Mounting evidence supports the rapid antidepressant efficacy of the N-methyl-D-aspartate receptor antagonist, ketamine, for treating major depressive disorder (MDD); however, its neural mechanism of action remains poorly understood. Subgenual anterior cingulate cortex (sgACC) hyper-activity during rest has been consistently implicated in the pathophysiology of MDD, potentially driven in part by excessive hippocampal gluatmatergic efferents to sgACC. Reduction of sgACC activity has been associated with successful antidepressant treatment. This study aimed to examine whether task-based sgACC activity was higher in patients with MDD compared to controls and to determine whether this activity was altered by single-dose ketamine. In Study 1, patients with MDD (N = 28) and healthy controls (N = 20) completed task-based functional magnetic resonance imaging using an established incentive-processing task. In Study 2, a second cohort of patients with MDD (N = 14) completed the same scanning protocol at baseline and following a 40 min infusion of ketamine (0.5 mg/kg). Task-based activation of sgACC was examined with a seed-driven analysis assessing group differences and changes from pre to post treatment. Patients with MDD showed higher sgACC activation to positive and negative monetary incentives compared to controls, associated with anhedonia and anxiety, respectively. In addition, patients with MDD had higher resting-state functional connectivity between hippocampus and sgACC, associated with sgACC hyper-activation to positive incentives, but not negative incentives. Finally, ketamine reduced sgACC hyper-activation to positive incentives, but not negative incentives. These findings suggest a neural mechanism by which ketamine exerts its antidepressant efficacy, via rapid blunting of aberrant sgACC hyper-reactivity to positive incentives.