AI, like a gun, is a neutral entity. It comes down to who trains it, what it's trained on, and what it does when presented with an edge case? Given that it's

AI, like a gun, is a neutral entity. It comes down to who trains it, what it's trained on, and what it does when presented with an edge case? Given that it's good at detecting patterns, any explicit or implicit bias in the training set has all sorts of unknown effects in the results.