I don't know why this isn't done more in the US. At least with thermal water, my understanding is that you don't need huge temperature differentials and this is cost effective on a pretty small scale.
And every gigajoule you take from Yellowstone is one less gigajoule for the next supervolcano eruption.
There are approximately twenty supervolcanoes around the world, so I wouldn't worry to much. There is an almost supervulcano in Italy that is also much closer to where power is needed [0].
Remoteness is not that much of a problem though: (ultra) high-voltage direct-current ("(U)HVDC") power lines have losses in the order of 3% per 1000km (that's 620mi) which is very acceptable. China has power lines that move the power equivalent of several nuclear power plants over thousands of kilometers for example.
[0] https://en.wikipedia.org/wiki/Phlegraean_Fields
[1] https://en.wikipedia.org/wiki/High-voltage_direct_current
And the Mediterranean can provide a lot of water for cooling at the same time the extra heat can easily be used for desalination.
Edit, as reply to child comment: here is the NASA report that concludes that it is possible to cool supervolcanoes:
https://scienceandtechnology.jpl.nasa.gov/sites/default/file...
Multiply any activity by a non trivial fraction of humanity and you get non trivial side effects.
We don't need to make a mess in Yellowstone, but 'massive energy loss by resistance' is just wrong.
The majority of generating losses (about 60% of input thermal energy) is due to Carnot efficiency losses, not transmission inefficiencies. There's also some loss in transformers (ramping voltages up or down), and rectifiers (converting AC to DC and vice versa).
But the biggest losses are in going from thermal to mechanical energy itself.
Keep in mind that the same general region is a major coal-producing zone presently, and much of that coal is burned locally for generation: it's cheaper to move the electricity than the coal used to generate it.
This also means that a substantial amount of the transmission infrastructure is already in place.
What if it disturbs whatever equilibrium the caldera's maintained and causes it to erupt sooner than it would have?
This is a genuine question, I know next to nothing about this.
We are not good enough in vulcanology yet to tickle the sleeping Yellowstone giant.
If we could take enough energy out of Yellowstone to make it less supervolcanic, we’d have a hard problem dealing with whatever waste heat would be left after consuming that much energy.
If a lot of extraction is done, geysers and other surface geothermal activity reduce or stop. Rotorua, New Zealand had these issues back in the 1980s when everyone had a private bore. Blocking them up helped a lot.
https://www.geothermal-energy.org/pdf/IGAstandard/Japan/1997...
You might poke a hole and get something hot and nasty gushing out but that would only ruin some machinery and perhaps the day for a few people.
However there are ways to mitigate the uncertainty. You can start small, observe, and if still stable, expand. I believe that is how most new geothermal plants are done today in Iceland.
Maybe increasing automation means that we can build more factories and data centers near these sources of massive renewable energy potential without struggling to convince workers to live in these remote locations (although that might not be a huge challenge for a beautiful place like Yellowstone).
It's far less expensive to build one high-capacity line than many low-capacity lines (e.g., serving wind farms or solar farms).
Geothermal has the added benefit that it's dispatchable power. It can be extracted on demand, when the need arises. As a complement to wind and solar, it's actually a great fit.
The grid doesn't simply want megawatts. It wants a balance of supply and demand, and a minimisation of outages.
Wind and solar cannot be spun up on demand. Hydro can be, and in general is (much lower-capacity hydro is grid smoothing). Hydro will even suck capacity off the grid where possible (pumped storage).
Geo which can load-match, disable when grid is oversupplied, and fill in for cloudy & windless days, makes all kinds of sense at the grid level.
With geothermal, though, you don't actually want to run flat-out all the time, as the geothermal reservoir does require time to recover, whether that's groundwater refresh or heating up after a period of high generation which extracts heat faster than it is transmitted through surrounding rock (most cases) or magma (not yet a major utilisation mode).
Soaking up excess generating capacity through interruptable heavy loads is somewhat better suited to wind or solar. If you have a task you can divert electricity that will be generated anyway to, so much the better. That's what pumped-hydro, desalination, grid-scale battery banks, or fuel synthesis represent.
I suspect the amount of energy involved in a volcano eruption can't possibly be compared to the energy you extract as geothermal.
Tree roots mess with the piping, and you have to be careful to not drive or set anything over it (much like a septic drain field, though not as fragile).