What we ought to do and what we ought to believe depends on what we know. That is the way many decision makers, scientific experts and ordinary citizens think: for instance, if we do not know whether a drug is safe enough, we should not market it. The idea is rejected, however, by dominant theories of rationality. Standard decision theory, bayesian accounts of scientific rationality and many ethical theories hold that rational choice and rational theory acceptance depend on what we believe or experience rather than what we know. Recent developments in epistemology have challenged these views. The project builds on them to develop a full-blown account of rationality in terms of knowledge. It aims at reconciling our fundamental theories of rationality with the way scientific experts, decision makers and citizens ordinarily think about rational decision and belief.