Shmi comments on Thinking soberly about the context and consequences of Friendly AI