AI systems should be designed and operated so as to be compatible with ideals of human dignity, rights, freedoms, and cultural diversity.