The AI machines might destroy us, but not out of revenge. It might just be because they intelligently realise that the human race has been the worst thing to happen to the planet. Many of the problems are our fault, so the way to solve all those problems might be to get rid of all humans. Similar to how the best way to be 100% sure to get rid of all spam emails would be to destroy the internet. AI machines might not feel anger towards us, but also wouldn't care to preserve us either, so the decision would be a purely logical one. Which measure is going to be the most effective?