Purpose <p> Posttraumatic stress disorder (PTSD) is characterized by heightened threat sensitivity, including attentional bias to threat, and associated with increased alcohol consumption. Given that alcohol consumption may be driven by or worsen PTSD symptoms, it is critical to consider how threat sensitivity relates to drinking in individuals with probable PTSD.</p> Methods <p> In this study, 687 adults with probable PTSD (PTSD Checklist for DSM-5 symptoms ≥ 33; M_age=30.01; 80% female) completed a 15-day app-based attention bias modification (ABM) protocol to reduce threat-related biases. Participants were randomized into personalized ABM, non-personalized ABM, or placebo training (days 2–7, 9–14), and completed dot-probe detection assessments on days 1, 8, and 15. Participants reported alcoholic drinks consumed in the past 24&#xa0;h every training day. We examined associations of ABM condition, threat sensitivity (self-reported, avoidance and vigilance reaction time biases), stress (past, anticipated), and PTSD symptoms with alcohol consumption (drink count, heavy drinking) using mixed models.</p> Results <p> Personalized ABM was not significantly associated with reduced drink count, relative to placebo (rate ratio <i>[RR]</i> = 0.72, 95% CI 0.51, 1.01). Higher self-reported threat sensitivity was associated with lower next-day drink count (<i>RR</i> = 0.93, 95% CI 0.87, 0.99; marginal after multiple corrections), while there was a non-significant effect between avoidance bias and lower next-day drink count (<i>RR</i> = 0.88, 95% CI 0.76, 1.02). Higher past and anticipated stress were associated with lower next-day drink count (<i>RR</i> = 0.92, 95% CI 0.87, 0.96) and heavy drinking (odds ratio <i>[OR]</i> = 0.91, 95% CI 0.82, 1.00), respectively.</p> Conclusions <p> Our findings link threat sensitivity and stress with subsequent lower alcohol consumption, necessitating further research on the intersection of these factors with alcohol use in populations with probable PTSD.</p>