RomanS comments on Three main arguments that AI will save humans and one meta-argument