Could Ai destroy humanity? Only in certain scenarios - and the RAND Corporation has been wargaming these scenarios.Do we need to ask - should we fear Ai - or, perhaps, humans who play "games" like this?