We described public views toward harm reduction among Canadian adults and tested a social exposure model predicting support for these contentious services, drawing on theories in the morality policy, intergroup relations, addiction, and media communication literatures. A quota sample of 4645 adults (18+ years), randomly drawn from an online research panel and stratified to match age and sex distributions of adults within and across Canadian provinces, was recruited in June 2018. Participants completed survey items assessing support for harm reduction for people who use drugs (PWUD) and for seven harm reduction interventions. Additional items assessed exposure to media coverage on harm reduction, and scales assessing stigma toward PWUD (α = .72), personal familiarity with PWUD (α = .84), and disease model beliefs about addiction (α = .79). Most (64%) Canadians supported harm reduction (provincial estimates = 60% - 73%). Five of seven interventions received majority support, including: outreach (79%), naloxone (72%), drug checking (70%), needle distribution (60%) and supervised drug consumption (55%). Low-threshold opioid agonist treatment and safe inhalation interventions received less support (49% and 44%). Our social exposure model, adjusted for respondent sex, household income, political views, and education, exhibited good fit and accounted for 17% of variance in public support for harm reduction. Personal familiarity with PWUD and disease model beliefs about addiction were directly associated with support (βs = .07 and -0.10, respectively), and indirectly influenced public support via stigmatized attitudes toward PWUD (βs = 0.01 and -0.01, respectively). Strategies to increase support for harm reduction could problematize certain disease model beliefs (e.g., “There are only two possibilities for an alcoholic or drug addict–permanent abstinence or death”) and creating opportunities to reduce social distance between PWUD, the public, and policy makers.