What jobs will never be taken by AI?

[RubyRubyRubyRubay]

So, my son is about to take his GCSE options and although I've been getting him to think of a trade he might like, he really wants to go to uni.

I can think of lots of trades that will be difficult for AI to totally take over:
painting
plumbing
roofing
building
cooking
sports coach

. . . but what about jobs you need a degree for?

I'm struggling to think of any that AI won't eventually decimate.

Any ideas out there?

It's astounding leap in just 30'odd years isn't it!

Its like the romans just went steady with tech, then suddenly it was lightspeed so to speak.

This is really worth a watch OP

The question isn't what jobs will survive AI, as most will change in some way/be eradicated, but which jobs we want to have a person do. Or the government thinks worthwhile for a human to do, which is not the same thing at all.

Teaching will survive AI, indeed the private school sector will boast they have real live humans teaching the children.

AI is the next, and arguably the most significant, Industrial Revolution. I don’t understand why our Government haven’t created a dedicated Minister for AI position yet. There should probably be an entire department dedicated to it!

That said, AI does not equal something to be feared. Yes things will change, just like the invention of steam power forever changed the way we did things.

Another really good listen is this episode of The Infinite Monkey Cage: https://www.bbc.co.uk/sounds/play/p0gr7lx0?partner=uk.co.bbc&origin=share-mobile

This is what we need
"Samaritan is a artificially intelligent mass surveillance computer system created by Arthur Claypool for the United States government in a project similar to the Machine. Samaritan's focus lies less in the identification of relevant and irrelevant as "The Machine" does, but more so on its artificial intelligence capabilities."

I'm old enough to remember when we used radioactive labelling to sequence DNA and being able to pour a sequencing gel was an important skill you had to learn to do a PhD in biochemistry. And we read the gels ourself. That was in my PhD in the mid 90s. In my postdoc (late 90s) we used fluorescent labelling and computers could read the sequence but we had to check because it could make mistakes. Now we sent away samples to specialist labs and I have no idea how they do it anymore, presumably it's all high throughput and completely automated.