Hawkins worries about AI accidentally going nuts and destroying stuff. He ignores that it's an existential threat for the same reason nuclear weapons are: not

Hawkins worries about AI accidentally going nuts and destroying stuff. He ignores that it's an existential threat for the same reason nuclear weapons are: not because of potential accidents, but because of potential on-purposes. Do you really want Bezos / Musk / Biden / Putin / your favorite baddie being the first to get control of these things? It doesn't seem wise.