People often have some degree of credence in false moral claims.  It is intuitive that, in some cases, it is morally wrong to act against these credences.  Theories which attempt to accommodate these intuitions have run into difficulties.  For one, they tend to have other counterintuitive implications.  They also struggle to explain why it matters that one does something that isn’t (in some sense) “really” wrong.  Finally, it’s hard to say what an agent should do when they have some degree of credence in radically different moral claims which are relevant to a given choice; this is sometimes called the problem of “inter-theoretic value comparisons.”  Discussions of what one should do in light of one’s false moral beliefs, or moral uncertainty, often try to remain agnostic about what the correct moral theory actually is.  I think we can make more progress without this agnosticism.  I discuss how, by starting with some relatively innocuous ideas in deontological ethics, it looks like we can get a view about false moral beliefs and moral uncertainty that fits a number of desiderata.  It says that false moral credences generate moral obligations in cases where it seems intuitive that they do, but not in cases where it is intuitive that they do not.  It also explains why these obligations exist, and it allows us to determine their weight (which is vital, since they can conflict with our other moral duties). Finally, it avoids the problem of inter-theoretic value comparisons.