" He acknowledged that "there is always some risk that in in fact wanting to advance (friendly) AI we may possibly build the issue we're worried about"; but Even so, that the most effective protection was "to empower as Lots of people as is possible to own AI. If Absolutely e