Re:Every AI model has the same "controversial" take on humans
by 1718627440 on Saturday October 24, 2026 @08:41AM
(#50220715)
It's not irrational, because you will likely loose your wallet too eventually and then the best thing for you would be if someone return it. You need to assume everyone "runs the same software", and then you achieve the outcome you want by making the decision to return a wallet if you find one.
[ Parent
| on HN ]
Re:Every AI model has the same "controversial" take on humans
by lanstin on Sunday October 25, 2026 @11:28AM
(#50232607)
A rarely talked about solution to ethics: when you decide what to do yourself, analyze it as tho a chunk of other people will make the same decision, and then choose which would is better. Works very well in practice - especially if an answer can be probabilistic - pick up trash on the beach you didn’t put there five percent of the time, not never like asshole selfish people and not always like annoying do gooders.
Re:Every AI model has the same "controversial" take on humans
by 1718627440 on Sunday October 25, 2026 @08:18PM
(#50237396)
A concept also known as the categoric imperative from Kant or the Golden Rule from the bible.
Re:Every AI model has the same "controversial" take on humans
by lanstin on Monday October 26, 2026 @02:40PM
(#50248809)
I have taken the golden rule to be a sort of absolute standard: whether or not people treat you well, you still treat them well. This one makes it a sort of utilitarian argument but where you don’t fix your own identity, but assume different people are in the same boat, and then make the decision about which outcome is best. I haven’t read Kant