> It just solves a conservation problem. This is news to me. The story I was taught was that Einstein's equations predicted a universe that wasn't static, i.e.

> It just solves a conservation problem. This is news to me. The story I was taught was that Einstein's equations predicted a universe that wasn't static, i.e. an expanding universe. Einstein didn't like that and introduced the cosmological constant to leave the possibility of a static universe. Then, it was found that it was indeed expanding. Einstein promptly regretted his decision ("biggest blunder") to introduce the cosmological constant. Then, it was found that the universe was not only expanding, but expanding in an accelerated fashion. That is at odds with Einstein's equations without the cosmological constant, so it was re-introduced, but with the opposite sign. If you interpret the cosmological constant as part of the already present energy-stress tensor, it could represent the energy density of vacuum itself. That is only one interpretation of "dark energy" though, as the term is an umbrella term for whatever is responsible for the acceleration. Energy conservation never entered the discussion. In fact, energy conservation generally isn't even expected to hold in cosmology. Energy is not even well defined, as it is directly associated with time translation invariance as per Noether's theorem, which doesn't exist in an expanding universe. See: https://www.preposterousuniverse.com/blog/2010/02/22/energy-is-not-conserved/ https://www.preposterousuniverse.com/blog/2010/02/22/energy-... > The calculation of vacuum energy by QFT isn't infinite. It's just really, really large. This is one of the largest errors between theory and observation in all of physics because the measured energy of a vacuum is ~120 orders of magnitude less. They are infinite if we sum up all frequencies naively. But just like with renormalization, we kinda say handwaveingly that we can probably cut off frequencies at the Planck scale, and then we arrive at this number 120.