One year into the UK's AI roadmap, progress is slow; infrastructure growth is sluggish.

The UK announced its National AI Strategy last January, at which time Prime Minister Keir Starmer described the plan as aiming to transform the country into an “AI superpower.” This was met with positive signs from major tech companies pledging investment in AI infrastructure, coupled with favorable government regulations for data centers. However, high energy costs and delays in accessing the national grid remain significant obstacles.

December 28, 2568- Back in January of this year, the United Kingdom launched its AI Opportunities Action Plan. This is a large-scale master plan that promotes the use of AI in all sectors of society. Prime Minister Keir Starmer It has been clearly announced that this strategy will transform the country into... "The AI ​​superpower"

One of the key pillars of this plan is accelerating the construction of massive data centers to support the enormous computing power demands of AI applications. This will be driven through projects called "AI growth zones," areas with relaxed urban planning regulations and improved access to energy.

CNBC reports that over the past year, companies like Nvidia, Microsoft, and Google have announced investments in AI infrastructure in the UK totaling billions of dollars. Four AI growth zones have been launched, and the British startup Nscale has become a key player in the industry.

However, critics warn that severely restricted access to energy through the national electricity grid, coupled with construction delays, could put the UK at risk of falling further behind global competitors in the AI ​​competition.

Ben Pritchard, CEO of AVK. A data center power provider told CNBC that... “Ambition and action are not aligned. Growth is severely constrained by energy limitations, particularly electricity grid bottlenecks, which slow down infrastructure development and prevent the UK from accelerating infrastructure installation fast enough to catch up with global competitors.”

Currently, the construction of AI infrastructure in the UK is still in its early stages, as most AI growth zones are still in their development phase. The area in Oxfordshire, first announced in February, has not yet begun construction and is awaiting proposals from development partners. Meanwhile, the area in the North East of England, announced in September, has already begun site preparation and official construction is expected to begin in early 2026.

In addition, two more areas in North and South Wales were announced in November. The first area is seeking investment partners, with the Department for Science, Technology and Innovation (DSIT) indicating that this will be finalized within a few months. The second area comprises a group of multiple sites, some of which are already operational, while others require further construction.

The UK government stated in July that it aims for key AI growth zones to be able to meet at least 500 megawatts of energy demand by 2030, with at least one expanding to more than 1 gigawatt of capacity.

Pritchard stated that the most serious obstacle to achieving these goals is the limitations of the national electricity grid, with developers expecting to wait 8–10 years for grid connections and the backlog of connection requests, particularly in areas around London, at unprecedented levels. Meanwhile, the workload of AI is drastically increasing energy demand as businesses and consumers increasingly adopt the technology, putting further pressure on an already strained energy system.

“This is no longer a distant risk, but one that is significantly slowing or hindering development projects nationwide,” Pritchard said.

Spencer Lamb from Kao Data specify that The open application process for AI growth zones has resulted in landowners with power poles or transmission lines running across their land applying to participate in the project. The consequence is that the national electricity grid is overwhelmed with connection requests from speculative sources with virtually no chance of success.

The National Energy System Operator (NESO), responsible for the UK's national electricity grid, has begun addressing this issue. Earlier this month, NESO announced plans to accelerate the access of hundreds of projects to the electricity grid. While NESO declined to comment on whether AI infrastructure projects were among those prioritized, it acknowledged that a significant proportion are data centers.

Meanwhile, tech giants have announced massive investments, prominently showcased by the UK government in September. Companies like Microsoft, Nvidia, Google, OpenAI, CoreWeave, and others announced billions of dollars in AI investments during US President Donald Trump's official visit, with plans to install the latest AI chips and open new data centers across the country.

Meanwhile, Nscale, a UK startup providing AI computing power and building a data center, announced an agreement to install tens of thousands of Nvidia chips in an AI facility outside London by early 2027.

Puneet Gupta, General Manager of NetApp in the UK and Ireland. He said that large-scale private sector investment has helped lay the groundwork and provided impetus for national research supercomputing and plans to build “AI gigafactories” in the UK. However, he noted that the real test is the speed at which these plans will translate into computing power that UK organizations can actually utilize.

Stuart Abbott, Managing Director for the UK and Ireland at VAST Data. It was stated that long-term success requires investment in a full-stack infrastructure, encompassing data pipelines, storage, energy, security, personnel, and skills.

"If the UK wants this to be sustainable, not just a fleeting trend, it must treat AI infrastructure as economic infrastructure."

He added that an operating system must be built that allows real institutions to securely use AI on a large scale. Challenges remain significant, with data center investment in Europe far lower than in the US, while the UK has the highest energy costs in Europe, around 75% higher than before the Ukraine war, and a stale electricity grid that will take years to connect new sites.

One alternative for projects that cannot connect to the national power grid is microgrids, which are independent electrical systems derived from energy sources such as internal combustion engines, renewable energy, and batteries.

AVK is currently designing two microgrid projects for its cloud computing partners, which are expected to take around three years to build and cost about 10% more than grid power. Abbott suggests that locating computing centers near existing power sources, rather than developing entirely new sites, could accelerate the deployment of AI infrastructure.

Spencer Lamb concluded that speed of action is the decisive factor. "If structural problems related to energy, pricing, AI licensing, and funding are not addressed quickly, the United Kingdom could miss out on one of the greatest economic opportunities of our time and risk becoming a marginalized nation in the AI ​​landscape on the world stage."

refer: cnbc.com

Read news related to Technology circles around the world can be found here.