The topic of this chapter is how rational agents should revise their opinions if they accept the gradualist hypothesis, and discover that others have moral opinions that differ from their own. I defend the following moral convergence thesis: All rational moral peers who accept the axioms of the probability calculus will adopt the same opinions with the same credence, regardless of what their initial opinions are. My argument for the moral convergence thesis builds on recent work in epistemology. I define moral peers as agents who have access to the same information and are equally likely to evaluate the acceptability of a moral claim correctly, but I leave it open what it means to say that a moral claim is evaluated “correctly”. The moral convergence thesis is a strong claim, but it can be supported by a general argument. This argument has some limitations, which are explored in a series of computer simulations.