It demonstrates a deep problem with a broad class of dark energy models. This provides a theoretical constraint on the evolution of dark energy with time; these serve as useful complements to existing observational limits.

The discovery of the acceleration of the universe seems to imply that there is something important that we do not understand about particle physics and/or gravity. The acceleration could be due to a new source of energy in the universe (dubbed "dark energy") or due to new gravitational physics. If it is due to dark energy, the important issue becomes how that energy evolves as the universe expands. We know that the density of dark energy must evolve gradually, if at all; but we don't know whether the density is slowly decreasing, slowly increasing, or precisely constant. The possibility that the density (the amount of energy in each cubic centimeter of space) might be increasing is intriguing, but violates some cherished beliefs about the behavior of energy in an expanding universe. Our paper examines the possibility that such models could be ultimately unstable, by calculating the rate at which empty space spontaneously decays into gravitons. If it decays in a time shorter than the age of the universe, then it won't work as a dark energy model. We find that this provides an extremely strong constraint on these models.