Invest1 distinct publisher3 min readUpdated
Three national projects need 24.7 GW of new generation, which the ministry calls 18 nuclear reactors' worth, on top of an April forecast that already assumed 40%-plus demand growth by 2040.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
Korea's Ministry of Climate, Energy and Environment said on the 15th that it will hold the fourth and fifth public policy debates on the 12th Basic Plan for Electricity Supply and Demand on the 20th, at Seminar Room 3 of the National Assembly Members' Office Building, with the fourth session starting at 10 a.m. [1]. The session that afternoon is where a revised 2040 demand forecast gets put on paper [2], and it matters because the plan is the top-tier framework from which generation capacity and transmission build-out follow [3].
The number being replaced was not conservative. In April the ministry put 2040 peak annual demand at 131.8 to 138.2 GW [4], against recent summer peaks of roughly 95 to 97 GW [5], which the ministry itself described as demand jumping more than 40% in about 15 years [6]. Two months after that forecast the government announced three national projects: four semiconductor fabrication plants in the southwestern region, early construction of the Yongin semiconductor cluster, and large-scale AI data centers [7]. The additional generation capacity for those three alone comes to 24.7 GW, which the ministry equates to 18 nuclear reactors of 1.4 GW each [8].
Run the arithmetic and the scale is clearer than the framing. That 24.7 GW is an 18 to 19 percent uplift on the April range [1], and roughly a quarter of everything the country currently draws at summer peak [2]. Added straight onto April's numbers, 2040 peak demand lands between 156.5 and 162.9 GW [3], which is 61 to 71 percent above the current summer peak [4]. The reactor comparison is also rounded up in the government's favour of drama: 18 units at 1.4 GW is 25.2 GW, and 24.7 GW is 17.6 units [5]. None of that yet includes electric vehicles, which the ministry says are spreading faster than expected, or the push to expand heat pumps [9].
The sequencing is the part operators should read closely. Power authorities will draw up the generation capacity plan from the demand figures presented at the hearing [10], and the renewable energy rollout plan through 2040 is unveiled the same day [2]. But how much of the non-renewable demand gets covered by new nuclear reactors is pushed to a separate debate around September or October [11], with the coal phase-out roadmap and electricity market reform each getting their own hearings later [12]. So the demand number is fixed first, the renewables number second, and the firm-capacity question that actually determines whether a fab gets energised is settled last. The plan is normally revised every two years, but the 11th plan was only finalised last year, a year later than scheduled, so the 12th is arriving in close succession [13]. The fourth debate is billed around the era of the "electricity state" ushered in by the AI revolution [14].
Three things to watch. First, whether the published 2040 figure is genuinely additive to April's range or whether the mega-project load is netted against existing assumptions. Second, the transmission and substation component, since the plan explicitly covers network build-out [3] and the 24.7 GW sits in specific clusters rather than across the country [7]. Third, the September-October nuclear session, which is where the gap between the new demand line and the renewables line has to be closed.
Ranked by verification strength, evidence, and original report placement.
The fifth debate, held on the afternoon of the 20th, will revisit power demand projections through 2040 and review the renewable energy rollout plan; the session will unveil an upwardly revised power demand forecast reflecting three major national projects and surging AI demand, and the new forecast is expected to far exceed the earlier projection.
The Ministry of Climate, Energy and Environment said on the 15th that it will hold the fourth and fifth public policy debates on the 12th Basic Plan for Electricity Supply and Demand on the 20th at Seminar Room 3 of the National Assembly Members' Office Building, with the fourth debate beginning at 10 a.m.
The electricity plan is Korea's top-tier framework underpinning mid- to long-term power policy, covering power demand over the next 15 years, plans for the generation capacity needed to meet it, and plans for building transmission and substation networks.
In April the ministry forecast that peak annual power demand in 2040, factoring in population change and economic growth, would reach 131.8 to 138.2 GW.
Recent summer peak power demand in Korea has run at around 95 to 97 GW.
The April projection implies power demand jumping more than 40% in about 15 years.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Concrete official figures, single unverified translated source
The load-bearing numbers (131.8–138.2 GW April forecast, 95–97 GW recent peak, 24.7 GW for three projects, hearing date and venue) are specific and attributed to the ministry, and the article's own notes record a numeric back-check. But the cluster has exactly one publisher, the piece is an automated Korean-to-English translation whose ministry name and byline romanization are flagged unconfirmed, no primary ministry document or third-party grid analysis is cited, and the central claim — the revised forecast — is unpublished.
Institutional process moving, physical capacity not committed
What is observable is procedural: hearings scheduled and dated, a plan cycle running in compressed succession, and mega-projects announced with a stated 24.7 GW requirement that has already forced a mid-cycle recalculation. Nothing in the source shows generation, transmission or renewables capacity actually procured, sited or built, and the nuclear share is explicitly deferred, so adoption is early-stage policy formation rather than deployment.
Framing runs slightly ahead of disclosed numbers
The story's headline tension rests on a forecast that has not been released — described only as 'expected to far exceed' the April range — and on an unquantified EV/heat-pump uplift. The ministry's own 18-reactor equivalence rounds 17.6 units up to 18, and the most dramatic arithmetic (156.5–162.9 GW, 61–71% above today's peak) requires an additive assumption nobody in the source makes. Modestly overstated rather than unfounded, because the 24.7 GW and April baseline figures are real and material.
Same ministry authors the forecast and the plan it justifies
The supplied material itself shows the structural incentive: the ministry issues the demand forecast, power authorities then build the generation capacity plan from the figures presented at that hearing, and the politically contested question of how much goes to new nuclear is deferred to a later session while the renewables plan is showcased now. A higher demand number therefore enlarges the build-out the same institution will authorize. No sponsorship, vendor funding or commercial interest is disclosed in the source, so the read stops at institutional positioning.
Directionally solid, single-source and forward-looking
The documented facts — plan scope, cadence, hearing schedule, April range, current peak, 24.7 GW requirement — are internally consistent and arithmetically checkable, which supports the story's core direction. Confidence is held down by one publisher, an automated translation with self-flagged naming uncertainty, no primary or independent corroboration, and the fact that the decisive forecast and supply mix are still unpublished.
invest
Korea's brokerages added 402 staff in a quarter while mortgages headed to 5.78%1 distinct publisher
invest
A 180kW rack, an 18-month transformer, and a shortfall nobody has shown the math for1 distinct publisher
invest
Bank of Korea says a $936bn swing in net external assets is arithmetic, not a solvency signal1 distinct publisher
invest
K-Beauty's Best Trade Is Now the Factory, Not the Brand1 distinct publisher