Oh is this being studied, the ramifications of AI reflecting and amplifying some of the negatives of humanity back at us? Today’s training sets might not know w

Oh is this being studied, the ramifications of AI reflecting and amplifying some of the negatives of humanity back at us? Today’s training sets might not know when something was created, and don’t see attitudes changing. Today’s AI has no concept of things it has seen examples of but should not repeat. Maybe we’ll get AI that will convincingly argue conspiracy theories. AI will never get tired of arguing. What if neural networks become better than people at spreading misinformation?