Britain Plots Atomic Reboot As Datacenter Demand Surges (theregister.com) 54
The UK is seeking to fast-track new atomic development to meet soaring energy demands driven by AI and electrification. According to a new report published by the government's Nuclear Regulatory Taskforce, excessive regulation has made Britain the most expensive place in the world to build nuclear projects. The report is calling for a sweeping overhaul to accelerate reactor construction -- everything from "streamlining regulation" to relaxing environmental and safety constraints. The Register reports: The document outlines 47 recommendations for the government, which come under five general areas: providing clearer leadership and direction for the nuclear sector; simplifying the regulatory approval process for atomic projects; reducing risk aversion; addressing incentives to delay progress; and working with the nuclear sector to speed delivery and boost innovation. Among the recommendations is that a Commission for Nuclear Regulation should be established, becoming a "unified decision maker" across all other regulators, planners, and approval bodies. The report also talks of reforming environmental and planning regimes to speed approvals, echoing the government's earlier decisions to streamline the planning process to make it easier for datacenter projects to get built.
It recommends amending the cost cap for judicial reviews and limiting legal challenges to Nationally Strategic Infrastructure Projects (NSIPs), while indemnifying nuclear developers against any damages they might incur as a result of proceeding with their project while a judicial review is still being decided. Another recommendation that may be cause for concern is that the government should modify the Habitats Regulations to reduce costs. These are rules created to protect the most important and vulnerable natural sites and wildlife species across the UK. The report also states that radiation limits for workers are overly conservative and well below what could be appropriately considered "broadly acceptable," claiming that they are many times less than what the average person in the UK normally receives in a year.
It recommends amending the cost cap for judicial reviews and limiting legal challenges to Nationally Strategic Infrastructure Projects (NSIPs), while indemnifying nuclear developers against any damages they might incur as a result of proceeding with their project while a judicial review is still being decided. Another recommendation that may be cause for concern is that the government should modify the Habitats Regulations to reduce costs. These are rules created to protect the most important and vulnerable natural sites and wildlife species across the UK. The report also states that radiation limits for workers are overly conservative and well below what could be appropriately considered "broadly acceptable," claiming that they are many times less than what the average person in the UK normally receives in a year.
News at 11, government agency needs importance (Score:3)
It's always odd how the agency, task force, blue ribbon commission, head of policy are X, ... in the government issues a statement or report that the general public would benefit from more of whatever its regulates or reports on.
Re: News at 11, government agency needs importance (Score:3)
To be fair I wouldn't expect someone like the Cheese Marketing Board to be pushing for more nuclear power.
What irks me is that we go from a report by an industry body to "Britainplots atomic reboot", as if this group is actually setting government policy. It's like the stories we used to get where one MEP floating some madcap scheme becomes a headline like "EU to ban/mandate [x]".
Raise the costs even more! (Score:5, Insightful)
Great, already the most expensive electricity and now they want to push that even higher. And for what? LLMs that are far from living up to expectations and in an over stretched bubble that'll going to burst long before any of these projects can ever come to fruition.
Need an external enemy (Score:2)
Elected officials in particular need some large external (to government policy or outside the government) trend, organization, company, threat, or more to endlessly debate about, form study groups and issue government reports on, and campaign for or against.
It's to justify the noise and attention seeking behavior of elected officials to secure reliable voting blocs and get reelected.
A larger guess is that since everyone uses electricity and water, including the millions of people who do not contribute to GD
Re: (Score:1)
Re: (Score:2)
"They'll need something to keep the lights on during a dunkelflaute. "
I don't think they have those in Britain. (the wind is always blowing somewhere, especially offshore)
Re:Raise the costs even more! (Score:5, Interesting)
Assuming equal regulatory burden, fission power is cheapest, not most expensive. For example, for an equivalent safety level, coal plants would need to capture every bit of reaction products (which currently are vented into the air), store them safely, and place somewhere where they'd no longer cause harm if released. Which for combustion products means forever. The plan itself would need decade-long studies wrt its localisation, many rounds of votes among the regional population -- etc. And throw in another 10x cost factor of bureaucratic costs.
The reason? Completely banning nuclear power was unfeasible politically, but adding layers after layers of "safety" was easy to be voted in.
Result? Hardly any new plants have been built. A good part of existing installed power dates back to the first generation -- which was indeed unsafe (as expected of any new technology). All three plants that failed have been built in the '60s.
Its only expensive in some places (Score:2, Troll)
Meanwhile in France 70% of their power comes from nuclear and their energy prices are lower than the UK.
Re: Its only expensive in some places (Score:1)
Re: (Score:2)
Meanwhile in France 70% of their power comes from nuclear and their energy prices are lower than the UK.
The province of Ontario in Canada is supplied by primarily nuclear power and their energy prices are lower than almost anywhere in Europe.
Re: (Score:2)
Re: (Score:2)
According to the regulator, in 2021, about 91% of electricity in Ontario was produced from zero-carbon sources: 55% from nuclear, 24% from hydroelectricity, 8% from wind, and 4% from solar.
https://www.cer-rec.gc.ca/en/d... [cer-rec.gc.ca]
To be fair I thought it was closer to 65%. Wind and solar have indeed displaced some of that in recent years. Hydro and nuclear are excellent backstops f
Re: (Score:2)
Government subsidies often do lower prices. Looking, it seems that the actual operating prices are pretty opaque, couple of relevant paragraphs from wiki,
Re: (Score:3)
It's not completely irrational, although arguable (I would be on this point of view for instance) it's now out of date. The idea is to pick the most expensive bid price for the spot market price, not the cheapest, in order to incentivise continued operation of difficult energy sources. W
Re: (Score:1)
Re: (Score:3)
Re: Raise the costs even more! (Score:1)
Re: (Score:2)
Re: (Score:2)
They seem to have bought into the SMR hype, despite nobody having even demonstrated a viable prototype commercial design yet.
Re: (Score:2)
AFAIK, nobody has demonstrated a viable SMR prototype of any kind. No, marine reactors do not count, they have the wrong characteristics and are far too uneconomic for this, even worse than civilian designs. The two that exist (Russian and Chinese) do NOT come with any or any believable cost figures. In addition, the he Russian one is a military design and the Chinese one is a highly experimental pebble-bed reactor based on German patents. The Germans wrecked three of these and two are still highly radioact
Re: (Score:3)
There are quite a lot of datacentres in the UK, but they're expensive because our energy is expensive (and so is land in a lot of locations). Our energy is expensive because we're (still) heavily tied to the price of gas, and we don't have a lot of gas and oil of our own left, so we have to import it all. We also have a lot of taxes applied to energy for environmental reasons as well as future investment in the grid and so on (our grid is for the most part pretty decent, even if the market for energy on it
Re: (Score:2)
Also, since it takes 10+ years to build a nuclear plant, nuclear power is kind of irrelevant.
Could the AI bubble do something good? (Score:5, Funny)
If this stupid-ass AI bubble makes governments rush to add lots of nuclear power capacity which will soon have no AI data center to feed, maybe something good could accidentally come from it?
Re: (Score:2)
We'll have an abundance of cheap energy once the bubble bursts or GPU get 10x more energy efficient.
Re: (Score:2)
I had a realisation a while back that it wasn''t AI research/dev per se thats driving this, its Nvidia thats driving it.
DeepSeek proved that you don't *need* the the "hyperscale" datacenters to develop good-enough AI. (Theres a lot of conspiracy theories about how DeepSeek must have had secret spooky mega sized datacenters doing all this, but they published their methods and training sets, and people have reproduced it, and it all checks out, you really can do this shit on the cheap).
And thats bad news for
Re: (Score:2)
Re: Could the AI bubble do something good? (Score:1)
Re: (Score:2)
They will just throw Rolls Royce some money to muck about with SMRs, before realizing what everyone already knows - they aren't better than traditional reactors, and nuclear in general is the most expensive form of energy we have.
Naturally the taxpayer and consumers will be on the hook for all this.
Re: Could the AI bubble do something good? (Score:1)
Re: (Score:2)
Theoretically, but in practice SMRs won't be useful for that. They still need a large and robust containment building, and nuclear grade security around it. They need a cooling pool or guaranteed supply of water.
A more practical idea for small fossil fuel stations is to turn them into spinning mass, to help provide inertia and a bit of energy storage. Or turn the site into a battery.
Re: (Score:1)
AI data centers aren't going anywhere (Score:3)
When the AI bubble bursts yes, you and I are going to bail out the banks that loaned doggy AI companies hundreds of billions (either that or they'll crash the global economy, remember, you're a hostage not a consumer)
But all that infrastructure you paid for with your tax dollars will just be bought up for cheap by whoever survives and you'll lose your jobs to it.
But hey, look over there! It
Re: (Score:3)
maybe something good could accidentally come from it?
Only if the money slated for nuclear is diverted to sources which make sense like wind and solar, which even when paired with batteries are now cheaper than coal, let alone nuclear.
Re: (Score:3)
They're both cheaper than fossil fuels, the main problem with nuclear has been that it can't be built in time to help with global warming and so can serve as a distraction that ties up resources that could've gone into renewables. A mad scramble to build them for the AI bubble could fix that, at least temporarily.
Re: (Score:2)
I agree that's the main problem in this context, but there are other large ones of course. The nuclear isn't just a problem in construction, it's also a problem in maintenance, and in decommissioning. Nuclear is also not cheaper than fossil fuels if you consider full lifecycle costs of operation. You might say it's cheaper because it's possible to contain the waste and that's not possible for fossil fuels, but fossil fuels shouldn't actually even be in the running.
Re: (Score:3)
maybe something good could accidentally come from it?
Like even more expensive power bills? Is that good?
Give it up. No nuclear plant will be built within the next 20 years in the UK as a result of this decision. Even the ones that are underway are horrendously resource constrained and literally none will meet time and budget.
Re: (Score:2)
"relaxing environmental and safety constraints" (Score:2)
Re: (Score:2)
The laws of physics are non-negotiable. Where to find the compromise between death by radiation and death by hypothermia is very negotiable. Similarly there is a broadband between living under a pine tree in the Lake District, and living on the slag heap from a blast furnace.
Re: (Score:2)
This will never happen (Score:3)
The UK has become so slow at developing infrastructure that it's now at the point where if you are in your late 30s you will NEVER benefit from anything they have not already put shovels in the ground for. They can literally talk about whatever they want - its takes so long you'll be on your way to the care home before it ever happens. I moved here 15 years ago when they were talking about 'making a decision on the third runway at Heathrow'. Today they announce that they are about to 'make a decision' on it again. There are things like the electrification of the Great Western line - which would benefit people for the next 100 years if they did it - and they still haven't done. The first electric trains on the underground happened nearly 130 years ago.
Re: (Score:3)
Yeah in my experience the UK sucks for any big projects. The boobs in charge will find a way to overcomplicate it, slow it down, make it fail, any cost the Government gives for a nuclear plant, you can quadruple it to get the price it will really cost - and double any timescales.
A mixture of incompetence and corruption, just like how we do software - the Govt could hire a team of 5 programmers for 6 months for a project, instead they will bid it out to their friends at a big corporation for 10's of million
Re: (Score:2)
Looking at the budget today, the UK is in managed decline. They are just slowly winding the country down, letting things age out and deteriorate. Keep kicking the can down the road for the next unlucky chancellor to deal with.
Summary should say target dose levels NOT limits (Score:4, Informative)
The worker dose limit is 20 mSv per year whilst the average public dose is 2.7 mSv. The worker dose target is 1 mSv per year. This represents what you should be trying to keep your workers exposure to, with the dose limit being the ceiling above which you'll be prosecuted if you exceed.
Re: (Score:2)
Presumably the dose target is an additional dose target i.e. a target of 3.7 mSv/year, unless workers are given lead-lined houses to live in.
Re: Summary should say target dose levels NOT limi (Score:1)
The motivations behind this clearly don't have worker health as a first priority. If they're thinking of changing the legislation so that the 20 mSv limit becomes a working ceiling then that's really quite bad. If you crunch the numbers, it would make nuclear work as dangerous as deep sea fishing, since
What happened to the wind turbines? (Score:1)
I thought a few wind turbines in Scotland were going to power the whole country.
Re: (Score:2)
They've kind of got to get around to building them first...
Britain will dominate the world... finally... (Score:2)
In 25 years when the first reactor comes online. It takes a great deal of effort to plan so far ahead to miss a boat. This is why America is better. The Second Amendment allows people to shoot themselves in the foot without waiting 2 decades.
Plotting maybe (Score:2)
But since they can't even build a single high-speed rail line for double the price that France paid for ALL THEIR LINES COMBINED, I don't see it happening.
Energy "too cheap to meter"... (Score:1)