Mass loss from the Greenland ice sheet (GrIS) has increased over the last two decades in response to changes in global climate, motivating the scientific community to question how the GrIS will contribute to sea‐level rise on timescales that are relevant to coastal communities. Observations also indicate that the impact of a melting GrIS extends beyond sea‐level rise, including changes to ocean properties and circulation, nutrient and sediment cycling, and ecosystem function. Unfortunately, despite the rapid growth of interest in GrIS mass loss and its impacts, we still lack the ability to confidently predict the rate of future mass loss and the full impacts of this mass loss on the globe. Uncertainty in GrIS mass loss projections in part stems from the nonlinear response of the ice sheet to climate forcing, with many processes at play that influence how mass is lost. This is particularly true for outlet glaciers in Greenland that terminate in the ocean because their flow is strongly controlled by multiple processes that alter their boundary conditions at the ice‐atmosphere, ice‐ocean, and ice‐bed interfaces. Many of these processes change on a range of overlapping timescales and are challenging to observe, making them difficult to understand and thus missing in prognostic ice sheet/climate models. For example, recent (beginning in the late 1990s) mass loss via outlet glaciers has been attributed primarily to changing ice‐ocean interactions, driven by both oceanic and atmospheric warming, but the exact mechanisms controlling the onset of glacier retreat and the processes that regulate the amount of retreat remain uncertain. Here we review the progress in understanding GrIS outlet glacier sensitivity to climate change, how mass loss has changed over time, and how our understanding has evolved as observational capacity expanded. Although many processes are far better understood than they were even a decade ago, fundamental gaps in our understanding of certain processes remain. These gaps impede our ability to understand past changes in dynamics and to make more accurate mass loss projections under future climate change. As such, there is a pressing need for (1) improved, long‐term observations at the ice‐ocean and ice‐bed boundaries, (2) more observationally constrained numerical ice flow models that are coupled to atmosphere and ocean models, and (3) continued development of a collaborative and interdisciplinary scientific community.