Politico reports that 63% of Americans see at least a moderate risk of AI destroying humanity, and that a plurality favor pausing development of more advanced models. A post sharing the poll gives the pause split as 48% yes, 31% no and 22% unsure, calling it a “mandate” for policymakers. One commentator predicts large, relatively bipartisan majorities concerned about AI risk, though for different reasons. Separately, a critic rejects the narrative that AI will kill every human.