I believe the thought is that the cost of regulatory and safety measures do not scale favorably with reactor size. You don't have separate facilities so much as 10 reactors on the same site, so you don't need to increase personnel 10x.

Actually proving this with a deployed design seems to be the challenge, though with strong headwinds against nuclear across the board, favoring proven designs over newer ones is at least a little faster. That's not a flaw of the design, but intrinsic to the social and political environment.