Exactly. Prior to ~1920, horses had lots of jobs. After that, not so much:
https://www.researchgate.net/profile/Terance-Rephann/publica...
(of those that do have jobs now, most are employed on a nonprofit basis for companionship and/or pasture decoration)
Moravec's paradox [1] explains that AI can perform the "complex" processes and cannot execute the "simple" processes of children.
y recognizing a face
y moving around in space
y judging people's motivations
n catching a ball[1]
y recognizing a voice
n setting appropriate goals
y paying attention to things that are interesting
y perception
y attention
y visualization
y motor skills
y social skill[2]
[0] https://en.wikipedia.org/wiki/Remote_manipulator[1] horses can enjoy matching velocity vectors with a struck ball, but are not into catching as dogs[3] are
[2] at least those sufficient for a feudalistic society, anyway...
[3] recall: Warren Teitelman eventually gave up on computer DWIM in favour of agility dogs
As someone who grew up in Latin America in the 80s and 90s: amen!
My country is still picking up the pieces, and some of the damage will persist for a very long time, especially in the areas of health and education - which means even if it does get fixed, the effects will reverberate for at least another generation or two.
Job market like lots of other things is complex and not linear, and has direct dependency on macro conditions. AI might reduce the need for lots of roles, but opens up a lot of new opportunities that are not conceivable by even almighty IMF researchers today.
It’s perhaps analogous to “cars will remove the need to lots of jobs” back in 1886.