I think all this model says is the it's currently profitable for bitcoin miners to build their own power stations to run their miners. But if they're doing that

I think all this model says is the it's currently profitable for bitcoin miners to build their own power stations to run their miners. But if they're doing that, why would they contribute to the grid? From looking at the model it just assumes the power will go to the grid first and bitcoin miners second, even though it would always be more profitable to run the miners (by 2x to 3x). I don't see this happening without government intervention. And from the point of view of the human race even this approach involves a huge amount of extra resources going into the mining hardware, an extra cost which only really benefits the miners (extra mining doesn't even benefit bitcoin at this point). I find it extremely worrying that 'bitcoin price' is not an input to this model: it's buried in the assumptions that it basically remains as it is. Bitcoin price going up means even less incentive to contribute to the grid, bitcoin price going down means the benefits it can bring under the broken assumption miners are altruistic go down. The only way I see this actually having a positive impact through market mechanisms is if there's a railroad fever and massive buildout of solar capacity for mining, followed shortly by a bitcoin crash which actually makes that energy available for the rest of us to use. (Also, solar is not a great source for mining to use at the moment: mining hardware still costs an appreciable sum relative to the energy cost of running it, especially considering it depreciates quickly. So if at all possible the miners will run them constantly, not adjust to grid supply vs demand)