A community based cross-sectional study inquiring into determinants and correlates of sexual risk taking was conducted from June to December 2003. The objective of the study was describing sexual behavior and assessing socio-economic and cultural correlates of high-risk sexual behavior among urban youth in Addis Ababa. A multi-systemic perspective, an approach that focuses on the reciprocal relations among the personal, family, and extra family systems on sexual behavior was employed. Data were collected using anonymous questionnaire based interview, seven focus group discussions and four case studies of PLWHA. A total of 925 youth aged 15 to 24 from inner and outer Addis Ababa participated in the survey. Females make 53.3% and mean age was 18.8 years. Over 60% of the respondents were in school and one third was employed in the labor force. Nearly a third (28.1%) were sexually active, 66.2% of them being males. Close to half (46.5%) of the sexually active youth had history of multiple sexual partners and 2.3% had history of STD. Condom usage at debut was 36.5% with 30.8% regular usage rate. One fifth (20.3%) of the sexually active males admitted having had sex with CSWs and same proportion of girls had been pregnant. Out of 12 HIV knowledge related questions, almost 75% scored 10-12 (adequate knowledge). Only 5.3% perceive personal risk and 3.8% were found to have favorable self-protective attitude. From the self system, being a male, age 20-24, taking part in the labor force, watching pornography, alcohol consumption, khat use and having taken the HIV test kept on having statistically significant association with potential sexual risk-taking behavior. Youth with strong parental monitoring and who discuss sexual matter with parents demonstrated less indulgence in sexual activity. Living in the inner city, and having sexually active friends and involvement with delinquent peers were correlates of sexual risk taking with in the extra-familial system. The FGDs and the cases also confirmed the survey findings and further revealed that socially disadvantaged youth such as CSWs and street youth were more at risk than others whereby the influence of familial and extra familial systems predominate. CONCLUSION: Many young people are engaged in sexual behavior that is considerably risky for HIV/AIDS. There is neither a single explanation for sexual risk taking among youth nor is there a single intervention to curb the problem of HIV/AIDS. There is a need to consider the entirety of personal, familial and extra familial factors as co-determinants of high-risk sexual behavior.