There's not enough comparison with the conventional ground-source heat pump here. There's not enough modeling of the expected system dynamics. I don't have a physics argument against it (right now at least) but I think that the author is trying way too hard to sell me on the idea of energy storage and not hard enough on why this proposal can work. And I don't think it's just me. Anyone reading the pitch for an energy storage startup in 2025 is probably aware of the basic goals, and more importantly is fatigued and suspicious after watching several dozen clever ideas go nowhere.

Surely you can write a short model of the system at the level of undergraduate thermo. If you have a pile of dirt this big (say about a thousand times the size of a spherical cow) with these pipes running through it, then at a storage temperature T your capacity is X, your leakage is Y, and your recovery rate is Z. Fill in the blanks.

GSHP/ASHP can't get to steam-generating temperatures on the hot side unfortunately - you're looking at a "heat concentrator" which doesn't really exist in the form of a commercially available heat pump today - possibly not even in the theoretical.

Heat pumps are a nonstarter here, because the goal here is minimizing capex, not maximizing efficiency.