OBJECTIVES: We examined the efficacy of a peer-driven intervention to increase rates of screening for AIDS clinical trials among African Americans and Hispanics living with HIV/AIDS. METHODS: We used a randomized controlled trial design to examine the efficacy of peer-driven intervention (6 hours of structured sessions and the opportunity to educate 3 peers) compared with a time-matched control intervention. Participants were recruited using respondent-driven sampling (n = 342; 43.9% female; 64.9% African American, 26.6% Hispanic). Most participants (93.3%) completed intervention sessions and 64.9% recruited or educated peers. Baseline and post-baseline interviews (94.4% completed) were computer-assisted. A mixed model was used to examine intervention effects on screening. RESULTS: Screening was much more likely in the peer-driven intervention than in the control arm (adjusted odds ratio [AOR] = 55.0; z = 5.49, P < .001); about half of the participants in the intervention arm (46.0%) were screened compared with 1.6% of controls. The experience of recruiting and educating each peer also increased screening odds among those who were themselves recruited and educated by peers (AOR = 1.4; z = 2.06, P < .05). CONCLUSIONS: Peer-driven intervention was highly efficacious in increasing AIDS clinical trial screening rates among African Americans and Hispanics living with HIV/AIDS.