Steven Byrnes comments on [Intro to brain-like-AGI safety] 10. The alignment problem