>>14329201There's an addendum to the Second Law of Thermodynamics called the 'Fluctuation Theorem' (FT). The 2nd Law states that entropy of a system always moves towards thermal equilibrium, however the FT states that the 2nd Law is a statistical law which only applies on average over time and length scales which are large relative to the characteristic scales of the system, and that there is a non-zero probability of local, transient movement of part of a system *away* from thermal equilibrium.
For example - if you open a window and let cold air into a hot room, the system will eventually try to move towards thermal equilibrium (everything will try to reach the same temperature), but there is a possibility that you could develop random hot or cold spots as that process happens or even after the room reaches equilibrium. The spots will be small and won't last very long, but those concentrations and gradients in the thermal energy can, in turn, trigger other physical responses.
Same concept for the 'creation' of a universe. You start with some background of matter-energy in a flat spacetime (maybe it's strings, maybe it's something else, don't know, don't care). You have *some* unit of matter-energy that at its basic level can be thought of as positive or negative, up or down, black or white. In most of this infinite expanse matter-energy is in a state of thermal equilibrium, flipping back and forth randomly between those two states (like the image on the left), not producing any net curvature in spacetime, no gradients to drive any physics.
FT implies there is a non-zero probability of everything in a region assuming the same state at the same time (like the image on the right). If that were to happen it would take some time for the system to return to thermal equilibrium, during which those concentrations and gradients in energy density could drive physical responses (like metric expansion, like formation of other types of matter-energy, etc.).
