Steven Byrnes comments on Is the argument that AI is an xrisk valid?