Alex Pascal and Nathan Sanders write that state governments’ greater trustworthiness and political stability make state-level Public AI an appealing model.