Plant responses to elevated atmospheric carbon dioxide (eCO 2 ) have been hypothesised as a key mechanism that may ameliorate the impact of future drought. Yet, despite decades of experiments, the evidence that eCO 2 reduces plant water use, yielding "water savings" that can be used to maintain plant function during periods of water stress, is mixed. In this Viewpoint, we highlight key literature and identify experimental challenges and limitations to our understanding of plant responses to drought under eCO 2 . In particular, we argue that future studies need to move beyond exploring whether eCO 2 played 'a role' or 'no role' in responses to drought, but instead more carefully consider the timescales and conditions that would induce an influence. We also argue that considering emergent differences in soil water content may be an insufficient means of assessing the impact of eCO 2 . We identify eCO 2 impact during severe drought (e.g., to the point of mortality), interactions with future changes in vapour pressure deficit and uncertainty about changes in leaf area as key gaps in our current understanding. New insights into CO 2 x drought interactions are essential to better constrain model theory that governs future climate model projections of land-atmosphere interactions during periods of water stress.