Nitrous oxide (N2O) is an important greenhouse gas and ozone depleting substance. Previous projections of agricultural N2O (the dominant anthropogenic source)show emissions changing in tandem, or at a faster rate than changes in nitrogen (N) consumption. However, recent studies suggest that the carbon dioxide (CO2) fertilization effect may increase plant N uptake, which could decrease soil N losses and dampen increases in N2O. To evaluate this hypothesis at a global scale, we use a process-based land model with a coupled carbon-nitrogen cycle to examine how changes in climatic factors, land-use, and N application rates could affect agricultural N2O emissions by 2050. Assuming little improvement in N use efficiency (NUE), the model projects a 24%–31% increase in global agricultural N2O emissions by 2040–2050 depending on the climate scenario—a relatively moderate increase compared to the projected increases in N inputs(42%–44%) and previously published emissions projections(38%–75%). This occurs largely because the CO2 fertilization effect enhances plant N uptake in several regions, which subsequently dampens N2O emissions. And yet, improvements in NUE could still deliver important environmental benefits by 2050: equivalent to 10 Pg CO2 equivalent and 0.6 Tg ozone depletion potential.