, they will become more open to amateurization, because skills will be easily assessable and outcomes predicatable before deploying a scenario in the real world. Of course, this will happen concurrently with AI/ML-replacing humans in these areas, so the amateurization if it will ever have time to happen will probably be a footnote in history, dwarfed by the bigger story of mostly getting humans out-of-the-loop... But the humans that will remain in the loop will likely be more and more expert-generalists, eg. the "platinum level" version of "amateurs"!