Project Summary The disproportionate incidence of HIV/AIDS among Black men underscores a dire need for technologically innovative and culturally-responsive approaches to reduce HIV in Black communities. There is evidence that social media platforms such as Twitter may offer an unprecedented opportunity to identify and recruit young Black men (YBM) at elevated risk, but to date there is a dearth of empirical evidence to support that social media based-recruitment using natural language processing is feasible and/or acceptable. On social media platforms like Twitter, young men share substance and sex related content. We propose mining these Twitter posts to identify YBM who may be at risk of elevated HIV, based on their substance and sexual risk posts. This study will serve as a proof of method to test the extent to which we can identify and recruit YBM based on their social media posts. We will extend and calibrate our “High-Risk Language” (HRL) algorithm to identify users who post HIV-risk related language on Twitter. Those who agree to participate and deemed eligible will complete a qualitative interview and in- person study visit which includes HIV testing, multi-drug screening, and a survey of their sexual risk and substance use behaviors. Our primary outcomes are; (1) condomless (and/or without PrEP) vaginal or anal sex and (2) illicit drug use before or during sex. We will assess the yield and accuracy of our HRL algorithm at identifying YBM at risk, compared to their biomarkers and surveys. This work will test cutting-edge research tools aimed at increasing the rigor and precision of online recruitment efforts to reduce HIV in the Black community.