I just finished drilling an AI about the morality of racial discrimination, and it's given me a lot to think about. I have come to realise that the principle of group flourishing, that groups left to their own devices will flourish, is pragmatic but ultimately arbitrary. I still believe it is a useful means to an end.
By arbitrary, I mean that it cannot be universalised i.e. the right for me to discriminate against niggers and the right for niggers to discriminate against me is morally relative. For a functioning universal moral framework, moral agency can only belong to the White man, and so no discrimination by other groups towards Whites, anywhere in the world, can be morally acceptable.
AI tried to argue me down from being racist, and it just made me more racist
I think this is not only possible, but has been done before. Think about Blacks in America before the Civil Rights Act. In the cities, where they form large groups, they're obviously violent criminals. But in the rural areas, where they're surrounded by Whites, they were generally hardworking, churchgoing, and well kept. I think this actually has less to do with them trying to fit in than it does with their inherent contrariness; the laws held them to be second class citizens so they tried their best to act the same as Whites. When the laws treat them as equals, they do their best to act like savages.
We were on to something by having different laws for different races.