Nuclear Malaysia

Malaysia has no nuclear power plant, no nuclear weapons, and no intention of acquiring either in the near term. What it does have is one small research reactor, a mature regulatory body, decades of nuclear medicine and industrial use, and, since the 13th Malaysia Plan (2026-2030), a formal government decision to study whether nuclear power belongs in the national grid.

That last point is what changed. For twenty years nuclear power in Malaysia was a recurring proposal that never survived contact with politics. The 13MP put it back on the table as a named strategy, and the Ministry of Energy Transition and Water Transformation (PETRA) followed by launching a feasibility study. Cabinet's position is that nuclear is one of the options for power generation after 2035, with officials describing a ten to fifteen year horizon.

So the honest summary is: a decision to study, not a decision to build. No site has been chosen, no reactor technology selected, and no construction approved. Anyone telling you Malaysia is building a nuclear plant is ahead of the evidence.

This guide covers four separate things people tend to conflate:

• Nuclear power, which Malaysia is studying
• Nuclear weapons, which Malaysia has renounced by treaty in about as many ways as a country can
• Nuclear technology in daily use, which is already everywhere in Malaysian hospitals and industry
• Radioactive waste and public trust, which is where Malaysia's real nuclear history lives, and it is not a happy one

Malaysia has been here before. A nuclear plant was floated in the 2010s, targeted for around 2021, and shelved after Fukushima and sustained public opposition. Understanding that history is why the current process looks deliberately slow.

What is decided:

• The 13th Malaysia Plan names nuclear as a potential clean electricity source under its energy strategy.
• PETRA has launched a feasibility study.
• MyPower, an agency under PETRA, has been appointed the Nuclear Energy Programme Implementing Organisation (NEPIO), the body the IAEA expects a country to create before it commits to anything.
• Malaysia is following the IAEA Milestones Approach, a three-phase framework that runs from considering a programme, to preparing for it, to actually building. Malaysia is early in that process.

What is not decided: site, reactor type or size, financing model, whether it will be conventional large reactors or small modular reactors (SMRs), who would operate it, and ultimately whether it happens at all.

Why it is being considered now. The National Energy Transition Roadmap commits Malaysia to 40% renewables by 2040, 70% by 2050, a full coal phase-out by 2045, and net zero by 2050. Meanwhile electricity demand is climbing steeply, particularly from data centres in Johor and the Klang Valley. Solar and hydro alone struggle to supply the round-the-clock baseload that grid operators need, and gas is a transition fuel rather than a destination. Nuclear is being examined as the one proven zero-carbon baseload option.

The counterarguments are serious and are not fringe. Capital costs are enormous and nuclear projects overrun famously. Malaysia has no trained operating workforce and would need a decade to build one. The country sits close enough to seismic activity to require careful siting. And there is no permanent high-level waste solution anywhere in the world, let alone here. Nuclear's advocates and critics in Malaysia are arguing about real things.

Pro Tip: When you read a headline about Malaysian nuclear power, check which of three verbs it uses: studying, deciding, or building. As of 2026 the honest verb is studying, and the gap between that and building is measured in a decade or more.

The argument for nuclear in Malaysia is not that it is cheap. It is that it is firm: a reactor produces the same output at 3am in a downpour as it does at noon. The argument against is that solar keeps getting cheaper and can be built now. Those two claims talk past each other unless you put a number on the gap.

So here is ours. The Baseload Replacement Ratio is how much Malaysian solar it takes to replace one megawatt of nuclear running round the clock.

A modern reactor runs at roughly a 90% capacity factor over its life. Malaysian solar runs at roughly 16%, from a specific yield of about 1,400 kWh per kWp per year, the same figure behind our residential solar payback maths. That alone is a factor of 5.6. Then storage has to carry the roughly fourteen hours a day the panels are not producing at full output, which is the part usually dropped from the comparison.

Use the calculator below to run it at any size, including SMR scale.

What the ratio is for: it makes the physical trade explicit. Nuclear buys firmness and buys back land. Solar buys speed and optionality and spends land. Neither of those is the cost argument, and the cost argument is the one that usually decides these things.

What the ratio is not: a case for either side. Batteries and panels have fallen in price nearly every year for a decade while nuclear construction costs have generally gone the other way, so a ratio that is unflattering to solar on land and capacity can still be the cheaper answer. It also assumes solar plus batteries has to replace nuclear one for one, when in a real grid gas, hydro, imports and demand response all do part of the job.

Pro Tip: Whenever someone quotes a solar capacity figure against a nuclear one, check whether they used capacity or output. Comparing 1,000 MW of solar to 1,000 MW of nuclear overstates solar by about five and a half times in Malaysian conditions, and it is the single most common sleight of hand in the debate.

No costed proposal exists for Malaysia, because nothing has been proposed. What exists is international experience, and it is not encouraging on price.

Recent Western builds have landed between roughly US$6,000 and US$12,000 per kilowatt of capacity, against vendor projections that were usually half that. Korean and Chinese builds have come in materially lower, in the US$2,500 to US$4,000 range, which is why any Malaysian programme would look hard at those vendors. The honest planning range for a first-of-a-kind plant in a country with no nuclear workforce sits at the higher end.

The number that matters for a household is not the capital cost, it is the tariff. Because nuclear's cost is dominated by capital and the cost of borrowing, the financing model decides almost everything. The same reactor financed by the state at sovereign rates and financed by a private consortium at commercial rates produce very different electricity prices. This is why the financing question, not the technology question, is the one to watch.

And the cost that never appears in a project budget: delay. Every year of overrun is a year of interest on money already spent with nothing being generated. That mechanism, more than any safety issue, is what has made recent Western nuclear projects go wrong.

Pro Tip: Treat any SMR cost per kilowatt quoted today as a projection, not a price. Almost none are operating commercially, so the figures come from vendor models rather than from built plants, and first-of-a-kind projects have a long record of exceeding them.

Malaysia does not have nuclear weapons, has never pursued them, and has bound itself against them through effectively every available treaty. This is one of the clearest positions in Malaysian foreign policy.

The Bangkok Treaty is the regionally distinctive one. It makes the whole of Southeast Asia, including territorial seas and exclusive economic zones, a nuclear-weapon-free zone. That EEZ coverage is precisely why the nuclear-weapon states have been slow to sign its protocol: it would constrain their naval movements through waters they consider international. Malaysia has been a consistent advocate for getting them to sign.

Malaysia also votes reliably for disarmament resolutions at the UN General Assembly and has historically been among the more vocal non-aligned voices on nuclear disarmament.

The proliferation episode worth knowing. Malaysia's one brush with the weapons world was not a weapons programme. In 2004 it emerged that a Malaysian company, SCOMI Precision Engineering, had manufactured centrifuge components that reached Libya through the A.Q. Khan proliferation network. A police investigation concluded the Malaysian firm had been given a civilian cover story and did not know the end use, and no charges were brought in Malaysia. The episode pushed Malaysia to tighten export controls, which eventually produced the Strategic Trade Act 2010 governing dual-use goods.

Two bodies are routinely confused, and the difference matters.

The regulator is the Atomic Energy Licensing Board (AELB, Lembaga Perlesenan Tenaga Atom), operating as the Department of Atomic Energy (ATOM Malaysia) under the Ministry of Science, Technology and Innovation (MOSTI). It licenses, inspects and enforces. If you own equipment containing a radioactive source, you answer to them.

The operator and researcher is the Malaysian Nuclear Agency (Agensi Nuklear Malaysia), also under MOSTI, based in Bangi, Selangor. It runs the research reactor, does the R&D, and provides irradiation and analysis services. It does not regulate itself; ATOM Malaysia regulates it.

The governing law is the Atomic Energy Licensing Act 1984 (Act 304), supported by regulations covering basic safety standards, radioactive waste management, transport and licensing. It is the reason a dental clinic's X-ray machine, a factory's thickness gauge and a hospital's cancer therapy unit all sit inside the same licensing regime.

A point observers raise consistently: Act 304 was written for research, medicine and industry. A power reactor is a different regulatory animal, requiring an independent, well-resourced regulator with the authority to halt a project backed by the state. Strengthening that framework is widely treated as a precondition for any power programme, and it is one of the IAEA milestones Malaysia would have to clear.

Pro Tip: If you need to verify whether an operator is licensed, or import anything containing a radioactive source, ATOM Malaysia is the authority, not the Nuclear Agency. Sending your application to the wrong one is the most common administrative mistake in this area.

Malaysia has exactly one nuclear reactor, and most Malaysians have never heard of it.

The PUSPATI TRIGA Reactor (RTP) at the Malaysian Nuclear Agency in Bangi is a 1 megawatt TRIGA Mark II research reactor. It first went critical on 28 June 1982 and has run for over four decades. TRIGA stands for Training, Research, Isotopes, General Atomics, which describes what it does.

To be clear about scale: at 1 MW thermal it is roughly a thousandth the size of a commercial power reactor, and it generates no electricity. TRIGA reactors use a fuel design that becomes less reactive as it heats, an inherent safety characteristic that is why this reactor type sits on university campuses worldwide.

What it is actually used for:

• Producing medical radioisotopes and research isotopes
• Neutron activation analysis, used in forensics, archaeology, geology and environmental testing
• Training the small pool of Malaysians with hands-on reactor experience
• Neutron radiography and materials research

That training function is the strategic part. A country cannot assemble a nuclear operating workforce on demand, and RTP is where Malaysia's existing nuclear expertise was largely built. It is also ageing, and its future is a live question inside the sector.

The debate about power plants obscures how much nuclear technology Malaysia already uses daily, safely and unremarkably.

Medicine. Nuclear medicine departments in major Malaysian hospitals use radioisotopes for both diagnosis and treatment: PET and SPECT scans for cancer staging and cardiac work, radioactive iodine for thyroid disease, and radiotherapy for cancer treatment. If you or a family member has had cancer care in Malaysia, nuclear technology was almost certainly part of it.

Industry. Non-destructive testing uses gamma sources to inspect pipeline and structural welds without cutting them open, which matters enormously in oil and gas. Nucleonic gauges measure thickness and density on production lines. Well logging uses radioactive sources downhole.

Agriculture and food. The Nuclear Agency has used irradiation for decades to extend shelf life and control pests, and mutation breeding, using radiation to accelerate natural variation, has produced Malaysian rice varieties. This is conventional agricultural science, not genetic modification.

Research. Neutron activation analysis supports environmental monitoring, forensic work and archaeology.

The reason this matters to the power debate: Malaysia is not starting from zero. It has licensed operators, a regulator with real inspection experience, and radiation-safety professionals. That is a genuine foundation, though a much smaller one than a power programme requires.

Any honest discussion of Malaysian nuclear policy has to start here, because this is what the public remembers.

In 1982 Asian Rare Earth Sdn Bhd, a venture involving Mitsubishi Chemical, began processing tin-tailings monazite at Bukit Merah, Perak, to extract yttrium. Monazite is naturally rich in thorium, so the process produced radioactive residue, and that residue was handled and stored badly, close to a residential community.

What followed became one of Asia's defining environmental cases. Residents documented birth defects, and a cluster of leukaemia cases in a community of roughly 11,000, with deaths attributed by campaigners to the contamination. Residents sued. The plant closed in 1994. The cleanup, widely described as the largest radioactive waste cleanup in Asia, cost on the order of US$100 million and was formally completed around 2011, with the waste entombed in a permanent repository that remains monitored today.

The legacy is trust. When the Lynas rare earth processing plant opened in Gebeng, Kuantan in 2012, opposition was immediate and drew explicitly on Bukit Merah. That dispute over residue storage has continued for over a decade. Our rare earth guide covers the Lynas story and Malaysia's critical minerals position in detail.

Why this matters for nuclear power: Bukit Merah was not a nuclear power accident, it was a rare earth refinery. But in public memory it is the answer to the question can Malaysian institutions be trusted to manage radioactive material honestly over decades? Any nuclear programme has to answer that question, and pointing out that the cases are technically different is not an answer.

Pro Tip: If you are researching this topic, keep the three cases distinct. Bukit Merah (1982-1994, closed, cleaned up) is a rare earth refinery. Lynas Gebeng (2012 to present, operating) is a rare earth refinery. Neither is a nuclear power plant, and conflating them is the single most common error in Malaysian nuclear commentary, in both directions.

Most of the current interest is in small modular reactors (SMRs), typically under 300 MW per unit and factory-built rather than constructed on site. The appeal for a country like Malaysia is lower absolute capital cost, a smaller grid footprint, and shorter build times.

The caution worth carrying: as of 2026 very few SMR designs are operating commercially anywhere. Much of what is marketed is a design on paper with a projected cost, and first-of-a-kind nuclear projects have a long history of costing more and taking longer than projected. A country buying an SMR in the near term is buying into an immature market.

Regionally, Southeast Asia is moving on this together. The Philippines has an established programme and is working toward restarting or replacing capacity. Indonesia and Vietnam have both revived nuclear plans. Singapore, constrained on land, has studied advanced reactors. This matters for Malaysia in a specific way: a serious regional accident anywhere in ASEAN would end the domestic conversation, and conversely, a neighbour operating safely makes the Malaysian debate easier.

The SEANWFZ treaty does not prohibit any of this. It bans nuclear weapons, and explicitly preserves the right to peaceful nuclear energy under IAEA safeguards.

Nuclear is a topic where confident wrong statements are common in Malaysian public discussion, on both sides.

Rookie mistakes

1. Saying Malaysia is building a nuclear plant. It is running a feasibility study, with Cabinet framing nuclear as an option after 2035. Study is not construction.
2. Confusing Bukit Merah and Lynas with nuclear power. Both are rare earth refineries dealing with naturally occurring radioactive material. Neither is a reactor.
3. Confusing AELB with the Nuclear Agency. One regulates, the other operates and researches. Sending a licensing question to the wrong one wastes weeks.
4. Assuming the Bangkok Treaty blocks nuclear power. It bans nuclear weapons and explicitly protects peaceful use under safeguards.
5. Treating SMRs as proven and cheap. Very few are operating commercially anywhere, and the quoted costs are largely projections.
6. Assuming Malaysia has no nuclear experience. It has run a research reactor since 1982 and uses radioisotopes routinely in medicine and industry.
7. Citing the old 2021 target as though it is live. That plan was shelved after Fukushima. Quoting it as current is a decade out of date.
8. Assuming a plant would go in Peninsular Malaysia by default. No site has been selected, and siting is one of the hardest and most contested parts of any programme.
9. Believing waste is a solved problem. No country operates a permanent high-level waste repository at scale. That is a genuine open issue, not anti-nuclear rhetoric.
10. Dismissing public opposition as ignorance. Given Bukit Merah, scepticism about institutional handling of radioactive material is an evidence-based position.

Pro tips

• Pro Tip: Go to primary sources. The IAEA publishes country profiles, ATOM Malaysia publishes the licensing framework, and the 13th Malaysia Plan states the policy directly. Malaysian nuclear coverage is unusually prone to recycled claims.
• Pro Tip: Watch MyPower, not the ministry. As the appointed NEPIO, it is where actual programme progress will be visible first.
• Pro Tip: Track the IAEA Milestones phases. They are the honest measure of how far along a country is, and they cannot be skipped by announcement.
• Pro Tip: If you work with radiation sources professionally, licensing runs through ATOM Malaysia under Act 304, and it applies to far more equipment than people expect, including many industrial gauges and clinic X-ray units.
• Pro Tip: Separate the cost argument from the safety argument. They are independent. A reactor can be safe and still be a poor financial decision, and most serious critiques of nuclear in Malaysia are economic rather than safety-based.

Concrete markers that would tell you the programme is real, roughly in the order they would appear:

1. The feasibility study reports, and its findings are published rather than summarised.
2. Legal reform begins. A power programme needs a regulator with independence and authority beyond what Act 304 provides. Watch for amendments or a new act.
3. Site studies start. Seismic, cooling water, population density and grid connection work. This is where the politics gets real, because every candidate site has residents.
4. A technology decision. Large conventional reactors and SMRs imply different costs, timelines and vendor relationships.
5. Workforce building. Scholarships, university programmes, secondments to operating countries. A programme without this by the early 2030s is not serious.
6. A financing model. Nuclear's cost is dominated by capital and the cost of borrowing. How it is funded determines whether it is viable.

And the things that could stop it: a change of government, a regional nuclear incident, a shift in the wider environmental and climate policy settlement, SMR costs failing to materialise as promised, or renewables plus storage becoming cheap enough that the baseload argument weakens.

The realistic reading in 2026 is that Malaysia has taken a genuine institutional step, appointing a NEPIO is not a gesture, while remaining a long way from commitment. Treat 2035 as the earliest plausible date for operation rather than a plan, and expect the timeline to move.