The “supreme intelligence” concerns about AI live on one end of a very wide spectrum of concerns. I don’t think we need anything approaching super intelligence

The “supreme intelligence” concerns about AI live on one end of a very wide spectrum of concerns. I don’t think we need anything approaching super intelligence for a large subset of concerns to become reality, and rest of that spectrum is filled with non-theoretical risks, many of which are based on capabilities already in production.