Incorrect comments on Trapping AIs via utility indifference