It has never panned out. Most recently the company furthest along the licensing process, NuScale, cancelled their project in Utah due to spiraling cost increases, before they even started to build.
THE FORGOTTEN HISTORY OF SMALL NUCLEAR REACTORS
https://spectrum.ieee.org/the-forgotten-history-of-small-nuc...
In both cases, the underlying technology has worked for a while, and the challenges are only tangentially related (NIMBY, lack of production scale, scaremongering for nuclear), but solveable.
All while we have solutions, on the market, that are cheaper, faster and easier to deploy.
The costs of the French nuclear scale-up: A case of negative learning by doing
https://www.sciencedirect.com/science/article/abs/pii/S03014...
This study does not include their most recent attempt at Flamanville.
> Construction on a new reactor, Flamanville 3, began on 4 December 2007. The new unit is an Areva European Pressurized Reactor type and is planned to have a nameplate capacity of 1,650 MWe. EDF estimated the cost at €3.3 billion and stated it would start commercial operations in 2012, after construction lasting 54 months. The latest cost estimate (July 2020) is at €19.1 billion, with commissioning planned tentatively at the end of 2022.
https://en.wikipedia.org/wiki/Flamanville_Nuclear_Power_Plan...
If everything was rosy in the renewables, then why not. But we don't have storage, we don't have high-capacity long-range (thousands of kms) energy network, most of the production capacity is in China ...
Both directions have its risks, I think a wise choice would be to pursue both to avoid putting all eggs into one basket.