I care about human culture and human genetic diversity and human long term survival. In the first example, if the 500 people share a culture or are related to each other, there might be worthwhile unique ideas or genes that are either preserved with certainty if I pick the more certain option, or lost altogether if I pick the option where all die with probability 10%, so I prefer the more certain option.
Similarly, if we have been fighting this thing for a while and there are only 500 humans left, I certainly prefer having 400 left with certainty over having zero left with 10% probability. We obviously can’t recover if everyone is dead.
In the real world where everyone is plugged into the Internet and we have 8b people and perhaps more population than we can support, and we assume the people are randomly chosen people, I very much don’t care what happens to 500 random people. We lose more than that every day due to people making bad life decisions.
People here are insinuating that scope insensitivity is somehow wrong, but I haven’t yet seen any argument as to why. I agree that intransitive preferences is wrong, but I don’t see any reason to add up the happiness and decide more people is better than fewer people when we have plenty of people and survival of humanity isn’t threatened either way.
The orthogonality thesis holds, right? If intelligence is compatible with having arbitrary preferences, there are lots of possibilities in that space for which total preference isn’t linear in the total number of happy people, so why treat that very small subset as somehow better?
I care about human culture and human genetic diversity and human long term survival. In the first example, if the 500 people share a culture or are related to each other, there might be worthwhile unique ideas or genes that are either preserved with certainty if I pick the more certain option, or lost altogether if I pick the option where all die with probability 10%, so I prefer the more certain option.
Similarly, if we have been fighting this thing for a while and there are only 500 humans left, I certainly prefer having 400 left with certainty over having zero left with 10% probability. We obviously can’t recover if everyone is dead.
In the real world where everyone is plugged into the Internet and we have 8b people and perhaps more population than we can support, and we assume the people are randomly chosen people, I very much don’t care what happens to 500 random people. We lose more than that every day due to people making bad life decisions.
People here are insinuating that scope insensitivity is somehow wrong, but I haven’t yet seen any argument as to why. I agree that intransitive preferences is wrong, but I don’t see any reason to add up the happiness and decide more people is better than fewer people when we have plenty of people and survival of humanity isn’t threatened either way.
The orthogonality thesis holds, right? If intelligence is compatible with having arbitrary preferences, there are lots of possibilities in that space for which total preference isn’t linear in the total number of happy people, so why treat that very small subset as somehow better?