1. Overview
Just about the only thing we can be sure of concerning energy is that it is conserved. It can neither be created nor destroyed; energy can only be transformed between the various forms we are all familiar with. A consequence of this is that the total energy in the universe should be unchanged over time. For this reason, the natural place to look for an understanding of energy is in cosmology, the branch of science concerned with the origin and development of the universe, and we can explore specifically the role of energy within established cosmological theories.
The state-of-the-art is a very impressive cosmological model called the consensus model (also referred to as the lambda-CDM model) which presents a picture of the universe that is in excellent agreement with observational data. What is even more impressive is that there is agreement not just with the way the universe is today but also as it was in the past (going all the way back more than 13 billion years). Data spanning the lifetime of the universe is available to us because the finite speed of light shows distant objects at the much earlier time when the radiation that is received now was emitted. Overall, the consensus model fits the historical data extremely well.
That is not to say the model is perfect - with more detailed data emerging there are discrepancies that are small but which cannot easily be resolved. This is because most of the flexibility in the model has already been taken up and there is little scope for a further simple adjustment that will not mess up other predictions. Amusingly, and rather Trump-like, cosmologists play down the issue and refer to 'tension in the theory' rather than admitting there is a real problem. Have a look at this paper by [Lin, Mack and Hou] for more information.
The physical basis for the model is pretty convincing. The special theory of relativity and the general theory (both undoubtedly correct) are fully incorporated. The expansion of the universe naturally emerges, an expansion which is certainly occurring.
But what does the model say about energy? This is problematic. Whilst energy conservation appears to be the basis for the model, there are certain areas where energy conservation is not absolutely demonstrated. For example, distant photons appear to lose energy as they travel. The standard answer from cosmologists to questions about energy non-conservation is, 'energy is not conserved in general relativity' (and the consensus model is a general relativistic theory). A strange response, because general relativity as it is normally applied does conserve energy. Have a look at the Johnson-like contortions performed by [Sean Carroll] as he patronisingly explains that non-conservation of energy does not actually mean that energy is not conserved but that it something to do with space-time that only an expert is capable of understanding!
This is a little alarming as the only thing we were certain about concerning energy is now undermined. Consequently, the accepted cosmological model merits a closer look and cannot just be assumed to be correct.