A still fuzzy and arbitrary but imo somewhat elegant cutoff would let feasibility be set by the market and technological development, in the sense of something not being worth throwing compute at _now_ because the timescale of work is slow enough that future classical efficiencies will progress (much) quicker.
The ROI was terrible, but it wasn't mere busywork when Galileo was mucking around with balls and ramps.
You perform busywork, if for marketing purposes usually of the kind or at least set up in a way favorable to your product, and on the presumption that it will in some manner transfer to real work.
The more synthetic the less likely that is, but it's still indicative of something and I don't see what you mean is so different here.
Meaning when supremacy soon is demonstrated even accounting for storage, it should be able to be undone for a while yet by, say, appropriating everyone’s phones and fridge storage, Silicon Valley style.
Edit: Thanks guys for pointing out where I misread the article, leaving mistake intact so the replies make sense.