Unrecognized HIV infections, where the individual is unaware of his or her HIV serostatus, may play the most important role in the continued spread of HIV. Using venue-level data from the National HIV Behavioral Surveillance System, we conducted preliminary bivariate analyses to identify variables associated with unrecognized HIV infection and subsequently entered those variables into multiple logistic regression models to then characterize unrecognized HIV infections in Los Angeles and San Francisco. In Los Angeles, public sex environments have the highest likelihood of having men who have sex with men (MSM) with unrecognized HIV infection (adjusted odds ratio (AOR) 3.2, 95% confidence interval (CI) = 1.3-7.9). For San Francisco, MSM with unrecognized HIV infection are most likely found in adult bookstores (AOR = 8.0, 95% CI = 2.8-39). We identified opportunities to promote HIV testing through specific locations where unrecognized infections among MSM are found in highest concentrations.