fp.

He is so blantly not a conservative. So with a 41% (tho falling) approval rating, the bigger question is: did those who claimed to believe in it really give up their principles so easily? Or never really believe in them to begin with?

delichon•2mo ago

In this case, does federalism give us fifty laboratories of democracy, where dozens of variations of AI regulation can be tested and compared? Or, since it so hard to identify locations, and to support fragmented models, do we get a de facto national regulatory scheme combining the most restrictive laws from each state?

treetalker•2mo ago

just like we learned in law school, executive field preemption

unangst•2mo ago

Do not go gentle into that AI dawn.

jonahbenton•2mo ago

Doubly incorrect. There is no power by which the executive can "ban" any state action. Is not a thing. But that isn't what the piece says. The EO would set up a "task force" to sue states that did X or Y. That is something that happens, and when the theories are at the level of quality this admin produces, those suits fail.

pfannkuchen•2mo ago

You seem to be overlooking the commerce clause and Wickard v Filburn letting the Feds control the states however they like (so long as there is a butterfly effect argument about interstate commerce impact, which here there easily is).

polski-g•2mo ago

This is good. Dealing with 50 different laws on AI regulation would be a nightmare for a startup. Not to mention that most AI regulation would be unconstitutional per Bernstein v DOJ. (The PGP case)