Anxiety disorders contribute substantially to the overall public health burden of psychopathology. Anxiety sensitivity (AS), a fear of anxiety related sensations, is one of the few known malleable risk factors for anxiety pathology. Previous AS reduction treatments have utilized highly trained clinicians. A completely-computerized AS treatment would reduce costs and increase dissemination possibilities. Cognitive bias modification for interpretation biases (CBM-I) interventions have shown clinically significant reductions in anxiety symptoms. Another emerging literature focused on learning has shown context-shifting tasks can greatly increase learning without adding logistical burden to an intervention. The current study evaluated a CBM-I for AS that utilized a context-shifting task to deliver twice the treatment dose of extant interventions.Single-site randomized controlled trial. Participants completed an intervention appointment, as well as one-week and one-month follow-up assessments.Individuals with elevated levels of AS.Single-session computer-delivered CBM-I for AS.Results indicate that the CBM-I for AS was successful in reducing overall AS (62% post-intervention) and these reductions were maintained through one-month post-intervention (64%). Results also revealed that individuals in the active condition reported significantly less incidents of panic responding to a physiological straw-breathing challenge and that change in interpretation bias significantly mediated the relationship between treatment condition and post-treatment AS reductions.Taken together, the results show that the current CBM-I intervention was strong in terms of immediate and one-month AS reductions. Given its brevity, low cost, low stigma and portability, this intervention could have substantial impact on reducing the burden of anxiety disorders.