Queues for the grid have grown by 460% as housing projects reckon with capacity constraints – and AI data centres could be to blame.
First reported by the BBC, the new proposal said projects that promise the most economic growth and jobs could be prioritised. However, the Home Builders Federation (HBF) said failing to prioritise connections for housing developments could be a “moratorium” on new homes.
So what could the true cost be of this surge in AI data centre demand?
The cost of economic growth
In the Autumn 2025 budget, UK Chancellor of the Exchequer Rachel Reeves said the government would be betting big on infrastructure across the country to boost economic growth.
“If you build here, Britain will back you,” she said. “We are putting money and power back in the hands of regional leaders.”
She added that “private investment is the lifeblood of economic growth,” pledging the government would be boosting its commitment to “make Britain the best place to start up, scale up and stay”.
Part of these plans included boosting AI Growth Zones across the country, showing how demand for AI continues to grow nationwide. It was hoped that funding for AI and data centres would lead to investment into local infrastructure, business and skills programmes.
But what happens if an economy is saturated with AI data centres?
Ofgem, after warning that the queue had moved way beyond “even the most ambitious forecasts for future demand,” is now consulting on allowing “strategically important” projects to skip the queue. These will include AI infrastructure and industrial site projects, in addition to changing the rules to join the queue in the first place.
It also noted that some of the projects in the queue were only “speculative” and don’t have proper financing, planning permission or land rights.
This news comes amid a recent investigation by The Guardian, finding evidence of “phantom investments” in the data centre industry. According to the report, the money being pumped into data centres isn’t necessarily real, potentially undermining announcements from data centre companies promising investments to turbocharge AI in the UK.
“We still have a long way to go when it comes to processes and planning for AI infrastructure,” said Matthew Grant, development director at Colt Data Centre Services. “Firms should be required to provide concrete evidence of their projects.
“Standardisation is critical. Without it, regulators can’t hope to manage applications effectively while benchmarking environmental impact. The current system makes distinguishing between credible projects and pipe dreams extremely difficult.”
From critical infrastructure to capacity constraints
The new evidence from Ofgem means these investments could also impact genuine projects that promise new jobs and local economic growth. This may include housing, as the National Housing Federation said 8.5 million people in England are without the housing they need.
“As we continue to face a housing crisis, it is frustrating that regulatory, planning and policy arrangements effectively prioritise energy-intensive data centres over energy-efficient homes for families,” Steve Turner, executive director at HBF, told the BBC.
The Planning and Regeneration committee of the London Assembly also argued that some new housing developments in west London had been delayed after the electricity grid reached capacity.
Regarding data centres, Grant added: “Without clear timelines, resource planning, and progress tracking for data centre projects, the UK will continue to overpromise and underdeliver when it comes to AI investment and the infrastructure powering it.
“Trust takes time to build but can be easily lost, and a lack of visibility into data centre projects risks undermining confidence in both government strategy and the sector as a whole.”
With data centres being classified as critical national infrastructure, they cannot be blocked by local objections. This reality could continue to be hard-hitting for some local areas that require more housing and job creation.
The UK remains eager to position itself as an AI hub, but the real challenge might lie in balancing that ambition with local infrastructure and community needs.
Good, future-leading innovation should consider those who may get left behind.
Related stories
Biomethane emerges as data centres’ flexible energy option
Pure DC & AVK plan to deploy Europe’s first data centre microgrid
