AIDS risk reduction programs are being conducted in many institutional settings, but rigorous evaluations of their effectiveness are lacking. This is particularly unfortunate in that these programs are expensive, and tend to be of lower intensity than those that have been shown to be effective. Further, risk reduction is generally regarded as entailing greater difficulty for women, who do not use condoms themselves but must negotiate their use with male partners. We used a quasi-experimental design to evaluate an institutional AIDS prevention program on a New Jersey college campus. Sexual behavior was assessed via linked, anonymous mailed surveys at the beginning and end of an academic year among 1st-year students on the campus and others on a nearby control campus. Responses from the spring survey indicated that intervention campus students had been exposed significantly more than control students to intervention components. While MANCOVA analyses indicated no main effect of treatment group on outcome variable, we obtained a significant group by gender interaction, indicating a significant effect on number of risky encounters for men but not for women. In fact, relative to women on the control campus, women on the intervention campus displayed reduced self-efficacy to perform safe sex at the end of the year. These results may indicate that although men can be effectively reached by low-intensity risk reduction programs, women may not be. In fact, interventions without adequate intensity to provide substantial and individualized negotiation skill training may cause women to experience failure in these efforts.