Why? A robot should be able to disprove free will by following the rules of correct reasoning, if such a thing is relatively easy, i.e. you would expect humans to ever solve it.