Quasi-Dilemmas for Artificial Moral Agents