It’s possible I’m misunderstanding your comment, so please correct me if I am, but there’s no reason you couldn’t do superbabies at scale even if you care about alignment. In fact, the more capable people we have the better.
I’m having trouble understanding how concretely you think superbabies can lead to significantly improved chance of helping alignment.
Kman may have his own views, but my take is pretty simple; there are a lot of very technically challenging problems in the field of alignment and it seems likely smarter humans would have a much higher chance of solving them.
It’s possible I’m misunderstanding your comment, so please correct me if I am, but there’s no reason you couldn’t do superbabies at scale even if you care about alignment. In fact, the more capable people we have the better.
Kman may have his own views, but my take is pretty simple; there are a lot of very technically challenging problems in the field of alignment and it seems likely smarter humans would have a much higher chance of solving them.