Rates of HIV transmission amongst gay men who are in relationships continue to rise despite many years of public health campaigns that are aimed at reducing unprotected anal intercourse, through always using a condom. Previous research suggests that condom use is associated with a number of psycho-social variables, including demographic variables, attitudes, perceived partner norms and self-efficacy. The present study introduces power as a possible influential variable on condom use, in particular because it has been shown to influence condom use in heterosexual relationships. Perceived risk is also examined in the context of condom use. Additionally, the first incidence of unprotected anal intercourse is examined. The present study uses a quantitative approach to investigate these relationships. Ninety participants downloaded the web-based self-report questionnaire. Analyses suggest that there was no relationship between power and condom use. Furthermore the only relationship shown to be significant was self-efficacy and condom use. Interventions regarding self-efficacy are examined. This failure to support the main hypotheses may be due to methodological factors. These methodological factors are examined at great length regarding the difficulty in investigating sexual behaviour, particularly regarding the definitions safe sex in relationships. A discussion of what direction future research may wish to take in light of these findings concludes this study.