The existence of adult sexual dimorphism is typically explained as a consequence of sexual selection, yet coevolutionary drivers of sexual dimorphism frequently remain untested. Here, I investigate the role of sexual dimorphism in host-parasite interactions of the brood parasitic diederik cuckoo, Chrysococcyx caprius. Female diederik cuckoos are more cryptic in appearance and pose a threat to the clutch, while male diederik cuckoos are conspicuous and not a direct threat. Specifically, I examine whether sexual dimorphism in diederik cuckoos provokes threat-level sensitive responses in Southern red bishop, Euplectes orix, hosts. I use experimentally simulated nest intrusions to test whether hosts have the capacity to differentially (i) detect, and/or (ii) discriminate between, male and female diederik cuckoos, relative to controls. Overall, I found no evidence that diederik cuckoos differ in detectability, since both sexes are comparable to controls in the probability and speed of host detection. Furthermore, neither male nor female hosts discriminate between sexually dimorphic diederik cuckoos when engaging in frontline nest defences. However, hosts that witnessed a male diederik cuckoo during the trial were more than twice as likely to reject odd eggs compared to those that saw a control. Moreover, hosts were more likely to reject experimental eggs when exposed to a male compared to a female diederik cuckoo: the reverse of a beneficial threat-level sensitive response. While the cryptic appearance of female diederik cuckoos does not differentially avoid detection by hosts, it does appear to provide the benefit of anonymity given the egg rejection costs of male-like appearance in the nest vicinity. These findings have implications for the evolution and maintenance of sexual dimorphism across the Cuculidae, and highlight the value of testing assumptions about the ecological drivers of sexual dimorphism.