51²è¹Ý

Artificial Intelligence will be used to deliver the UK’s national priorities under  the government’s Plan for Change and position the country as an AI maker rather than an AI taker - accelerating economic growth and transforming public services, as a new strategy looks to bolster the country’s compute capacity to power new breakthroughs in AI.  

Businesses and researchers use compute – essentially the computer chips that process huge amounts of data – to train and build AI models or process prompts and questions through AI to discover everything from new drugs which treat and beat diseases to new tools to tackle climate change. Demand for cutting-edge compute power is already expected to surge by 5.7x between now and 2035, with the government taking vital steps to ensure the UK can stay ahead of the curve as the technology develops.  

Published today (Thursday 17 July), the Compute Roadmap will deliver on the £1 billion set aside in the Spending Review to increase the UK’s compute infrastructure - allowing us to drive forward AI development on our own terms to ensure the technology can deliver for the British people. This will mean reducing our reliance on foreign computing power to deliver the transformations which will improve public services and help to fix the foundations of the economy. The Roadmap also builds on the ambition of the 10-year infrastructure strategy and the Modern Industrial Strategy to put the government’s vision into action - increasing investment and growing the industries of the future.

Compute is the raw processing power that drives AI’s development. Without enough power, we cannot deliver the breakthroughs to treat and beat diseases, make industries cleaner and greener, or find new ways to fight climate change. To help deliver on these shared national priorities, we will expand the UK’s AI Research Resource (AIRR) twenty-fold over the next 5 years. The system, delivered in partnership with UK Research and Innovation (UKRI), Nvidia, HPE,  Dell Technologies and Intel, brings together the country’s most powerful supercomputers – Isambard-AI based in Bristol and Dawn in Cambridge.  

The Technology Secretary flicked the switch on the Isambard supercomputer at its formal launch in Bristol today, meaning the AI Research Resource (AIRR) is now fully up and running - transforming the UK’s public compute capacity by being able to process in one second what it would take then entire global population 80 years to achieve. When the AIRR’s planned expansion is complete in the coming years, it will be vastly more powerful than the world’s current leading supercomputers. 

University College London researchers are already using Isambard to line up pioneering AI tools which could revolutionise NHS cancer screening. Using prostate cancer as its initial test case, they are harnessing the system to develop one of the first scalable AI models dedicated to medical imaging - using AI to analyse MRI scans and identify patients in need of treatment sooner.  

Secretary of State for Science, Innovation, and Technology Peter Kyle said:  

Britain has top of the class talent in AI and our plan will put a rocket under our brilliant researchers, scientists, and engineers – giving them the tools they need to make Britain the best place to do their work.

This will mean we can harness the technology in Britain to transform our public services, drive growth, and unlock new opportunities for every community in the country.

Chancellor of the Exchequer, Rachel Reeves, said:

We are harnessing the power of AI to transform our public services, drive innovation and fuel economic growth that puts money in people’s pockets.

As technology advances, our Plan for Change is ensuring we are ahead of the curve, expanding our sovereign AI capabilities so we can make scientific breakthroughs, equip businesses with new tools for growth, and create new jobs across the country.

The AIRR will see the UK’s compute capacity increase to 420 AI exaFLOP by 2030 - the equivalent of one billion people spending 13,316 years doing what the full AIRR will do in one second. That means all one billion people would have needed to start calculating more than 8,000 years before Stonehenge was built, without taking a break. Projects that matter most to the UK and align with national priorities will be prioritised access to the AIRR to help deliver the Plan for Change - as well as those which will have a real-world impact and deliver breakthroughs that change lives and grow the economy. 

Researchers at the University of Liverpool meanwhile have been using Isambard to develop their EIMCRYSTAL system. Their model harnesses AI to speed up the discovery of new chemical reactions for use in industry, sifting through 68 million chemical combinations to find new solutions which will decarbonise British industry to make it greener, cleaner, and more sustainable. Isambard is already supporting other areas of highly ambitious AI research. The Sovereign AI Unit has launched an early pilot supporting academic researchers in AI for biosciences, foundational AI research, and advanced materials. These will be some of the most compute-intensive training runs that academics have carried out on UK infrastructure. 

Working alongside the AI Research Resource, a network of National Supercomputing Centres will also be set up across the country - with the first based in Edinburgh, the future home of the UK’s most powerful research supercomputer. These will work as dedicated centres of expertise, connecting users not only with access to cutting-edge processing power, but catalysing greater collaboration between industry, academia, and researchers. They will help to build stronger links with existing talent in their regions - giving all areas of the country a supporting role in the UK’s ability to be an AI ³¾²¹°ì±ð°ù.ÌýÌý

To further support the UK’s AI sovereignty ambitions, the Sovereign AI Unit has been established in the Department for Science, Innovation, and Technology, backed with £500 million of funding. Strengthening the UK’s domestic AI capabilities, including by developing the UK’s compute ecosystem, will be a key focus for the unit. 

The strategy set out today and the work of the Sovereign AI Unit will ensure the UK can roll out the next generation of champions in compute technology – sparking the creating of leaders in a range of fields to put British innovation and expertise on the map. 

Today’s Compute Roadmap also puts Scotland and Wales in the frame to benefit from billions in private investment and thousands of new jobs as future homes to AI Growth Zones. These dedicated AI hotbeds offer accelerated planning permissions to speed up the roll-out of data centres, which will be powered by responsible and cutting-edge energy sources like small modular reactors (SMRs).Ìý

AI Growth Zones will not only deliver the infrastructure we need but also support the technology’s evolution in a range of other areas. These will include R&D and Innovation Platforms, Adoption Testbeds and taking on a role as skills and talent hubs which will give people the tools they need to develop, use, and work with the technology. Further details of where these Growth Zones will be based in Wales and Scotland will be confirmed in due course.   

Capitalising on the ambition of today’s announcements, the Technology Secretary is also launching a dedicated AI for Science strategy. This will set out the clear steps the government will take forward to cement the UK’s position as a global leader in AI-enabled science breakthroughs, explore ways to boost adoption of the technology across the science sector and spark new commercial opportunities created by AI for science. 

An expert group of senior academics, industry leaders and representatives of science institutions will advise on the strategy:

• Alison Noble CBE FRS FREng, Vice-President of the Royal Society and Professor of Biomedical Engineering at the University of Oxford.
• Antony Rowstron, Chief Technical Officer at the Advanced Research and Invention Agency.
• Charlotte Deane, Executive Chair of the Engineering and Physical Sciences Research Council and Professor of Structural Bioinformatics at the University of Oxford.
• Chris Bishop, FRS FREng FRSE and Technical Fellow, Microsoft Research AI for Science.
• Pushmeet Kohli, VP, Science and Strategic Initiatives, Google DeepMind.

Published in the Autumn, the strategy will help to accelerate the pace of scientific discovery through AI, maximising its potential to drive innovation and growth.  

The roadmap set out today lays the groundwork for a golden age for British AI – supporting innovation, growth, and new opportunities in all sectors of the economy. It is a plan which delivers certainty to researchers, industry, and investors alike, cementing the UK’s position as a world leader in artificial intelligence.  

Reaction to today’s announcements

On the Compute Roadmap

Josh Payne, CEO, Nscale said:

Nscale strongly welcomes the UK Government’s compute roadmap.

As the only full stack sovereign AI infrastructure provider in the UK, we are delighted that the Government recognises the importance of sovereign capability in this area.

We look forward to working with the Government and our partners to deliver this ambitious agenda.

Professor Sir Peter Mathieson, Principal and Vice-Chancellor of the University of Edinburgh said:

To be named the UK’s first national supercomputing centre is a significant recognition of the University of Edinburgh’s longstanding leadership in advanced computing. For more than thirty years, we have hosted the UK’s national supercomputer and further developed our globally respected expertise in computer science and artificial intelligence.         The new designation as the first national supercomputing centre will provide new opportunities for research and innovation across the UK, attracting further investment and talent. We look forward to working alongside the UK government and partners to bring this ambitious plan to life.

Carolyn Dawson OBE, CEO of Founders Forum Group and Tech Nation:  

We know the UK’s AI ecosystem is brimming with talent and ambition, but to lead globally, we must anchor this ambition in cutting-edge, sovereign compute infrastructure.

Bold investment in compute power is exactly what’s required to accelerate innovation and secure a leading role for Britain in the global AI race. By bringing together world-class supercomputers in partnership with industry leaders like Nvidia and Intel, and expanding access through National Supercomputing Centres and AI Growth Zones in Scotland and Wales, this roadmap demonstrates the UK’s ambition to shape the future of AI.

Julian David OBE, CEO of techUK, said: 

This ambitious roadmap, underpinned by actions with dates for delivery, shows that the UK Government is serious in its ambition to deliver innovative and real-world impact through transformative AI, compute and cloud technologies.

We are particularly encouraged to see alignment between compute recommendations and AI Growth Zones - a vital move to better connect expertise, support UK innovators, and maximise the value of UK research and innovation. 

While there are still certain aspects to be explored, such as how these Growth Zones will develop the testbeds and platforms to help the most innovative emerging tech businesses grow and scale, techUK remains committed to working with government and our members to build on this ambition to power the next generation of AI.

Walter Goodwin, founder and CEO of Fractile, said: 

I wholeheartedly welcome the Compute Roadmap. The Roadmap is a joined up strategy that will both drive an immediate expansion of AI compute capacity in the UK, but further will ultimately see pull-through of breakthrough AI compute platforms being built by UK semiconductor companies, like Fractile’s AI accelerators, into widespread commercial deployment.

This will close the loop on sustainable sovereign compute capacity and ensure the UK will be an AI compute maker, not just a taker.

On the AI for Science Strategy

Dr Antony Rowstron, CTO of ARIA, said:

I’ve built my career at the intersection of computing and science, and seen firsthand how the right technological leap can redefine what’s possible. AI represents just such a leap – a once-in-a-lifetime opportunity to transform the speed of research and invention.

I’m looking forward to bringing that experience, and my perspective from ARIA, to help put the UK at the forefront of this revolution.

Chris Bishop, FRS FREng FRSE and Technical Fellow, Microsoft Research AI for Science said: 

I personally believe that scientific discovery represents the most important and promising opportunity for AI in our generation. The consequences are far-reaching, from the discovery of life-saving drugs to the efficient design of sustainable materials.

I am therefore delighted to participate, alongside other leading experts, in the new government strategic advisory panel on AI for Science. Together, I know that we will ensure that the UK remains at the forefront of AI development, in an area that is key to the future success of our society.

Pushmeet Kohli, VP, Science and Strategic Initiatives, Google DeepMind said:  

Science can help us address some of humanity’s greatest challenges, from climate change to disease.

I’m excited to collaborate with the UK government and other industry leaders, experts and academics to help the nation leverage AI to accelerate scientific progress, and build upon the UK’s strong history of scientific leadership.

Professor Alison Noble, Vice-President of the Royal Society, said:  

The Royal Society welcomes the government’s commitments to growing the UK’s computing power and AI research resources. Today’s launch of the government’s AI for Science Strategy is an important step to advance the responsible use of AI across scientific disciplines. 

From drug discovery to robot-assisted laboratories, AI is already reshaping how science is done and enabling new discoveries that were previously out of reach. To fully realise its benefits, we must ensure that advances in speed and scale do not come at the expense of rigour, transparency, or trust.

By embedding principles of openness, reproducibility, and collaboration, this strategy could help ensure AI-based science has a strong foundation.

Charlotte Deane, Executive Chair of the Engineering and Physical Sciences Research Council and Professor of Structural Bioinformatics at the University of Oxford said: 

AI will completely change the way research is done, from the way we ask questions to the questions we can ask. It has the power to transform so many areas across science and innovation, and we need to ensure that the UK is at the forefront of this change.

It is an exciting time to be involved in driving the potential of AI in science and for me an honour to be part of trying to make this change happen.