Advances in neuroimaging techniques have significantly improved our understanding of the neural basis of autism spectrum disorder (ASD). Several attempts have been made to label the main neuroimaging phenotype of ASD, mostly by anatomical and functional activation studies, but none of the frameworks have been without controversy. Over the past decade, a renewed interest for rest brain functioning has emerged in the scientific community, reflected on a large number of resting state fMRI (rs-fMRI) studies, but results remain heterogeneous. It is possible today to investigate rest brain functioning by measuring rest cerebral blood flow (CBF) with MRI using arterial spin labeling (ASL). Here, we investigated rest CBF abnormalities using noninvasive ASL-MRI in 18 children with ASD without cognitive delay (10.4 ± 2.8 y) and 30 typically developing children (10.6 ± 3.0 y). Following quality control, images from a final sample of 12 children with ASD (11.2 ± 2.9 y) and 28 typically developing children (10.1 ± 2.5 y) were analyzed. Whole brain voxel-by-voxel analysis showed significant rest CBF decrease in temporal regions, mainly in the superior temporal sulcus (STS), in children with ASD. This hypoperfusion was individually detected in 83% of children with ASD. Finally, negative correlation was observed between ASD severity scores and rest CBF in the right posterior STS. Strikingly, despite the small sample studied here, our results are extremely similar to previous PET and SPECT findings describing decreased rest CBF in the same superior temporal regions at group and individual levels, as well as correlation with symptoms severity. The congruence between these results, with different methods and in different ASD profiles, reinforce the strength of rest functional abnormalities within these superior temporal regions in ASD and strongly indicates it might be a core characteristic of the disorder. Identifying a core dysfunctional region in ASD bears direct implications to the development of novel therapeutic interventions, such as transcranial magnetic stimulation. In addition, if confirmed in a larger sample, rest temporal hypoperfusion could become a reliable brain imaging biomarker in ASD.