A recent paper examines the derivation of Einstein's field equations and proposes that the original assumption of a divergence-free energy-momentum tensor may not be valid anymore due to the discovery of dark matter and dark energy. The authors derive new field equations in which [itex]T_{\mu\nu}[/itex] isn't divergence-free.

This step has several striking consequences. For example, the new Schwarzschild solution contains extra terms that describe the effect of dark matter, and even the repulsive effect of dark energy. The result fits observation, as it correctly describes Rubin's law for galactic rotation curves, and a rough estimation done in the paper predicts the real gravitational mass to be about 4 times the amount of visible mass. Also, the distances at which dark matter and dark energy effects become significant appear to fit observation as well. Furthermore, according to the revised theory, the nature of dark matter and energy has to do with a new quantity in the field equations, a scalar potential energy density, that is shown to reflect the non-uniform distribution of matter in the universe.

Here's the paper: http://arxiv.org/abs/1206.5078 (Conclusions on page 31)

I've read this paper several times by now and am quite excited about these new ideas. I have long suspected there might be something slightly but fundamentally wrong with general relativity. Now dark energy and dark matter naturally pop out of the new equations and are part of the normal gravitational field of matter in the universe.