rohinmshah comments on [AN #80]: Why AI risk might be solved without additional intervention from longtermists