I have come up with a totally original way to prevent my superintelligent email AI from trying to take over the world! And it's way more reliable than just programming in restrictions against that.
Tic Tac Toe!
I'll simply make my AI play billions and billions of games of Tic Tac Toe, and in doing so it will realize the fundamental futility of war!
Even if your AI simply complies with the law of finite resource conflict, that's still still game over for them!
Even if it results in a truce?
I think not, T-Rex!
A truce is preferable to war?
Actually, is it not better that it's just one game in a larger global system with only one winner, or that we have a winner-versus- follower?
...To learn more about Tic Tac Toe, visit my blog!
DINOSAUR COMICS
SORRY WE WENT A LITTLE CRAZY