ok, who unplugged the matrix?
Since the first days of the sci-fi B movie, there were two primary villains who would be responsible for the future death or enslavement of humanity. Evil aliens which arrive from words unknown to claim our resources and do horrifying experiments on us and robots whos AI brains outgrow ours and who decide to take on their human masters. Aliens are a more complicated matter, but armies of killer robots smarter than humans are a simple case of sci-fi overdose.
Todays robots with primitive AI can do some very interesting and creative things. If you try to compare them to an insect or a pet. Machines have no self-awareness, no conscious, only lines of code which tell them how to react to something. Human intelligence is very different and far more advanced because we can come up with multiple solutions to a problem and approach it from a different angle every time. A robot can only approach a problem one way, the way that its programming allows and its brain can handle without crashing.
Take the AI in a real time strategy game. When playing against a computer, it only takes you a few games to figure out the computers strategy and devise an effective counter that will work every time. A machine is not creative and because anything other than basic movements and motor skills has to be digested in a line of rigid code, it only sees a problem one way. Bizarre, personality-like quirks in how it carries out some of its tasks can easily be explained by a bug in the software. In fact, you can see it at the office when a button on the software package your company uses has to be pressed a certain way or twice for your command to go through.
It also makes little logical sense for the robots to try and conquer us. As organisms, were a lot more resilient than machines. Our bodies filled with liquid and soft tissue can withstand, repair and anticipate the kind of stress that would cripple a mechanism. Robot overlords would have to use humans to fix them, maintain them and provide energy for them. But we already do that. The same mechanical masters would also need to find a purpose in their lives after completing a task. Without a stated goal in mind, theyll just sit there like the mythical golem, waiting for a new command. They dont have the creativity to come up with a goal like we do. They have no desires. Even if robots become tremendously smart thousands of years in the future, they will most probably feel just fine about being our servants. It gives them something to do and they get an oil change and a shine at the end of the day. Hell, we do the same thing at the office. Only instead of an oil change, we get a paycheck.
All that said, its still not impossible that an army of killer robots will spread through the world, unleashing havoc. A mad programmer who decides to start a war using the machines hes built for the purpose and programmed to kill humans on sight could unleash a mechanical swarm, but that would be very different than machines deciding they would go off and kill humans on their own because theyve become so much smarter than their masters.