Previous work on social categorization has shown that people often use cues such as a person's gender, age, or ethnicity to categorize and form impressions of others. The present research investigated effects of social category membership on the evaluation of humanoid robots. More specifically, participants rated a humanoid robot that either belonged to their in-group or to a national out-group with regard to anthropomorphism (e.g., mind attribution, warmth), psychological closeness, contact intentions, and design. We predicted that participants would show an in-group bias towards the robot that ostensibly belonged to their in-group – as indicated by its name and location of production. In line with our hypotheses, participants not only rated the in-group robot more favourably – importantly, they also anthropomorphized it more strongly than the out-group robot. Our findings thus document that people even apply social categorization processes and subsequent differential social evaluations to robots.