Charlie Steiner comments on [AN #170]: Analyzing the argument for risk from power-seeking AI