This kind of science/tech investment is and has been catnip for UK government, regardless of political party, for years. They're out of ideas to stimulate growth, and AI is their hail Mary. No one involved seems to be able to explain why it will work.
Ok let’s assume that AI will be a game changer. How does hosting it in the UK change anything? Just feels like something to brag about, but doesn’t have any impact.
The UK is suffering from persistently expensive electricity (https://grid.iamkate.com/, see All-Time). How does hosting AI data centres help with that?
In Norway I've noticed that stringent requirements for privacy make it much easier to run things in the cloud if the physical location of said cloud is actually in Norway.
So if OpenAI is hosting their services within the borders of the UK, then they would also be beholden to UK law. Makes it easier for the financial sector, government and healthcare to use their AI models than if they would have to send their data to a datacenter in the US.
Well, they're aiming to do the power via SMR reactors, and they have a large government supported company (Rolls) making those. I think there's a bit of a hail mary that that will be a successful export story.
Yeah it’ll be great when it happens. But work on the SMRs was only approved in June this year. It will take 10+ years of development and construction according to their current estimates. Every nuclear project usually takes twice as long and thrice as costly. So we could see those SMRs start to come online in like 2040.
The chips in these data centres would have been EOL-d by then.
I don't know much about this but do these prices reflect electricity costs for a datacenter? Maybe the infra is different than for general public use.. maybe the last mile problems don't exist for data centers.
From what I gather the UKians have a real chip on their shoulders from being left out of every tech revolution of the recent decades (chips, EVs and assorted green energy tech, rockets and now AI)
I am convinced that a lot of people have developed a completely distorted view of the UK by spending too much time on x.com where Elon Musk pushes his “come on, have a civil war already” bullshit and paints a picture of a country in its death throes.
I might've misspoken, I didn't mean to imply the UK didn't contribute significantly to these fields, it's that they didn't manage to establish themselves as major players.
I assure you I've heard these sentiments from real Brits.
While I'm not going to question the contributions of UK people and companies to these fields, but with the exception of RR, none of these companies are actually still in the hands of Britons, DeepMind is Google, ARM has been brought out by SoftBank (and Imagination that used to do the GPUs for the iPhone has been kind of sidelined).
The UK is not percieved as a mover and shaker in the rocketry industry like the USA is with SpaceX and the like, and not even to the extent other players like Russia/China are.
Sota AI is made today by US and Chinese firms, while the UK might have an extensively built out green infrastructure, it's made out of foreign-made equipment, no UK company makes EVs like the way Tesla or VW does etc.
Well, there's something to be said about the inability to capitalize long term.
Britain has managed to be at the forefront of all those revolutionary technologies, but when the real break out happens somehow the Americans swoop in and buy the winners. That's shaky ground for future ingenuity.
ARM seems to have been very important to Cambridge at large, but it's not Silicon Valley. I can sort of understand why politicians would look back at things and ponder what could have been done differently.
Not bad for a small country that's never been involved in a space race and which stopped trying to be a global Great Power before the space age even started.
I'm from Britain and can't think of anyone I know who has a chip on their shoulder about technology, largely because most of us either went to work for successful American firms the moment we graduated, or in the case of my brother, made a successful tech startup, grew it to be a profitable business and then sold it for a large sum of money (to the Americans again). Many of us have managed to achieve great life success by taking part in the tech industry, and were rewarded with small ownership stakes in those firms as a result. The fact that we didn't found those companies is a pity and a genuine source of relative weakness, but the reality is that the internet makes for global markets in which for any given product category there can only be a few winners. People can't really handle more than about four or five brands vying for attention simultaneously, which means it's just not mentally possible for every country to have a successful tech company in every category. The places that managed to grow competitors to the big US success stories all relied on either language barriers or government interference.
As for the rest, note that the USA is holding onto chip manufacture by its fingernails right now, an obsession with green tech is exactly the reason there aren't many AI datacenters in the UK to begin with, and Britain birthed one of the world's top AI labs. Yes, owned by Google because only great powers can invest the sums required, but that's OK. The collaboration between Britain and America on AI has been superb nonetheless.
Don't get me wrong. The UK is in a terrible state right now, the result of decades of leftward drift after the 1980s that consistently prioritized everything except economic success. Just turning around the Titanic would take years even if the process were to start tomorrow, which it won't, and the cultural gap is real. But there are still some strong foundations there. A whole generation of Brits have learned what great companies look like by working for the Americans. That's not reflected in their politics yet because politics is in both countries dominated by the old, and currently revolves around the issue of mass immigration. Economic success is on the backburner for now. But it'll come back. And when it does, there will be people who are ready to lead.
The UK didn't exactly get left out, it just failed to invest in growth and productisation. Some great research comes out of there and then moves overseas or gets bought up (like ARM).
If there are any chips on shoulders, it's more to do with the government talking itself up, throwing around some feel-good bullshit, usually entirely mis-targeted (see for example Sunak talking about making a safe space for blockchain companies) and then failing to get out of the way when the country actually does produce something amazing.
This is so sad. Like an addict who is hoping for that easy win to get them out of the terrible hole they are in...
This government (as the last, to be fair) is betting on unproven technology they barely understand and sometimes horribly misunderstand rather than making tough policy decisions.
The parts of this that actually materialise are very probably a good thing for the country, but this should really be business as usual. Alongside that I hear MPs, government ministers and commentators alike saying things like "when AI allows us to make vast efficiency gains in that government department...", "AI will obviously help to reinvigorate the Welsh economy...", "with AI we will be able to transition our industry to clean energy quicker..." (?!)
The fact that this very much feels like the country is lucky to have Nvidia/OpenAI rather than vice versa just demonstrates how badly managed the UK is.
This just feels to me like you’re starting from a conclusion (“the UK is being badly managed”) and working backwards in a somewhat nonsensical manner.
Yes AI is new but it has contributed to something like 1/6th of total US economic growth over the last year and of course any government will be courting more investment of that type.
To the Americans I would note that your country is currently being even more badly run than the UK at the moment given that since Trump’s inauguration the UK has grown faster than you (and every other G7 economy).
The percentage of GDP growth attributed to investments in computer and communications equipment, including chips, and data centres from the official statistics.
But then where hasn’t been? When I was born we were poorer than all of France, Japan, and Italy on a per capita basis and now we are not. We have broadly kept pace with US growth since the end of world war 2. We have failed to figure out strategies to close the gap with the US by growing faster but we are doing no worse.
Why not have this located in northern Scotland taking advantage of the surplus wind power being wasted due to north-south grid capacity, instead it’s likely going to be powered by gas generators.
In general the UK government talks growth but so far has delivered anti growth.
Tech outsourcing seems to have really picked up since the NI changes.
Isn't there a half-decent changes Scotland votes for independence from the UK in the foreseeable future? Seems like that is in opposition to the goal of data and compute soverignty.
Even before mass online discourse, divisive seperatist figures like Alex Salmond and Nigel Farage were regularly paid to appear on Russia Today. Salmond even had his own show!
To add to the other replies, the SNP have been the biggest political driver of the independence efforts and they seem to be falling out of favour at the moment.
It made me laugh when Nicola Sturgeon, who was the leader of the SNP for a long old time and who has been at the forefront of the Scottish Independence movement for many years, one of its most recognisable voices, recently announced she might retire to London because Scotland was now feeling a little bit suffocating.
So I think the idea of Scotland splitting from the UK is on the back burner for the forseeable future. It had a real moment in 2014 but didn't make it over the line. I don't think the UK government will be allowing another vote any time soon.
Even if it does somehow happen, there would need to be agreements in place for these cross border transactions to continue. It would shatter the economies to draw a hard line between the two nations.
OpenAI seem very non-commital about the whole thing. They will "explore offtake", which effectively means they'll think about it. This lack of commitment in a world where we seemingly can't build new data centers fast enough is a bit worrying. They clearly anticipate scenarios in which they won't want this capacity either because of poor pricing or lack of demand.
They seem to be betting on multiple things happening at once - as indicated by recent projections by their CFO, following Oracle's own insane projections. They are betting on demand rising exponentially and somehow keeping up with it using a consistent monetization model - which is not there yet.
I am not a finance person, but I wish more people with qualifications - and hopefully unaffiliated with any party - scrutinized all those claims and projections.
Only 8,000 GPUs to start and potentially 31,000 over time.
This is very small compared to Stargate Norway (announced almost 2 months ago), which starts out with 100,000 GPUs and planned (!) to have 250,000.
It sounds like this will include Teesworks, one of the dodgiest land deals of recent years. For a small investment two people essentially held the entire thing to ransom and secured options on an astonishing amount government money.
Absolutely. Exactly like the Stargate datacenters being built in the US have no affiliation with the Stargate project because Oracle is not involved in them. Just another way trying to make that big announcement credible without anything being done for that 500B project.
I also wish we were attracting industries that weren't going to significantly push up electricity consumption on windless days, which will have an outsized effect on electricity prices everyone else pays. At least this says the datacentres will be up north, hopefully not exacerbating transmission issues.
Very disappointed OpenAI isn’t actually producing a generated tv show, and for that project choosing to make a UK version of the US tv show Stargate, a la the US version of the UK tv show the office. I feel like if tech companies were run by people with a sense of humor such things would sometimes happen.
Especially when you see that the entire economy is leveraged on AI- and will likely cause another gfc (within the silent depression we're already in)
“We’re now locked into a particular version of the market and the future where all roads lead to big tech,” says Amba Kak, co-executive director of the AI Now Institute, which studies AI development and policy. Indeed, the success of major stock indexes—and perhaps your 401(k)—is resting on the continued growth of AI: Meta, Amazon, and the chipmakers Nvidia and Broadcom have accounted for 60 percent of the S&P 500’s returns this year. "
People wont be able to handle this. Small scale community is the only way to get through what is coming.
Because none of our most advanced llm models need more than a few h100s or whatever. They would be better building exascale computing not centred on llms. It's blatantly promising undeliverable results.
This kind of science/tech investment is and has been catnip for UK government, regardless of political party, for years. They're out of ideas to stimulate growth, and AI is their hail Mary. No one involved seems to be able to explain why it will work.
Ok let’s assume that AI will be a game changer. How does hosting it in the UK change anything? Just feels like something to brag about, but doesn’t have any impact.
The UK is suffering from persistently expensive electricity (https://grid.iamkate.com/, see All-Time). How does hosting AI data centres help with that?
In Norway I've noticed that stringent requirements for privacy make it much easier to run things in the cloud if the physical location of said cloud is actually in Norway.
So if OpenAI is hosting their services within the borders of the UK, then they would also be beholden to UK law. Makes it easier for the financial sector, government and healthcare to use their AI models than if they would have to send their data to a datacenter in the US.
But as a US company it’s also under US jurisdiction. So the alphabet agencies still get their data handed on a silver platter.
Well, they're aiming to do the power via SMR reactors, and they have a large government supported company (Rolls) making those. I think there's a bit of a hail mary that that will be a successful export story.
Yeah it’ll be great when it happens. But work on the SMRs was only approved in June this year. It will take 10+ years of development and construction according to their current estimates. Every nuclear project usually takes twice as long and thrice as costly. So we could see those SMRs start to come online in like 2040.
The chips in these data centres would have been EOL-d by then.
I love small modular reactor reactors
I don't know much about this but do these prices reflect electricity costs for a datacenter? Maybe the infra is different than for general public use.. maybe the last mile problems don't exist for data centers.
This is true. Data centres usually sign deals with power plants. But for the country as a whole, it does exacerbate the problem with a lack of energy.
And I’m perfectly happy to spend the energy as well! It’s just unclear what the benefit is to hosting within the UK.
Royal Dutch Shell for the win!
Yes, all the usual suspects (politicians and political pundits of all stripes) have moved seamlessly from “the green revolution” to AI.
It’s all irrelevant to the fundamental economic problems the country is facing, which do not have an easy solution.
>They're out of ideas to stimulate growth
You've gotta help us doc. We've tried nothing and we're all out ideas
From what I gather the UKians have a real chip on their shoulders from being left out of every tech revolution of the recent decades (chips, EVs and assorted green energy tech, rockets and now AI)
Are you saying this with an awareness that both DeepMind and ARM were founded in the UK?
I am convinced that a lot of people have developed a completely distorted view of the UK by spending too much time on x.com where Elon Musk pushes his “come on, have a civil war already” bullshit and paints a picture of a country in its death throes.
I'm assuming you are a hunter because your gathering skills suck.
Green energy like the world’s largest offshore windfarms which are in the UK? Or the SMRs Rolls-Royce are developing?
Chips like those designed by Arm that can be found in almost everything these days?
AI like DeepMind?
I might've misspoken, I didn't mean to imply the UK didn't contribute significantly to these fields, it's that they didn't manage to establish themselves as major players.
I assure you I've heard these sentiments from real Brits.
While I'm not going to question the contributions of UK people and companies to these fields, but with the exception of RR, none of these companies are actually still in the hands of Britons, DeepMind is Google, ARM has been brought out by SoftBank (and Imagination that used to do the GPUs for the iPhone has been kind of sidelined).
The UK is not percieved as a mover and shaker in the rocketry industry like the USA is with SpaceX and the like, and not even to the extent other players like Russia/China are.
Sota AI is made today by US and Chinese firms, while the UK might have an extensively built out green infrastructure, it's made out of foreign-made equipment, no UK company makes EVs like the way Tesla or VW does etc.
Well, there's something to be said about the inability to capitalize long term.
Britain has managed to be at the forefront of all those revolutionary technologies, but when the real break out happens somehow the Americans swoop in and buy the winners. That's shaky ground for future ingenuity.
ARM seems to have been very important to Cambridge at large, but it's not Silicon Valley. I can sort of understand why politicians would look back at things and ponder what could have been done differently.
People of the UK are referred to collectively as Britons or Brits.
Number of objects launched into orbit by country, both cumulative and annual:
1. USA
2. Russia
3. China
4. Britain
https://ourworldindata.org/grapher/cumulative-number-of-obje...
Not bad for a small country that's never been involved in a space race and which stopped trying to be a global Great Power before the space age even started.
I'm from Britain and can't think of anyone I know who has a chip on their shoulder about technology, largely because most of us either went to work for successful American firms the moment we graduated, or in the case of my brother, made a successful tech startup, grew it to be a profitable business and then sold it for a large sum of money (to the Americans again). Many of us have managed to achieve great life success by taking part in the tech industry, and were rewarded with small ownership stakes in those firms as a result. The fact that we didn't found those companies is a pity and a genuine source of relative weakness, but the reality is that the internet makes for global markets in which for any given product category there can only be a few winners. People can't really handle more than about four or five brands vying for attention simultaneously, which means it's just not mentally possible for every country to have a successful tech company in every category. The places that managed to grow competitors to the big US success stories all relied on either language barriers or government interference.
As for the rest, note that the USA is holding onto chip manufacture by its fingernails right now, an obsession with green tech is exactly the reason there aren't many AI datacenters in the UK to begin with, and Britain birthed one of the world's top AI labs. Yes, owned by Google because only great powers can invest the sums required, but that's OK. The collaboration between Britain and America on AI has been superb nonetheless.
Don't get me wrong. The UK is in a terrible state right now, the result of decades of leftward drift after the 1980s that consistently prioritized everything except economic success. Just turning around the Titanic would take years even if the process were to start tomorrow, which it won't, and the cultural gap is real. But there are still some strong foundations there. A whole generation of Brits have learned what great companies look like by working for the Americans. That's not reflected in their politics yet because politics is in both countries dominated by the old, and currently revolves around the issue of mass immigration. Economic success is on the backburner for now. But it'll come back. And when it does, there will be people who are ready to lead.
The UK didn't exactly get left out, it just failed to invest in growth and productisation. Some great research comes out of there and then moves overseas or gets bought up (like ARM).
If there are any chips on shoulders, it's more to do with the government talking itself up, throwing around some feel-good bullshit, usually entirely mis-targeted (see for example Sunak talking about making a safe space for blockchain companies) and then failing to get out of the way when the country actually does produce something amazing.
Maybe they think it's like Peter Pan, if enough people believe in it, it will work.
Everytime somebody denies the AI revolution, a Bored Ape loses its value
This is so sad. Like an addict who is hoping for that easy win to get them out of the terrible hole they are in...
This government (as the last, to be fair) is betting on unproven technology they barely understand and sometimes horribly misunderstand rather than making tough policy decisions.
The parts of this that actually materialise are very probably a good thing for the country, but this should really be business as usual. Alongside that I hear MPs, government ministers and commentators alike saying things like "when AI allows us to make vast efficiency gains in that government department...", "AI will obviously help to reinvigorate the Welsh economy...", "with AI we will be able to transition our industry to clean energy quicker..." (?!)
The fact that this very much feels like the country is lucky to have Nvidia/OpenAI rather than vice versa just demonstrates how badly managed the UK is.
This just feels to me like you’re starting from a conclusion (“the UK is being badly managed”) and working backwards in a somewhat nonsensical manner.
Yes AI is new but it has contributed to something like 1/6th of total US economic growth over the last year and of course any government will be courting more investment of that type.
To the Americans I would note that your country is currently being even more badly run than the UK at the moment given that since Trump’s inauguration the UK has grown faster than you (and every other G7 economy).
> 1/6th of total US economic growth over the last year
What economy indicator exactly are you referring to?
The percentage of GDP growth attributed to investments in computer and communications equipment, including chips, and data centres from the official statistics.
The UK has been badly managed for as long as I've been alive, which is half a century.
This isn't party political. The tories have been worse than Labour in many respects but it's a shit show all round imo.
But then where hasn’t been? When I was born we were poorer than all of France, Japan, and Italy on a per capita basis and now we are not. We have broadly kept pace with US growth since the end of world war 2. We have failed to figure out strategies to close the gap with the US by growing faster but we are doing no worse.
"we are doing no worse" : is that the royal we? Because normal people absolutely are doing a lot worse over that last couple of decades.
I fear “we are not the richest country in the world” is an unreasonable standard for claiming chronic mismanagement.
Again with the "we". Per capita statistics are great distraction from neofeudalism.
What was the per capita GDP of Wessex?
Why not have this located in northern Scotland taking advantage of the surplus wind power being wasted due to north-south grid capacity, instead it’s likely going to be powered by gas generators.
In general the UK government talks growth but so far has delivered anti growth. Tech outsourcing seems to have really picked up since the NI changes.
Isn't there a half-decent changes Scotland votes for independence from the UK in the foreseeable future? Seems like that is in opposition to the goal of data and compute soverignty.
Unlikely. Don’t believe the online discourse to be representative of the people.
See: https://www.telegraph.co.uk/politics/2025/06/25/scottish-ind...
Even before mass online discourse, divisive seperatist figures like Alex Salmond and Nigel Farage were regularly paid to appear on Russia Today. Salmond even had his own show!
To add to the other replies, the SNP have been the biggest political driver of the independence efforts and they seem to be falling out of favour at the moment.
It made me laugh when Nicola Sturgeon, who was the leader of the SNP for a long old time and who has been at the forefront of the Scottish Independence movement for many years, one of its most recognisable voices, recently announced she might retire to London because Scotland was now feeling a little bit suffocating.
So I think the idea of Scotland splitting from the UK is on the back burner for the forseeable future. It had a real moment in 2014 but didn't make it over the line. I don't think the UK government will be allowing another vote any time soon.
Not really. Scotland's status is in the gift of the UK government. They've given them one vote already, they won't get another for a long time.
Even if it does somehow happen, there would need to be agreements in place for these cross border transactions to continue. It would shatter the economies to draw a hard line between the two nations.
[flagged]
OpenAI seem very non-commital about the whole thing. They will "explore offtake", which effectively means they'll think about it. This lack of commitment in a world where we seemingly can't build new data centers fast enough is a bit worrying. They clearly anticipate scenarios in which they won't want this capacity either because of poor pricing or lack of demand.
They seem to be betting on multiple things happening at once - as indicated by recent projections by their CFO, following Oracle's own insane projections. They are betting on demand rising exponentially and somehow keeping up with it using a consistent monetization model - which is not there yet. I am not a finance person, but I wish more people with qualifications - and hopefully unaffiliated with any party - scrutinized all those claims and projections.
At last, a local AI for Tubbs and Edward.
Finally, a can of can't
Only 8,000 GPUs to start and potentially 31,000 over time. This is very small compared to Stargate Norway (announced almost 2 months ago), which starts out with 100,000 GPUs and planned (!) to have 250,000.
It sounds like this will include Teesworks, one of the dodgiest land deals of recent years. For a small investment two people essentially held the entire thing to ransom and secured options on an astonishing amount government money.
You can hear Richard Brooks summarise the situation (in 2023) here: https://www.private-eye.co.uk/podcast/76
It isn't "sovereign" if the parent organization is controlled by a foreign company.
Yeah weird, I thought the company called OpenAI is all about naming things in a way that accurately represents their real status.
Openai is handling their demand without any stargates. Then they announce a mega facility in the USA saying they are future planning.
Then they announce one for middle east, and perhaps the assumption is $ being thrown at them but surely they dont need that much capacity.
Now the UK? Confirming superintelligence.
Is the “Stargate” naming here just another attempt to mislead/allude to the Stargate Project that was announced a couple of months ago?
And ruin a good 90’s film & series’ SEO in the process.
Indeed.
Absolutely. Exactly like the Stargate datacenters being built in the US have no affiliation with the Stargate project because Oracle is not involved in them. Just another way trying to make that big announcement credible without anything being done for that 500B project.
Hopefully not too hindered by the high energy prices.
I expect they'll bypass our sickeningly broken energy market pricing mechanisms by just producing their own power.
Does this mean everyone will have to delete their photos so it'll have enough cooling water to run? https://www.theverge.com/science/758275/drought-delete-files...
"Labour struggling to get handle on Goa'uld immigration"
[Daily Mail]
Lord Hermer says Lord Ba'al is entitled to refugee status. [Guardian]
Given https://www.gov.uk/government/publications/ukusa-agreement-o... , what does sovereignty even mean here?
I also wish we were attracting industries that weren't going to significantly push up electricity consumption on windless days, which will have an outsized effect on electricity prices everyone else pays. At least this says the datacentres will be up north, hopefully not exacerbating transmission issues.
Very disappointed OpenAI isn’t actually producing a generated tv show, and for that project choosing to make a UK version of the US tv show Stargate, a la the US version of the UK tv show the office. I feel like if tech companies were run by people with a sense of humor such things would sometimes happen.
The AI grift is so obvious now. The smart are abusing the dumb.
Yeah I’ve definitely started to see the winds change. It’s gonna hurt everything.
Especially when you see that the entire economy is leveraged on AI- and will likely cause another gfc (within the silent depression we're already in)
“We’re now locked into a particular version of the market and the future where all roads lead to big tech,” says Amba Kak, co-executive director of the AI Now Institute, which studies AI development and policy. Indeed, the success of major stock indexes—and perhaps your 401(k)—is resting on the continued growth of AI: Meta, Amazon, and the chipmakers Nvidia and Broadcom have accounted for 60 percent of the S&P 500’s returns this year. "
People wont be able to handle this. Small scale community is the only way to get through what is coming.
Can you explain how this is a grift?
- using the “stargate” name to make it sound related to the Stargate Project (which itself is vaporware)
- very non committal verbiage
- very light on details, but enough to entice new -marks- investors (“please bro we’re so close to AGI just $20B more”)
Because none of our most advanced llm models need more than a few h100s or whatever. They would be better building exascale computing not centred on llms. It's blatantly promising undeliverable results.
[dead]