The practical downside of fearing superintelligent AI is that we might slow down tools that could fix real problems, like disease or climate stuff. I get the worry, but it feels too hypothetical, and honestly I am not sure how to argue it better. Fear might cost us more than it saves.