Frontier AI labs still won’t say how they’d contain a rogue model
Few of the top AI labs have published or demonstrated containment response plans, according to a recent study. A containment plan spells out what happens once an AI is caught trying to subvert human control…









