The understanding of food cue associated neural activations that predict future weight variability may guide the design of effective prevention programs and treatments for overeating and obesity. The current study investigated the association between brain response to different food odors with varied energy density and individual changes of body mass index (BMI) over 2 years. Twenty-five participants received high-fat (chocolate and peanut), low-fat (bread and peach) food odors, and a nonfood odor (rose) while the brain activation was measured using functional magnetic resonance imaging (fMRI). BMIs were calculated with participant's self-reported body weight and height collected at the time of the fMRI scan and again at 2 years later. Regression analyses revealed significant negative correlations between BMI increase over 2 years and brain activation of the bilateral precuneus and the right posterior cingulate cortex (PCC) in response to high-fat vs. low-fat food odors. Also, brain activation of the right supplementary motor area (SMA) in response to food vs. non-food odor was negatively correlated to subsequent BMI increase over 2 years. Taken together, the current findings suggest that individual differences in neural responsivity to (high calorie) food odors in brain regions of the default mode and motor control network serve as a neural marker for future BMI change.