There are reasons to not take these projections at face value though:
National Grid guy asked how much of this they expected to really come online, ERCOT guy couldn't give a straight answer, but later I thought I heard someone give a historical number of ~30%. They didn't clearly detail on what basis they made this forecast, either.
On the call it sounded like maybe 30% of this comes online.
Papers on DR suggest a DR program has diminishing returns after certain size. If payouts shrink, that becomes a constraint on DR participation, too. I don't think ERCOT is considering the next bitcoin halving either.
the task force was created specifically to address crypto IX's. So there is likely some real data center load in there but i'd wager 80%+ of it is mining
To clarify, this is referring to bitcoin mining occurring on the ERCOT (Texas) grid.
Today, all bitcoin mining combined might be in that approximate range of 17 GW, using CBECI's estimate:
https://ccaf.io/cbeci/index
There are reasons to not take these projections at face value though:
And then there's this:
And this was a surprising take:
The full contents of that Tweet:
The chart is from a document (see below) titled:
Overview of Large Load Interconnection Requests
This Tweet suggests performing word substitution is not entirely wrong:
Source for the chart:
https://www.ercot.com/files/docs/2022/04/25/InterimLargeLoadInterconnectionProcess_AggregateData_04262022.pptx
Which is, with other documents, shared here:
ERCOT Large Flexible Load Task Force
https://www.ercot.com/calendar/event?id=1649295264273