Background: This research aims to identify the most common morphology of the upper premolars in a group of local and nonlocal people in the UAE using cone-beam computed tomography (CBCT).Methods: CBCT images of a total of 214 intact maxillary premolars were analysed with 3D reconstruction GALAXIS viewer software. The Pearson Chi squared test and the two samples t test were applied.Results: In all, 90% of the maxillary first premolars had two roots (92% females, 88% males, 83% locals, 96% nonlocals). A total of 52% of maxillary second premolars had two roots (47% females, 56% males, 38% locals, 67% nonlocals). Two cases had three roots among local males. A significant difference of one root in the maxillary second premolar was observed between locals (63%) and nonlocals (33%), while two roots were higher among nonlocals (67%) than locals (37%, p = 0.002). Three roots were discovered in two teeth of the local UAE first premolar. The most common canal morphology in the maxillary first premolar group among local UAE subjects was type V (52%), and among nonlocal UAE subjects was type V (59%). Additional types included 7 (13.2%) identified and unrelated to the Vertucci classification. The dominant canal morphology in the maxillary second premolar among local UAE subjects was type II (32%), and among nonlocal UAE subjects was type V (25%); among the additional types, 12 (23.1%) were identified and not related to the Vertucci classification. There was a significant difference in the distance from the pulp chamber roof to the cementoenamel junction (CEJ) and the measurement between the single-rooted maxillary premolar CBCT for local UAE subjects was 1.36 mm (SD ±0.75) and for nonlocal subjects was 1 mm (SD ±0.27, p = 0.049). There were statistically significant differences in all measurements between multiple rooted maxillary premolar CBCTs for local UAE and South Asian subjects (p < 0.05)Conclusions: The results suggest a more quantitative approach to maxillary first and second premolar access cavity preparation in the UAE population to prevent errors and iatrogenic damage when identifying the canals; using CBCT measurements for more precise access preparations may be advantageous.