An enormous amount of four dimensional effective field theories, many of them with interesting phenomenological implications, can be obtained from compactifications of string theory. The set of these theories is known as the landscape of string theory. It has been suggested that, beyond this landscape, there exist consistent-looking effective field theories that cannot be embedded into UV complete theories of quantum gravity, in particular into string theory. They are said to belong to the swampland. In this talk I will describe some of the conjectures that define the boundaries of the swampland and implications they have on phenomenology, in particular in models of dark matter and inflation. Moreover, I will discuss tentative explorations on the consistency of non-local field theories and their possible relations to the swampland program.