BACKGROUND: A public health concern regarding HIV pre-exposure prophylaxis (PrEP) is sexual risk compensation (ie, increased unsafe sex among PrEP users that may undermine prevention efforts). METHODS: This demonstration study (NCT#01761643; initiated in 2013) included 398 men who have sex with men who initiated PrEP and were followed over 48 weeks at 4 sites in Southern California. Wilcoxon signed-rank tests compared previous 30-day number of sex partners and condomless insertive anal sex and receptive anal sex (CIAS and CRAS, respectively) acts at weeks 4, 12, 24, 36, and 48 to baseline. At 2 sites, PrEP users were also compared with a lagged, comparison group of 99 men who have sex with men who did not receive PrEP over 24 weeks using linear regression models, adjusting for age, race/ethnicity, education, and respective baseline scores. Logistic regression compared week 24 sexually transmitted infection (STI) rates. RESULTS: Over 48 weeks in the PrEP group, there were significant decreases in the number of unknown HIV status sex partners and increases in CRAS at all study visits; there was no consistent change in number of HIV+ sex partners or CIAS. Among participants at 2 sites, there were no significant differences between PrEP and non-PrEP users in change in number of partners, CIAS, CRAS, or STI rates at week 24. CONCLUSIONS: Among early adopters of PrEP, there is some evidence for sexual risk compensation. Results support current guidelines of regular STI screening and behavioral risk reduction and adherence counseling with the provision of PrEP.