Roger Penrose (along with two other co-authors) recently released a paper on the arXiv which proposes a test of quantum gravity with a Bose-Einstein condensate acting as a very massive 'particle'. The aim is to investigate some interactions and hopefully rule out some theories of quantum gravity.
One theory they consider throughout the papers is Gravity-induced Quantum state reduction (GQSR). It seems like the idea is to keep gravity a classical theory, and have gravitation interactions interaction with a wavefunction, causing the wavefunction to evolve in a non-unitary way.
It didn't take me while before I noticed that such a theory implies that information is not conserved -- that's pretty much the definition of a unitary evolution. It seems to me that if we lose information conservation, energy conservation also falls (the argument would we a Szilard's engine style argument).
I wouldn't really take seriously a theory that doesn't agree with information and energy conservation, so why is this a viable or interesting theory, and how does it deal with this glaring problem?
