Artificial Intelligence and the Safety Debate

Futurists were optimistic about getting powerful AI, but none of them really seriously talked about the safety issues. The problem right now seems to be that the field is a bit divided between AI ethicists and those who are working on everyday ethical issues such as bias in algorithms. It's pretty obvious that if you can make an artificial system that can understand human social more well enough then that is also helpful for figuring out how it works. And I think we need some deep architectural discussions rather than just trying to change the training date or something like that.