This study aimed to determine the acceptability of the ACASI approach to risk assessment and the impact of personal preference regarding mode of interview on reporting risk behaviors among drug users entering treatment in Rio de Janeiro, Brazil. We assessed 268 substance users who completed the ACASI arm in a randomized trial comparing the ACASI with the Interviewer-Administered Questionnaire (IAQ). The vast majority of interviewees (90.7%) reported no problem using the computer, and 37.3% felt that their privacy was best protected by the ACASI (vs. 16.4% who preferred the IAQ). Nearly half (45.5%) reported that the computer interview would produce more "honest" answers, whereas 30.6% selected the IAQ. In the adjusted regression analysis, problems using the computer were associated only with lower educational level (p<0.05). We found no evidence that preference had an impact on reporting risk behaviors or drug use. Our study showed both good feasibility and acceptability of the ACASI for interviewing drug users in Brazil. The findings extend our understanding of the role of the ACASI method by suggesting the utility of this approach in assessing HIV risk among low-to middle-income drug users in a cultural setting quite different from previous studies.