Joseph Miller comments on Against Yudkowsky’s evolution analogy for AI x-risk [unfinished]