Gay male couples (N = 47) completed self-administered questionnaires at 2 time points (6 months apart) regarding their levels of general and HIV-specific social support and sexual behavior. HIV-specific social support measured partner support pertaining to HIV risk behavior. The sexual-risk-behavior outcome encapsulated monogamy, serostatus, and unprotected anal sex for each partner. The authors used an analytic approach that maintained the couple as the unit of analysis. General social support was an inconsistent predictor of HIV risk behavior. However, couples that reported greater levels of HIV-specific social support engaged in less HIV risk behavior at each time point as well as longitudinally. This study demonstrated the utility of measuring HIV-specific social support and its predictive ability related to HIV risk behavior.