Chest CT's are widely used to diagnose and assess disease severity in patients with COVID-19. In this study, chest CT images from 781 hospitalized patients were analyzed using an AI-based quantitative CT (AIQCT) that automatically segmented the lungs to quantify total airway count (TAC) and pneumonia volume. The study found that patients with critical COVID-19 outcomes exhibited significantly higher TAC than non-critical cases. Patients were further divided into four groups based on cutoff values for pneumonia volume and TAC, and their clinical characteristics and complications were compared. The group with both high pneumonia volume and high TAC demonstrated the highest frequency of critical outcomes. Overall, these results show that AI-based quantification of TAC and pneumonia volume on chest CT is an effective predictor of critical COVID-19 outcomes.