Exosomes are virus-sized nanoparticles (30–130 nm) formed intracellularly as intravesicular bodies/intralumenal vesicles within maturing endosomes (“multivesicular bodies”, MVBs). If MVBs fuse with the cell’s plasma membrane, the interior vesicles may be released extracellularly, and are termed “exosomes”. The protein cargo of exosomes consists of cytosolic, membrane, and extracellular proteins, along with membrane-derived lipids, and an extraordinary variety of nucleic acids. As such, exosomes reflect the status and identity of the parent cell, and are considered as tiny cellular surrogates. Because of this closely entwined relationship between exosome content and the source/status of the parental cell, conceivably exosomes could be used as vaccines against various pathologies, as they contain antigens associated with a given disease, e.g., cancer. Tumor-derived exosomes (TEX) have been shown to be potent anticancer vaccines in animal models, driving antigen-specific T and B cell responses, but much recent literature concerning TEX strongly places the vesicles as powerfully immunosuppressive. This dichotomy suggests that the context in which the immune system encounters TEX is critical in determining immune stimulation versus immunosuppression. Here, we review literature on both sides of this immune coin, and suggest that it may be time to revisit the concept of TEX as anticancer vaccines in clinical settings.