Adolescents' contact with the criminal justice system is associated with a number of long-term negative adult outcomes: unemployment, substance abuse, and incarceration (Moffitt, 2002). The most recent statistics indicate that nationally, there are at least 634,100 youth on probation and 39,000 youth in LA County are on probation (Office of Juvenile Justice & Delinquency Prevention, 1999). Delinquent youth on probation have multiple sexual partners, frequent unprotected sexual risk acts, high rates of sexually transmitted diseases and substance use, as well as gang involvement (Morris et al., 1995). HIV prevention is highly efficacious with adolescents (NIH, 1997); however, a successful program has not been developed for these youth. Building on the results of three efficacious HIV prevention programs with those at high HIV risk (CDC, 2001; LIGHT, Adolescent LIGHT, & Street Smart) and the PI's recent R21 project, this proposal aims to examine the relative efficacy of LIGHT in an interpersonal or a computerized delivery format, in contrast to a standard care HIV prevention condition. This application is in response to a request for applications (RFA-DA-04-012) to address critical gaps in research on HIV/AIDS prevention, treatment, and related health issues among highly vulnerable youth. The project will proceed in two phases. During Phase 1, the existing computerized LIGHT prevention program will be enhanced further to further ensure acceptability, assessment measures will be finalized and programmed on a combination ACASI/CAPI program, and 1KB applications finalized within each school, recruitment scripts and tracking programs finalized, and orientation sessions conducted for all school personnel, parents, and stakeholders. The main intervention trial, Phase 2, will utilize a complete randomized block design with youth recruited over a two-year period. Twenty four schools (n=960 students) in the Los Angeles Department of Education will be randomly assigned into one of three experimental conditions: 1) Interpersonal Light in 8 schools; 2) Computerized Light in 8 schools (n=320 youth); or 3) a standard HIV control condition. Youth in each condition will be assessed at a baseline interview, and reassessed longitudinally at 3, 6, and 12 months. The results of this study will inform researchers and the public health community about the generalizability of an HIV preventive intervention and feasibility and acceptability of an interpersonal intervention delivered in classrooms compared to delivery on a CD-ROM at computers within each classroom.