geoffreymiller comments on AI alignment with humans… but with which humans?

[geoffreymiller]

Thanks Mitchell, that’s helpful.

I think we need a lot more serious thinking about Plan B strategies.