If #1 disappears, you will not continue wanting to do #2 for itself.
Yet, English uses the word "want" for both – leads to confusion in discussions about AI.
Take a thought experiment, where there's a species with a sex ratio of 10/90 for some reason – maybe some researchers edited the genes – that is, 10% of the time a sperm meets an egg, the result is male, and 90% of the time, the result is female.
Will it stay on 10/90 in the long run?
Well, what happens for an individual who makes a bit more males than females? That individual's genes will be more successful. It's the same old story. The only stable state is 50/50, which is where the ratio ends up after some number of generations. But why 50/50, not, say, 60% males 40% females?
I want examples. I think some species make a skewed adult sex ratio by eating some of their young? Or?
Rene Descartes (1596–1650) thought that when we're told a proposition, we would first comprehend what it meant, then consider it, and finally accept or reject it.
Baruch Spinoza (1632–1677) disagreed: he thought that we first passively accept a proposition in the course of comprehending it, and only afterward actively disbelieve it if upon consideration we reject it.
Descartes' suggestion sounded more logical and intuitive, so that's what most philosophers have assumed true, but as of the 21st century the evidence is in: Spinoza was right.
Gilbert, Tafarodi, Malone paper: You Can't Not Believe Everything You Read. See www.greaterwrong.com/posts/TiDGXt3WrQwtCdDj3/do-we-believe-everything-we-re-told?hide-nav-bars=true
A takeaway: you may believe something outrageous if you're told it while distracted with some other cognitive work, so that your normal filters never apply.
l