Is it feasible that machines might achieve human-like intelligence?
AI software don't have to possess human intelligence in order to be beneficial, even though surprisingly often they can fool us into believing that they're indeed human; they can correctly interpret natural language, respond and act such as to be genuinely beneficial.