rk comments on AI development incentive gradients are not uniformly terrible

[rk]

I split out the comments into areas of concern over on lesswrong. I think it would be a bit too noisy to duplicate that over here, but do feel free to bring up any issues!