UK nuclear industry leaders unite in response to Chancellor’s call for sector-wide culture reset

Leaders from across the UK nuclear industry have come together to set out how the sector will drive a fundamental change in culture,
.
in response to an open letter from the Chancellor calling for a sector-wide reset.
.
Office of Nuclear Regulation , 12 November 2026
The joint response, sent on behalf of the newly formed Nuclear Industry CEO Forum, builds on individual replies already issued by organisations including the Office for Nuclear Regulation (ONR), and marks the first time industry and regulators have come together to respond collectively to the Chancellor’s call for reform.
The Chancellor’s letter, issued in response to the Nuclear Regulatory Review 2025, asked regulators and industry leaders to treat organisational culture as a strategic priority, and to report back on how they are addressing behaviours such as excessive risk aversion, proceduralism and a reluctance to challenge established ways of working.
In its joint reply, the CEO Forum notes how the UK is entering the largest expansion and renewal of its nuclear sector in decades, and that the industry’s current approaches to regulation and delivery need to evolve further to keep pace with the country’s needs.
Without urgent, system-wide reform, they warn, ‘the UK risks an energy security gap, diminished defence capability, escalating costs and the loss of a once-in-a-generation economic opportunity.’
Changes already underway across the sector include:
moving from resource-intensive periodic reviews to continuous assurance,- the safe application of digital technologies and AI to improve decision-making
- streamlining engineering and safety processes to remove duplication
- strengthening early collaboration between industry and regulators to resolve technical issues, reduce uncertainty and support innovation without compromising standards.
To sustain this momentum, the sector has established the Nuclear Industry CEO Forum, led by the ONR, which will develop and share case studies of effective culture change, capture lessons learned, and build confidence in new ways of working.
Leaders said they would hold themselves personally accountable for the transformation, fostering an environment where people are supported to challenge established practice and make proportionate decisions……………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………………..
Nuclear Industry CEO Forum – Co-signatories
- AWE Nuclear Security Technologies
- Babcock International Group including Cavendish Nuclear
- BAE Systems Submarines
- Great British Energy – Nuclear
- NNB Genco (HPC)
- Nuclear Decommissioning Authority
- Nuclear Restoration Services
- Rolls Royce SMR
- Rolls Royce Submarines
- Sellafield Limited
- Sizewell C Limited
- UK National Nuclear Laboratory
.
Fitch Ratings Revises Sizewell C Limited’s Outlook from Stable to Negative

Fitch Ratings – London – 10 Nov 2026:
Fitch Ratings has revised Sizewell C Limited’s (SZC) Outlook to Negative from Stable, while affirming its Long-Term Issuer Default Rating (IDR) at ‘BBB+’. The rating action is a direct result of the revision of the Outlook of Electricite de France (EDF) to Negative from Stable (IDR: BBB+), in turn related to the sovereign action on France (A+/Negative).
During the construction period, SZC relies heavily on Framatome, which is contracted to provide the nuclear island. This caps the rating of the company at the level of EDF, as EDF guarantees Framatome’s (capped) financial obligations towards SZC under its contract.
KEY RATING DRIVERS
For an overview of the issuer’s credit profile, including key rating drivers, see “Fitch Affirms Sizewell C Limited at ‘BBB+’; Outlook Stable” published 20 October 2026 on www.fitchratings.com.
KEY RATING DRIVERS
For an overview of the issuer’s credit profile, including key rating drivers, see “Fitch Affirms Sizewell C Limited at ‘BBB+’; Outlook Stable” published 20 October 2026 on www.fitchratings.com.
RATING SENSITIVITIES
Factors that Could, Individually or Collectively, Lead to Negative Rating Action/Downgrade
– Negative rating action on EDF as guarantor of Framatome
– The projected Fitch Rating Case (FRC) net debt/regulatory asset base (RAB) above 65%, which gives the Secretary of State the right to suspend elements of the UK government’s (AA-/Stable) support package…………………………………………………………………………………………………………………………………………………………. https://www.fitchratings.com/research/infrastructure-project-finance/fitch-revises-sizewell-c-limited-outlook-to-negative-affirms-rating-at-bbb-10-11-2026
Turkey, AtkinsRealis sign deal for study on possible nuclear plant

Reuters 13 Nov 26, https://www.theglobeandmail.com/business/article-turkey-atkinsrealis-sign-deal-for-study-on-possible-nuclear-plant/
AtkinsRealis ATRL-T said on Thursday it had signed an agreement with Turkish state-owned nuclear energy company TUNAS to conduct a feasibility study for a potential expansion of Turkey’s nuclear power program.
The company said in a statement that it would carry out a comprehensive study to be completed by the end of 2027 for a potential multiunit nuclear project, as well as front-end engineering and design (FEED) work.
Turkey is looking to expand its nuclear program beyond the 4.8-gigawatt Akkuyu plant under construction by Russia’s Rosatom in the Mediterranean province of Mersin, with sites in the Black Sea province of Sinop and the northwestern Thrace region identified as potential locations for new plants.
The Akkuyu project – Turkey’s first nuclear power plant – has faced repeated delays, with Rosatom now expecting the first reactor to achieve its first self-sustaining nuclear chain reaction before the end of this year.
Joe St. Julian, president of nuclear at AtkinsRealis, told Reuters that the parties were currently discussing a six-unit plan for a 4.5-gigawatt plant in Turkey, adding that they planned to complete the engineering design study by 2028, with that leading to a final investment decision.
There was also agreement on a road map for financing the project, St. Julian said.
The agreement was signed by AtkinsRealis subsidiary Candu Energy on the sidelines of the COP31 climate summit in the southern Turkish province of Antalya.
Ankara has said it was in talks with Canada, South Korea, Russia, Japan and China on a potential plant. It has signed memorandums of understanding with Canada and South Korea on nuclear co-operation, with financing and costs being a key factor.
Last week, unveiling the country’s national energy plan, Energy Minister Alparslan Bayraktar said Ankara aims to raise nuclear capacity to 10 GW by 2040 and expects nuclear power to meet 11 per cent of electricity demand by then. He has also described 2027 as a decision year for plans in Sinop or Thrace.
72.8 billion euros according to EDF, 680 billion according to Greenpeace: what is the true cost of the next six nuclear reactors?

Le Figaro, By Elsa Bembaron, 08/11/2026
Subject to negotiations between France and Europe, at the heart of tensions between presidential candidates, the cost of new nuclear power is exacerbating the debate
It was almost a year ago. Bernard Fontana, the
CEO of EDF, unveiled the estimate for the construction of six EPR 2 reactors in France: a
maximum of €72.8 billion . A figure that has been constantly challenged ever since. The 2027 draft budget (PLF) proposes a sum between €114 and €161.8 billion. .
Greenpeace estimates that the cost of reviving nuclear power is between €380 and €650 billion, a figure widely adopted by La France Insoumise (LFI) and the Greens. As the presidential election approaches, anti- and pro-nuclear groups are clashing with a barrage of figures. Which ones should we consider?
They’re not all talking about the same thing. EDF expresses its estimate in 2020 euros. This is a technical choice, intended to allow comparison with initial estimates, excluding the effects of inflation. A look back is necessary to understand this choice. In May 2021, an initial estimate, described as “incomplete,” put the cost of 6 new reactors at €51.7 billion. This was………………………………………………………………………………………….. (Subscribers only – https://www.lefigaro.fr/societes/72-8-milliards-d-euros-selon-edf-680-milliards-selon-greenpeace-quel-est-le-veritable-cout-des-six-prochains-reacteurs-nucleaires-20261108
Coulport workers begin overtime ban over pay dispute
Greenock Telegraph, 9th November, By Tempany Grace
Industrial action is set to begin at the Royal Naval Armaments Depot (RNAD) Coulport, threatening to disrupt the UK’s nuclear response capabilities after union members voted to stop working overtime in a dispute over pay.
Staff employed by companies which form the ABL Alliance, which manages and maintains the UK’s Trident nuclear weapons system, started action on Monday, November 9.
Around 70 Unite the union members employed by companies which form the ABL Alliance – AWE plc, Babcock Marine, and Lockheed Martin – have started a continuous overtime ban and work only their contracted hours.
Unite says that the specialist nature of the services means that even action short of a strike could bring Coulport operations to a standstill.
The industrial action is due to escalate to three days of strikes on November 16, 17 and 19 unless an enhanced pay offer is tabled.
Industrial action was backed by 98.2 per cent of Unite members………………………………………………………………………………………………………………………………
The UK Navy operates four Vanguard-class submarines, which were planned to be decommissioned in 2022 as part of the strategic defence and security review (SDSR) but this was extended until 2028. https://www.greenocktelegraph.co.uk/news/26710599.coulport-workers-begin-overtime-ban-pay-dispute/
Global stocks plunge as fears over US AI slowdown spread to Asia and Europe

BBC, Rachel Clun,, Osmond Chiaand, Nick Edser,Business reporters, 5 Nov 26
Global stock markets have tumbled after investors sold tech shares on fears that the US midterm elections would slow the development of artificial intelligence (AI).
The sell-off, which started in the US and then spread to Asia and Europe, has seen trillions of dollars wiped off the stock value of major tech firms.
In the US, the tech-heavy Nasdaq share index sank 5.5% on Thursday. Asian markets also saw hefty falls on Friday and European markets opened lower.
Investors are worried about tougher regulation of AI, after Democrat candidates opposed the data centres needed by the technology. The share falls also come on top of concerns about whether the huge amounts being spent on AI will pay off.
The sell-off in US stocks meant the so-called magnificent seven tech companies, including Apple, Nvidia and Tesla, shed a trillion dollars in value in a single trading day, with Nvidia shares sliding more than 9%.
Shares in Elon Musk’s companies SpaceX and Tesla also plunged, falling 8% and nearly 11% respectively and shaving more than $300bn off the firms’ market value.
Musk had spent nearly $450m backing Republican campaigns, including at least $45m in support of Ken Paxton for the Texas senate seat, but Texas turned blue in the Democratic tsunami on Tuesday.
The sharp fall in the Nasdaq meant it lost an estimated $2.5tn (£1.9tn) in market value. The index is now in what market watchers call correction territory, which is where an index or company value falls at least 10% below its recent high.
However, the huge gains seen by tech stocks in recent times – driven by the enthusiasm for AI – mean that even after Thursday’s fall, the Nasdaq is still higher than it was a year ago, and up about 4.6% from where it was at the start of this year.
The US sell-off spread into Asia on Friday. The Nikkei share index in Japan closed down 6% as shares in memory chip makers Kioxia and Tokyo Electron saw big falls.
In South Korea, the Kospi index plunged more than 8% at one point, triggering a so-called circuit breaker, a mechanism that temporarily halts trading to curb panic selling.
The Kospi eventually closed down 6%, with shares in chipmaker SK Hynix falling more than 10%.
European markets fell on Friday, although the declines were not as severe as the indexes contain fewer tech stocks. The UK’s FTSE 100 index fell nearly 2% while Germany’s Dax index was down 2.4%.
Dan Coatsworth, head of markets at AJ Bell, said investors were worried about the huge amounts of money being spent on AI with “no clear idea of when, or whether, there will be payback”.
“The midterm election result muddied the waters as the Democrats now have more influence and could slow down the pace of data centre roll-outs and look at new ways to regulate the sector,” he added.
“The AI story has all been about speed, and now there is a danger of it moving to a lower gear, and some of the wheels falling off.”
Pressure on data centres
Democrats in races across the US campaigned to slow the growth of data centres and AI, and there were dozens of measures on ballots around the country aimed at restricting the development of the hubs crucial for powering the technology.
They won both the US Senate and House of Representatives in the midterms, taking control of Congress for the rest of President Donald Trump’s presidency.
“Americans don’t want Big Tech oligarchs wrecking the environment, gutting jobs and threatening humanity’s future. NATIONAL DATA CENTER MORATORIUM NOW,” Democrat Senator Bernie Sanders said in a post on X on Thursday…………………………………………………………………………………………..
Some tech leaders have said they were caught off guard by the backlash against data centres in the midterms.
Some tech leaders have said they were caught off guard by the backlash against data centres in the midterms.
OpenAI president and co-founder Greg Brockman said he believed everyone had been surprised by the reaction to data centres, and more work needed to be done to ensure communities were on board.
“A lot of our shift over the course of this year has really been to realise that from San Francisco trying to build infrastructure in Texas, Georgia, and other places, you can’t just tweet and say: ‘This is going to be good.’ You need to be in those communities,” Brockman said at the WSJ Tech conference in Napa, California.
https://www.bbc.co.uk/news/articles/cmn04705rd8jo
Britain faces ‘substantial’ threat of recession if AI boom falters

Spending on the technology is propping up growth, economists warn
Soaring investment in data
centres and other AI infrastructure has propped up growth since the start
of last year, Oxford Economics said. Without it, growth would be tepid at
best. The warning comes after a heavy tech stock sell-off in recent weeks
driven by concerns that AI spending will not deliver the level of returns
hoped for. October marked the worst month for the tech-heavy Nasdaq index
on Wall Street in more than a year and a half.
Telegraph 3rd Nov 2026,
https://www.telegraph.co.uk/business/2026/11/03/britain-faces-threat-of-recession-if-ai-boom-falters/
French industry eyes a fresh chunk of UK nuclear

EDF-owned Framatome is negotiating the purchase of land on one of the U.K.’s prime nuclear sites.
October 28, 2026 , By Nicholas Earl, https://www.politico.eu/article/france-nuclear-industry-bid-uk-expansion-framatome-edf/
LONDON — A firm owned by French nuclear giant EDF is locked in talks with the U.K. government as it tries to buy a prime site for nuclear power development.
Four industry figures familiar with nuclear planning in Whitehall, granted anonymity to discuss sensitive negotiations, said EDF-owned company Framatome is bidding to purchase land at Oldbury in Gloucestershire, southwest England, a site earmarked by the British government for nuclear development.
Senior U.K. officials have been talking up the potential for private investment on the site for months — but some insiders warn the French proposal could restrict other developments at Oldbury just as ministers are desperate to get low-carbon nuclear power running as quickly as possible.
EDF, which is owned by the French state, operates the U.K.’s entire nuclear fleet. It has an 80% stake in Framatome.
Framatome said in 2023 it wanted to set up a facility in the U.K. for making the fuel used by nuclear reactors, but did not specify a location.
“We have been supporting the U.K. nuclear power market for over 25 years and it is a focal point of business for Framatome,” said Bernard Fontana, the company’s boss at the time, adding that it was “in engagement” with the Department for Energy Security and Net Zero.
Fontana was named chief executive of EDF in 2025, reflecting the close ties between the companies.
Framatome is holding talks with the Nuclear Decommissioning Authority (NDA) over the purchase of land owned by the regulatory body at the site, according to the four industry figures.
The U.K. has its own nuclear fuels sector and uranium enrichment industries, but relies on France to supply some nuclear fuels and equipment used in reactor supply chains. Framatome signed a deal in 2025 with nuclear station Sizewell C, which is under development, on the understanding that fuel made in factories in France would come from a U.K. factory once it was up and running.
Then-Energy Secretary Ed Miliband said earlier this year that the government could offload Oldbury, a state-owned site, to private investors. “I would be very encouraging and positive about [opportunities at] Oldbury,” he told MPs.
Whitehall row
Two of the industry figures cited above said Great British Energy Nuclear (GBE-N), the arm’s-length body responsible for delivering U.K. small modular reactor (SMR) plans, had raised concerns over the sale of the land, worried it would hamper potential new nuclear generation at Oldbury.
Framatome has been “trying for ages but it’s land which may be needed for an SMR there,” one of the figures said.
One of the two figures said the decision over a potential sale NDA land now rested with Nuclear Minister Michael Shanks.
Oldbury was home to one of the U.K.’s early nuclear power stations, which closed in 2012 for decommissioning — a process expected to last until the 2080s.
The site was officially earmarked for future development in 2011 under the coalition government. But a chunk of its land was sold by the NDA to Horizon Nuclear Power in 2012, whose plans to set up a new nuclear plant on the site collapsed in 2020.
Financing structure for Sizewell C a ‘terrible deal’ – Public Accounts Committee

The MPs also took aim at the corporate and financing structure the government had selected for Sizewell C, called the Regulated Asset Base (RAB).
28 Oct, 2026, By Tom Pashby,
https://www.newcivilengineer.com/latest/financing-structure-for-sizewell-c-a-terrible-deal-public-accounts-committee-28-10-2026/
Sizewell C, the 3.2GW nuclear power plant under construction on the Suffolk coast, represents “a terrible deal for the UK taxpayer”, Members of Parliament (MPs) on the Public Accounts Committee (PAC) have said.
In its report, Sizewell C, published today, 28 October, the financing of the £38bn project, which exposes the public to up to £54bn in costs, was heavily criticised.
The first line of the report said: “Sizewell C represents a terrible deal for the UK taxpayer.”
It went on: “Successive governments’ failure to deliver new nuclear has left the UK in a position where the best terms it could secure to deliver Sizewell C mean that consumers and taxpayers shoulder significant financial risks, while private investors enjoy protected profits.
“The UK must never be put in this position again.”
Sizewell C, minority owner EDF, the Department for Energy Security and Net Zero (Desnz) and other proponents of the project have regularly argued that Sizewell C should be delivered more easily because it is copying so much of the design work from its ‘sister project’ Hinkley Point C in Somerset.
The PAC report said: “Desnz argues that applying lessons learned from the UK’s other current major nuclear project, Hinkley Point C, and replicating the design of its pressurised water reactors, will reduce risks at Sizewell C.
“However, no comparable nuclear power station anywhere in the world has been delivered on time and on budget. Projects using the same reactor design have experienced delays of up to 14 years and substantial cost overruns.”
The MPs also took aim at the corporate and financing structure the government had selected for Sizewell C, called the Regulated Asset Base (RAB).
“The Department has chosen a novel delivery model for Sizewell C in which it relinquishes majority control, despite the public providing most of the financing and bearing most of the risk,” the report said
Korea Hydro & Nuclear Power (KHNP)ends Daewoo-Doosan talks on $18.7 billion Czech nuclear plant
Korea’s state-run nuclear power operator ended talks with the Daewoo-Doosan consortium over key terms and plans to select a new contractor for the $18.7 billion Dukovany project.
October 29, 2026, https://www.koreajoongangdaily.com/business/khnp-ends-daewoodoosan-talks-on-187-billion-czech-nuclear-plant/12946437
Korea Hydro & Nuclear Power (KHNP), Korea’s state-run nuclear power operator, said Thursday it has ended negotiations with a local consortium on a construction contract for a nuclear power plant project in the Czech Republic after the two sides failed to agree on key terms.
“The company has decided to terminate negotiations with a consortium comprising Daewoo Engineering & Construction and Doosan Enerbility for the Dukovany nuclear power plant project. The breakdown is due to differences over key contract terms,” KHNP said in a press release.
KHNP said it will formally conclude the negotiations and begin preparations for a tender to select a new contractor.
“We aim to sign a contract with a new consortium for the Czech project during July of 2027,” Choi Il-kyung, executive vice president of KHNP’s overseas business division, said.
The Dukovany project, valued at about 25 trillion won ($18.7 billion), involves the construction of two 1,000-megawatt APR1000 reactors, a Korean-designed model, in the Czech Republic.
A Korean consortium led by KHNP won the project last year. KHNP subsequently negotiated a construction contract with the Daewoo-Doosan consortium, which had been designated as the construction contractor.
“The end of the negotiations does not mean the overall nuclear power plant project has been canceled as KHNP is preparing follow-up measures,” the company said.
The consortium expressed regret over KHNP’s decision, saying the nuclear operator had introduced revised contract terms in early October and rejected its request to extend the Thursday deadline before unilaterally ending negotiations.
It argued that the revised terms would unfairly burden construction contractors with additional costs arising from design changes, despite their lack of involvement in the reactor’s design.
It also accused KHNP of violating the principle of good faith by refusing further talks unless the consortium accepted the revised terms, urging the nuclear operator to resume negotiations in the broader national interest.
In a Facebook post, Industry Minister Kim Jung-kwan said he respects KHNP’s decision and pledged government support for the successful construction of the nuclear power plant in the Czech Republic.
Sizewell nuclear plant is ‘terrible deal’ for British taxpayers, MPs conclude.

Rachel Millard in London, Oct 28 2026, https://www.ft.com/content/bc330ee4-f151-4552-88a8-8879a26d8fdd?syn-25a6b1a6=1
The £40bn Sizewell C nuclear plant is a “terrible deal” for the UK taxpayer and a financing model that must never be repeated, a powerful parliamentary inquiry has concluded.
Taxpayers and consumers are shouldering “extraordinary” financial risk and footing the vast majority of the bill for the facility being built by a consortium that includes the UK government and the French utility EDF, while private investors “cash in”, according to the Public Accounts Committee made up of cross-party MPs.
However, the government had no option but to provide such attractive terms to investors owing to the failure of successive governments to lay the groundwork for nuclear investment, the MPs said, urging ministers to build confidence in the sector so better terms can be negotiated in future.
“The high level of risk borne by taxpayers and consumers for the construction of Sizewell C is extraordinary,” PAC chair Sir Geoffrey Clifton-Brown said as he outlined the committee’s findings on Wednesday.
Private investors would receive “significant returns on their investments, while customers are likely to be saddled with higher energy bills”, he continued. “Sizewell C represents a terrible deal for the UK taxpayer,” the PAC said. “The UK must never be put in this position again.”
Almost all of Britain’s existing nuclear plants are steadily coming to the end of their working lives, with efforts to replace them complicated by high costs and the length of time it takes to build new projects.
The 3.2-gigawatt Sizewell C project in Suffolk is due to provide electricity for the equivalent of 6mn homes for at least sixty years when it becomes operational at the end of the 2030s. It is one of only two nuclear plants approved in a generation after Hinkley Point C, being built by EDF in Somerset.
The Sizewell C construction cost, excluding finance costs, is estimated at £38bn in 2024 prices, or about £40bn in today’s money. The vast majority of this will come from the UK government, which is providing £3.8bn in equity and almost £37bn in loans funded by government borrowing.
EDF, which is supplying the reactor technology, is contributing £1.05bn while private investors La Caisse, Centrica and Amber Infrastructure are putting in a further £3.65bn.
The power plant is being built under a so-called regulated asset base model that means consumers begin paying for it from the outset of construction. The idea is to lower finance costs by reducing the risk for developers.
The National Audit Office has estimated Sizewell C’s finance costs are about five percentage points below those of Hinkley Point C.
Previous projects using the pressurised water reactor design being built at Sizewell C have been hit by heavy delays and cost overruns, including Hinkley Point C.
Yet despite providing the majority of the overall funding, the UK government has a minority stake in the delivery body that controls the project, an arrangement that the parliamentary committee said may make the government less able to intervene if things go wrong.
Its analysis shows that if costs rise to £47.7bn, the upper end of a potential cost range, consumers could pay £9.8bn more during construction and £12bn more over Sizewell C’s operating life, “while private investors may only need to put in an extra £30mn”.
Alison Downes at Stop Sizewell C, a local campaign group, and Steve Thomas, emeritus professor of energy policy at the University of Greenwich, urged the government to immediately cancel Sizewell C “before any more public money is wasted”.
Nigel Cann, chief executive of Sizewell C, said the plant was “the right investment for Britain” that would build energy independence, lower bills and provide reliable low-carbon electricity for decades to come.
AI’s real bottleneck is electricity. Who will pay for it?

In effect, households are already paying higher electricity capacity charges, partly because of projected demand from facilities that may never become operational
In the United States, the debate over AI infrastructure is becoming increasingly centred on who ultimately bears the cost of its electricity demands,
As Big Tech turns to nuclear power to fuel its AI ambitions, the cost of keeping data centers running raises a question that extends far beyond Silicon Valley.
By AlHasan AlSammarraie, Founding & Managing Partner – Applab. 30 Oct 2026, https://www.aljazeera.com/opinions/2026/10/30/ais-real-bottleneck-is-electricity-who-will-pay-for-it
Within a week, two of the three largest cloud computing companies in the world signed 20-year contracts for nuclear electricity. Amazon has taken 690 megawatts from Constellation’s Calvert Cliffs plant in Maryland, including 190 megawatts of additional generating capacity expected between 2030 and 2032. Google has signed an agreement that will bring 890 megawatts of additional nuclear capacity into the United States. Constellation is investing more than $4.3bn to upgrade 11 reactors across Illinois, Pennsylvania, and New Jersey. Neither agreement, however, involves building a new nuclear reactor. Rather, they are paying to generate more output from reactors that already exist – some of which are even older than the companies themselves.
A few years ago, such agreements would have seemed inconceivable. Nuclear power had fallen out of favour following accidents at Three Mile Island in 1979, Chornobyl in 1986 and Fukushima in 2011. Germany phased out nuclear power, while Japan shut down its reactors after Fukushima, although it has since begun restarting them. In many democratic countries, advocating for nuclear energy became politically difficult.
Today, an undamaged reactor at Three Mile Island is being brought back into service, with Microsoft contracted to buy its output for 20 years. The US Department of Energy is providing up to $4.4bn in loans to support the restart of three shuttered plants, with the aim of restoring about 2.25 gigawatts. Palisades in Michigan, which was closed in 2022, is contracted to be generating again by March 2027. If successful, it would become the first US commercial nuclear plant to restart after entering decommissioning.
What has changed?
The quiet reversal
In the last decade, the industry’s answer to AI’s energy problem was efficiency – things such as better chips, smaller models and smarter cooling seemed appealing and feasible. Industry conferences featured discussions about performance per watt, with modern artificial intelligence (AI) chips performing more computation per unit of electricity than their predecessors.
Yet those efficiency gains did not reduce overall electricity demand. Instead, each gain was quickly put towards something larger. Even when DeepSeek demonstrated that a competitive model could be trained at a fraction of the computing cost many had expected, the result was not a reduction in industry spending. Instead, laboratories continued planning more training runs, models that reason for longer, agents that work continuously in the background, and video generation.
This would not be the first time something like this has happened in history. In 1865, William Jevons observed that more efficient steam engines made Britain burn more coal, not less, because cheaper power created uses that had not existed before. The International Energy Agency estimates that global data centre electricity demand will more than double to around 945 terawatt-hours by 2030, even assuming continued improvements in efficiency.
Hence, the industry has changed strategy – unable to shrink demand, companies are now buying up supply.
The arithmetic of waiting
Money, however, cannot solve every problem quickly.
A new large reactor takes a decade or more from decision to electricity. Tech companies are investing in small modular reactors to power their data centres, although many of these projects are not expected to deliver electricity until the early 2030s. So far, very few of these have been built anywhere in the world. Thus, while reactor restarts and upgrades are the faster option, the “fast option” still means 2027, 2028, 2029, or beyond.
AI capital expenditure, however, does not operate on that same calendar. The chips being installed this year could be overtaken by newer technology long before the 20-year electricity contracts powering them expire. This discrepancy underpins all of these announcements, yet the industry has not addressed the implications if demand weakens while contractual obligations remain.
Who pays in the meantime
This is where the story stops being about technology.
Most of these deals sit inside PJM, the grid operator serving roughly 67 million people across 13 American states and Washington, DC. PJM uses capacity auctions to secure sufficient electricity-generating resources for future demand, with the associated costs ultimately passed on to consumers through their electricity bills. Capacity prices have reached the regulatory price cap in three consecutive auctions. According to the grid operator’s independent market monitor, data centre demand contributed approximately $29.4bn in capacity costs across the four most recent auctions. In the latest auction alone, around $6.3bn of these costs were attributed to data centre demand, including forecasts for facilities that have yet to be built.
In effect, households are already paying higher electricity capacity charges, partly because of projected demand from facilities that may never become operational. The Natural Resources Defense Council estimates that, by 2028, the typical household within PJM could face an additional $70 per month in electricity costs if current trends continue. These rising costs are also having political consequences. Recent US polling shows growing public dissatisfaction with data centres, with electricity bills among the main concerns. In response, Virginia has established a separate electricity rate class for data centres, Ohio has introduced minimum billing requirements, and Illinois is debating legislation.
The aim of the Google agreement is to tackle this objection. Its underlying principle is that major new electricity consumers should bring additional generation capacity rather than compete with existing users for an already constrained supply. This approach represents a constructive step towards mitigating the pressure that data centres place on electricity systems.
However, the arrangement remains voluntary. In October 2026, US federal regulators rejected most of PJM’s proposal to require large new electricity consumers to provide their own generation capacity. As a consequence, the industry is voluntarily offering a commitment that regulators have so far declined to impose through PJM’s proposed rules. This poses the question: how long will such voluntary commitments remain viable if the costs of fulfilling them begin to rise?
What the Gulf should take from this
For Gulf countries preparing their own AI infrastructure expansion, the US experience offers a warning about the costs of getting these arrangements wrong.
For countries with abundant energy resources, the expansion of AI infrastructure presents a significant opportunity.
Qatar’s sovereign fund has committed to a $20bn joint venture with Brookfield for AI infrastructure, beginning with a first phase of roughly 300 megawatts and potentially expanding to near 1.5 gigawatts. Abu Dhabi is pursuing plans for up to five gigawatts of AI data centre capacity. Saudi Arabia is targeting almost two gigawatts through HUMAIN. The region’s competitive advantage is substantial. As demand for AI computing grows, access to electricity is becoming an increasingly important factor in determining where computing infrastructure is built. The Gulf is well positioned to capitalise on this demand.
However, the scale of the proposed developments warrants scrutiny. If the proposed 1.5-gigawatt build-out is completed in Qatar, it would represent electricity demand equivalent to approximately 12 percent of the country’s installed generating capacity. Electricity allocated to AI infrastructure cannot simultaneously serve other purposes, meaning the opportunity costs of these projects deserve serious consideration.
In the United States, the debate over AI infrastructure is becoming increasingly centred on who ultimately bears the cost of its electricity demands, with many of the major commitments already in place. The Gulf is in a rare position of being able to settle that question first: what the industrial tariff is, who funds the new generation, what happens to the contracts when the hardware ages, and what long-term economic benefits the country would see beyond energy revenues and a limited number of jobs.
While hosting AI infrastructure can generate considerable economic value, the greater strategic opportunity lies in determining the conditions under which that infrastructure operates. For Gulf countries, the opportunity to establish these terms exists now, before large-scale commitments constrain their options.
Welsh share of Wylfa nuclear contracts unspecified as developers set 70% UK target
27 Oct 2026, Nation.Cymru staff,
https://nation.cymru/news/welsh-share-of-wylfa-nuclear-contracts-unspecified-as-developers-set-70-uk-target/
Developers behind the planned nuclear project at Wylfa have set a target for UK companies to supply 70% of its content, but have not specified how much should go to Welsh businesses.
The ambition was outlined as more than 400 suppliers gathered at Coleg Menai’s Llangefni campus to explore opportunities linked to Gwyndod, the project to build the UK’s first small modular reactors on Ynys Môn.
The trouble with small nuclear reactors

Proponents believe the technology is central to the AI data centre boom. Others think it is a waste of time and money.
In Darlington, Ontario, an expensive experiment is taking place.
Construction of the first of four small nuclear reactors, each around the
size of a football field, is under way. Connection to Canada’s grid is
planned for the end of the decade. One day, each reactor could power
300,000 homes. The project is meant to cost $15bn (C$21bn).
But critics do
not have much faith in that estimate. “This is a ‘first of its kind’
project and we don’t really know what it’s going to cost,” says Mark
Winfield, a professor in the environment faculty at Canada’s York
University. “The risks to ratepayers and taxpayers are enormous.”
The Darlington site is one of a handful of so-called small modular reactors
(SMRs) under construction that are being closely monitored by investors and
policymakers. Some believe the technology — essentially a scaled-down
version of a traditional nuclear reactor — is instrumental to the
cleantech revolution and the future of AI.
Others think SMRs are a colossal
waste of time — a distraction from cheaper, more efficient technologies.
The huge costs involved, they say, are unrealistic. And many designs rely
on fuel from hostile or rival states.
To date, billions of dollars, both
public and private, have flowed into SMRs. Last year, private capital
funding for the technology reached a record $2.9bn, according to PitchBook
data.
Proponents argue that the technology’s time has come. Unlike
renewables, nuclear produces a constant and reliable power stream that does
not need to be paired with battery storage. SMRs are also expected to be
quicker to build than conventional nuclear and take up less space.
In 2021, one SMR developer backed by Amazon, X-Energy, said that its demonstration
project would cost around $2.5bn, a figure that rose to almost $6bn in
2023. X-Energy declined to provide updated estimates but warned in June of
continued “price increases from our suppliers”.
FT 30th Oct 2026,
https://www.ft.com/content/da1af660-c1d0-4c8b-8f94-e84052344448
CANDU’s nuclear reactor business is unsupportable without overseas sales.

Magnificent projections of nuclear expansion within Canada have vanished in a puff of smoke. The industry tantalizes politicians with great expectations of fabulous success just around the corner. Without any public debate, the politicians are quite easily suckered in – only to find that the promises were based on engineering euphoria and little else.
Gordon Edwards, 29 October 26
There have been no domestic sales of CANDU reactors since 1979. And the overseas sales have been minimal – 6 reactor sales in 48 years (since 1979), on average one sale every eight years. That’s not enough business to keep the industry alive.
Overseas customers have been India (2), Pakistan (1), South Korea (1), Argentina (1) and Romania (1) – all before 1979 – and South Korea (3), China (2), and Romania (1) after 1979. The Türkiye sale, if it proceeds, would be new territory for CANDU sales.
Here’s what the World Nuclear organization has to say about the Monark reactor, which is essentially a reactor of the Darlington CANDU design – with modifications – but encased in a single reactor building (without a vacuum building) similar to the Point Lepreau CANDU-6 design.
From https://world-nuclear.org/information-library/country-profiles/countries-a-f/canada-nuclear-power
“In late 2023, AtkinsRéalis (formerly SNC-Lavalin) announced its new Candu Monark design, with several passive safety features. The 1000 MWe design is a heavy water cooled and moderated Candu reactor core of 480 channels, with a two-loop ‘figure of eight’ heat transport system, similar to the design used for the Darlington plant. The fuel is natural uranium in 37-element bundles (37M). The conceptual design was completed in September 2024, and in October 2024 AtkinsRéalis signed a service agreement with the CNSC for familiarization and planning of a future pre-licensing design review. A CNSC preliminary regulatory design assessment of the Candu Monark-D, the variant referenced to the refurbished Darlington plant, began in March 2026. In the same month Candu Energy signed a memorandum of understanding with Türkiye Nuclear Energy Company (TÜNAŞ) to assess Candu deployment there.”
Back in 1982, I wrote two articles in business magazines based on a study by the federal government about the sad state of affairs for the CANDU industry. One was printed in the
Canadian Business Review (Cost Disadvantages of Expanding the Nuclear Industry 1982), the other was for the Canadian
Journal of Business Administration.(Canada’s Nuclear Dilemma 1983). Magnificent projections of nuclear expansion within Canada have vanished in a puff of smoke. The industry tantalizes politicians with great expectations of fabulous success just around the corner. Without any public debate, the politicians are quite easily suckered in – only to find that the promises were based on engineering euphoria and little else.
There were other articles about CANDU export sales from the same period of time. They are illuminating in terms of past examples of overblown hype from the nuclear industry, promises of expansion and foreign sales that did not turn out well. It is useful to know something about the past history….
Feeding the Nuclear Dragon: CANDU exports to China
By Dave Martin, 1995
Documents related to the Sale of CANDU reactors to ChinaMiscellaneous, 1996
Exporting Disaster: The Cost of Selling CANDU ReactorsBy Dave Martin, 1996
Nuclear Sunset: The Economic Cost of the Canadian Nuclear IndustryBy Dave Martin and David Argue, 1996
The CANDU Syndrome: Canada’s Bid to Export Nuclear Reactors to Turkey
By Dave Martin, 1997
Appeal regarding the Akkuyu Nuclear Plant in Turkey
By Dave Martin and the Nuclear Awareness Project
-
Archives
- November 2026 (156)
- October 2026 (302)
- September 2026 (339)
- August 2026 (330)
- July 2026 (355)
- June 2026 (287)
- May 2026 (306)
- April 2026 (356)
- March 2026 (251)
- February 2026 (267)
- January 2026 (308)
- December 2025 (358)
-
Categories
- 1
- 1 NUCLEAR ISSUES
- business and costs
- climate change
- culture and arts
- ENERGY
- environment
- health
- history
- indigenous issues
- Legal
- marketing of nuclear
- media
- opposition to nuclear
- PERSONAL STORIES
- politics
- politics international
- Religion and ethics
- safety
- secrets,lies and civil liberties
- spinbuster
- technology
- Uranium
- wastes
- weapons and war
- Women
- 2 WORLD
- ACTION
- AFRICA
- Atrocities
- AUSTRALIA
- Christina's notes
- Christina's themes
- culture and arts
- Events
- Fuk 2022
- Fuk 2023
- Fukushima 2017
- Fukushima 2018
- fukushima 2019
- Fukushima 2020
- Fukushima 2021
- general
- global warming
- Humour (God we need it)
- Nuclear
- RARE EARTHS
- Reference
- resources – print
- Resources -audiovicual
- Weekly Newsletter
- World
- World Nuclear
- YouTube
-
RSS
Entries RSS
Comments RSS






