TurnTrout comments on Some conceptual highlights from “Disjunctive Scenarios of Catastrophic AI Risk”