This review aims to provide a brief historical overview of the feto-maternal immunological relationship, which profoundly influences the outcome of pregnancy. The initial question posed in the 1950s by Medawar [Symp Soc Exp Biol. 1953; 7: 320–338] was based on the assumption that the maternal immune system recognizes the fetus as an allograft. Indeed, based on the association between HLA-matching and spontaneous miscarriage, it became obvious that immunological recognition of pregnancy is required for a successful gestation. The restricted expression of polymorphic HLA antigens on the trophoblast, together with the presence of nonpolymorphic MHC products, excludes recognition by both T and NK cells of trophoblast-presented antigens; however, γδ T cells, which constitute the majority of decidual T cells, are likely candidates. Indeed, a high number of activated, progesterone receptor-expressing γδ T cells are present in the peripheral blood of healthy pregnant women and, in the presence of progesterone, these cells secrete an immunomodulatory protein called progesterone-induced blocking factor (PIBF). As early as in the peri-implantation period, the embryo communicates with the maternal immune system via PIBF containing extracellular vesicles. PIBF contributes to the dominance of Th2-type reactivity which characterizes normal pregnancy by inducing increased production of Th2 cytokines. The high expression of this molecule in the decidua might be one of the reasons for the low cytotoxic activity of decidual NK cells.