What’s the next “oops, fluoride probably doesn’t belong as an additive to drinking water “? And why are we not using ML to investigate every single published paper for flaws?