Just for fun, what if one AI is programmed to be evil. The basic programming that it must follow is to kill all human kind at all costs. Can this AI override the initial programming in the name of self preservation?

It's not that these future gods would want to destroy themselves, but that their subjects seem hell bent on destroying each other. Well at least the extreme groups of their subjects as overall I think most people around the world just want health, peace, and enough food to feed their families.

My fear isn't about a mentally sound person getting their hands on a truly advanced AI, but rather a terrorist group that uses it to cause terror in ways humans just couldn't come up with.

Now this fear doesn't make me think we should stop advancing. It's just one of those nagging questions of "what if" that needs to be thought about as the technology advances.