Ontario Power Generation’s final investment decision for the first BWRX-300 at its Darlington site put the cost of that single 300-megawatt unit at roughly 7.7 billion Canadian dollars, about 5.6 billion US dollars, close to 18,700 dollars per installed kilowatt before any of the standardization and repeat-build savings a European fleet is meant to eventually deliver. That is the only real-world cost figure available for the reactor design SGE, GE Vernova, Hitachi and Samsung C&T agreed this week to jointly advance across Central and Eastern Europe and the United Kingdom.
The memorandum signed on the sidelines of the United Nations General Assembly commits the four companies to assess commercial opportunities and refine market strategies for the BWRX-300, combining GE Vernova and Hitachi’s reactor design, Samsung C&T’s international construction capacity, and SGE’s position as exclusive regional development lead across Central and Eastern Europe. SGE’s most cited evidence of momentum, a Polish decision in principle covering 26 BWRX-300 units, describes a considerably earlier stage of commitment than the figure alone conveys. That decision, initially issued in December 2023 for up to 24 SMRs across six potential sites and now cited by SGE and its partners as covering 26 units, is a preliminary governmental endorsement of siting and technology selection, not a construction license, a binding funding commitment, or a firm equipment order. Of those 26 units, only 14, spread across three specific Polish locations, have reached the stage of Orlen Synthos Green Energy actually applying to Poland’s energy ministry for a Contract for Difference subsidy, a step the company itself describes as “the first phase” of the broader 26-unit ambition, with the first unit near Wloclawek targeted for completion in 2032, six years from now.
SGE’s parallel UK proposal uses an identical number. The company and a partner team including Samsung C&T, Laing O’Rourke, Aecon Group and Google Cloud have outlined plans for 14 BWRX-300 units across three UK sites under the country’s Advanced Nuclear Framework programme, a proposal still under government consideration rather than an approved program. Two separate national fleets, in two different regulatory and grid environments, arriving at the same 14-unit, three-site structure suggests SGE is working from a standardized fleet template adapted to each market’s framework rather than each figure reflecting country-specific grid or siting analysis independently arrived at.
The genuine, verifiable progress behind the European push sits entirely at Darlington. Construction began in April 2026 following a construction license the Canadian Nuclear Safety Commission issued in April 2025 and Ontario’s provincial approval the following month, and the project has since cleared documented milestones, including a regulatory hold point on the reactor building foundation removed in March 2026 and the arrival and completion of a purpose-built tunnel boring machine’s journey to the site. Ontario Power Generation applied for a 20-year operating license in March 2026, well ahead of the first unit’s expected completion by the end of the decade, standard practice for a project this far along but also a sign of how much regulatory runway a Western SMR project requires even after clearing its construction license. GE Vernova Hitachi’s own description of Darlington as providing “the confidence that comes from real project execution” is a fair characterization of that documented progress, distinct from claims about projects still at the decision-in-principle or proposal stage elsewhere.
What Darlington does not yet demonstrate is the cost profile a standardized European fleet would need to compete on. The first unit’s 7.7 billion Canadian dollar cost sits well above the per-kilowatt figures typically cited in SMR industry marketing, and while Ontario’s approval of three additional units at a combined estimated 13.2 billion Canadian dollars, roughly 4.4 billion each, shows a real cost reduction of about 43% per unit from the first-of-a-kind build to repeat units on the same site, the blended cost across all four Darlington units, roughly 20.9 billion Canadian dollars for 1,200 megawatts of combined capacity, still works out to more than 12,600 US dollars per kilowatt. That figure describes the most mature BWRX-300 deployment anywhere, built by the same design and construction partners now proposing to replicate the model across multiple European countries. Whether a fleet spanning Poland, the UK and other Central and Eastern European markets achieves costs meaningfully below what Darlington’s own repeat-unit data shows, rather than simply replicating that cost structure at a larger geographic scale, is the specific economic question this week’s memorandum defers to the “commercial opportunities” assessment it commits the four companies to carrying out next.
