We present a simple mathematical model for assessing the effects of introducing a microbicide as an HIV infection protective method. As very little is known about the in vivo efficacy of microbicides, we ran sample scenarios for microbicides of various efficacies. We found that, in general, if existing condom usage in a community is low, introducing a microbicide will most likely have a positive impact on HIV incidence as abandonment of condom use in favor of microbicides will not play a significant role. If condom use in a community is high, though, attrition of condom users could play a role large enough to overwhelm any added risk reduction afforded new microbicide users. Our model illustrates the importance of knowing key behavioral parameters, such as the proportion of the population that uses condoms, before microbicides can be safely introduced. These parameters include the proportion of condom users likely to maintain condom use and the proportion of condom nonusers likely to adopt microbicides, as well as the efficacy of the candidate microbicide.