Might be a worthwhile experiment to see if the opposite is possible. We definitely should be developing artificial ethics before artificial general intelligence.