Korea's SMR review standards trail the first permit filings by four years

The commission's staffing plan shows how a 2032 rulebook turns into a queue. It estimates that each reactor design needs 15 to 20 reviewers ^[11], and it wants 33 SMR staff in 2027 ^[12]. Subtract the 18 already working on the i-SMR ^[10] and 15 people are left for the three designs due for prereview that year ^[2]. Assuming a prereview needs the same headcount as a full review, that is about five per design, a third to a quarter of what the commission says each design needs ^[21]. The target peaks at 39 in 2029 and 2030 ^[12]. That covers about two designs at once ^[22], against nine applications planned for 2027 and 2028: eight prereviews plus the i-SMR construction permit ^[23].

None of those hires is secured. Talks between ministries on the staffing have been pushed back to the second half of this year ^[13]. The US Nuclear Regulatory Commission has turned its advanced reactor unit into a full office with 310 to 330 staff ^[14], roughly eight times Korea's peak target ^[24]. Britain's Office for Nuclear Regulation is training more than 30 inspectors for its Advanced Nuclear Technologies team while it handles six designs ^[16].

The rulebook will take longer to fix than the staffing. The commission plans a technology-inclusive, risk-informed and performance-based review method, with 2032 as the target date ^[7]. Companies filing for construction permits from 2028 would likely wait three to four years for it ^[8], and TerraPower and other developers plan to file in 2028 and 2029 ^[5]. In March the NRC issued TerraPower a construction permit for Kemmerer Unit 1 in Wyoming, its first approval of a commercial non-light-water reactor in more than 40 years ^[15].

"Our SMRs are recognized worldwide for their technology, but they are stuck at the threshold of moving on to demonstration and construction," Hwang said ^[19]. "To review them on time while keeping safety intact, the regulator needs the expertise and the staff so that the critical window for commercializing SMRs is not lost." ^[20]

The calendar could break three ways. If the staffing talks deliver, the 2027 filings could proceed before final standards exist, because a prereview is a check of documents for what an application still needs ^[3]. Alternatively, the commission could finish the i-SMR first and take other designs in turn, as its current allocation of reviewers already implies ^[10]. The third possibility is that the 2032 target slips.

I'd expect the second, and its cost falls on non-light-water entrants. TerraPower's US permit was for that kind of reactor ^[15], and that kind is the hardest to fit into a rulebook written for light-water plants ^[6]. The counter-case is that prereview is paperwork and the permit filings come a year or two later ^[2][5], so a staffing deal this year could narrow the gap to whatever interim review the commission can run. This view would be wrong if a signed agreement for the 33 staff and a published interim route for non-light-water designs both arrive before the 2028 filings. The reporting does not include what SK Innovation or Korea Hydro & Nuclear Power has committed to the venture ^[4], so the cost of a three-to-four-year wait cannot yet be put in won.