Using baseline data of the Engage Cohort Study, a Canadian study of sexually active gay, bisexual and other men who have sex with men (GBM), we evaluated the association between sexual behavior and risk perception among HIV-negative participants and whether HIV treatment optimism moderated this relationship. Participants were recruited by respondent-driven-sampling (RDS). We defined high-risk sexual behavior in the past six months as any condomless anal sex with a casual partner (i.e. not the participant's main partner) with either unknown HIV-status where neither used pre-exposure prophylaxis or with a partner living with HIV having detectable/unknown viral load. We assessed HIV treatment optimism-skepticism using a 12-item scale. RDS-II-weighted adjusted logistic regression models examined associations with risk perception measured by the question "How would you assess your current risk of getting HIV?" (response options were on a 6-point Likert-scale ranging from "very unlikely" to "very likely", dichotomized into "No Perceived Risk" (very unlikely/unlikely) and "Perceived Risk" (somewhat likely/likely/very likely/I think I already have HIV). Of 1961 participants, engagement in high-risk sexual behavior was reported by 155 (17.0%), 62 (12.4%), 128 (17.2%) of participants in Montréal, Toronto, and Vancouver, respectively. High-risk sexual behavior increased the odds of perceived HIV risk (pooled adjusted odds ratio = 2.9, 95%CI = 2.2-3.8). HIV treatment optimism-skepticism scores moderated the relationship: for GBM engaging in high-risk sexual behavior, higher HIV treatment optimism-skepticism scores increased perceived HIV risk. Promoting awareness around advances related to HIV prevention and treatment is important for appropriate risk assessment and for increased engagement in prevention interventions.