Research has shown that people tend to take more risks when in the presence of their friends than when they are alone. The current research examined whether mere and nonconscious activation of friendship concepts is sufficient to increase risk taking. In Experiment 1, participants who were subliminally primed with their friend’s name showed higher risk-taking responses in a modified longshot paradigm. Experiments 2 and 3 conceptually replicated the findings of Experiment 1 using a behavioral measure of risk taking – the Balloon Analogue Risk Task (BART). In addition, Experiment 3 revealed that the nonconscious friendship-risk taking effect remained even when implicit interpersonal threat expectancies were activated. The results also showed that mood did not mediate the friendship-risk taking effect.