Clément L comments on Utilitarian AI Alignment: Building a Moral Assistant with the Constitutional AI Method