Even though some SMIA have changed their behavior in response to AIDs and HIV prevention programs, improved interventions are needed to promote safer sex among SMIA in the community. A long history of research has shown that persons who are engaged to voluntarily advocate a position often change their attitudes and behavior to become congruent with their advocacy position. Another body of research has shown that sexual risk practices of SMIA are influenced by their perceptions of condom use and social or peer group norms. Taken together, these findings--and results of community-level HIV prevention research--suggest that an intervention that trains SMIA to function as HIV prevention advocates or endorsers to their peers may both help the advocate better maintain safer behavior and also influence the risk characteristics of the advocate's social network members. Following formative studies undertaken to delineate HIV risk issues facing SMIA, the investigator will initiate a study to test the hypothesis that HIV prevention outcomes for intervention participants and members of their social networks will be most successful among SMIA who are taught to advocate behavior change recommendations and norms to their network members favoring risk avoidance. The study will recruit 348 SMIA (174 men and 174 women) who have engaged in high-risk sexual transmission acts in the past 90 days. The two closest personal network members of SMIA participant will also be recruited. Following baseline assessment of sexual practices in the past 30 days and other risk-related characteristics, participants will be randomly assigned within gender and in equal numbers to a comparison cognitive-behavioral risk reduction intervention in which participants attend seven 90-minute group sessions that focus on making behavior changes to reduce risk for AIDS; or a 7-session group intervention that teaches risk reduction behavior changes but also teaches and engages participants to effectively communicate AIDS prevention messages to their friends and acquaintances (""""""""advocacy training""""""""). Follow-up risk assessments will be conducted with participants and their network members 3, 6, and 12-months postintervention. The investigator hypothesizes that participants in the advocacy training intervention will show greatest change in HIV risk-related characteristics; will show greatest reduction in unprotected sexual acts; and will most influence others in their immediate networks to also lower their HIV risk practices. If successful, this research will suggest new methods for community HIV prevention approaches for SMIA and their social network members.