Back to stories
Industry

OpenAI Pauses Stargate UK Data Center Over Energy Costs and Regulation

Michael Ouroumis2 min read
OpenAI Pauses Stargate UK Data Center Over Energy Costs and Regulation

OpenAI has put the brakes on its flagship British data center project, Stargate UK, citing prohibitive energy costs and regulatory uncertainty. The decision, reported on April 9, deals a significant blow to Prime Minister Keir Starmer's strategy of positioning the United Kingdom as a global AI powerhouse.

What Is Stargate UK?

Announced in September 2025 during a visit by U.S. President Donald Trump that generated £150 billion (approximately $200 billion) in overall inward investment pledges, the Stargate UK project was a partnership between OpenAI, Nvidia, and cloud infrastructure firm Nscale. It was designed to strengthen Britain's sovereign compute capabilities — the ability to develop and run advanced AI models from domestic data centers — and accelerate AI adoption across the country.

OpenAI described the initiative at the time as a "major step" in the technology partnership between the U.S. and UK.

Why the Pause?

In a statement, OpenAI said it continues to "see huge potential for the UK's AI future" but added that it would only "move forward when the right conditions such as regulation and the cost of energy enable long-term infrastructure investment."

Britain's electricity prices remain among the highest in Europe, a persistent challenge for power-hungry AI data centers that can consume hundreds of megawatts. The regulatory landscape has also been a point of friction, with the UK yet to pass comprehensive AI legislation while simultaneously trying to attract the same companies it may eventually regulate.

Impact on UK AI Strategy

The pause is an awkward development for the Starmer government, which has made AI infrastructure a cornerstone of its economic growth agenda. A UK government spokesperson responded by saying they were "continuing to work with OpenAI and other leading AI companies to strengthen UK compute capacity," but offered no specifics on how the concerns would be addressed.

The decision also raises questions about OpenAI's broader "OpenAI for Countries" program, which offers similar sovereign AI arrangements to nations including Australia, Greece, the UAE, Slovakia, and Kazakhstan. If energy costs and regulation are sticking points in Britain — one of the world's largest economies — smaller nations may face even steeper hurdles.

What Comes Next

OpenAI has not provided a timeline for restarting the project, and the pause appears open-ended. For the UK, the challenge is clear: balancing its ambition to be a leading AI hub with the practical realities of energy pricing and the political complexity of crafting regulation that satisfies both safety advocates and the industry.

With competitors like France, the UAE, and Saudi Arabia aggressively courting AI infrastructure investments with subsidized energy and streamlined approvals, Britain risks falling behind in the global race for AI compute if it cannot resolve these issues quickly.

Learn AI for Free — FreeAcademy.ai

Take "AI for Business: Practical Implementation" — a free course with certificate to master the skills behind this story.

More in Industry

Anthropic to Close $30B+ Round at $900B Valuation, Vaulting Past OpenAI
Industry

Anthropic to Close $30B+ Round at $900B Valuation, Vaulting Past OpenAI

Anthropic is set to close a funding round of more than $30 billion at a pre-money valuation above $900 billion as soon as the week of May 26, surpassing OpenAI to become the world's most valuable private AI company.

3 hours ago2 min read
AMD Bets More Than $10B on Taiwan's Chip Ecosystem to Scale AI Packaging and Helios Racks
Industry

AMD Bets More Than $10B on Taiwan's Chip Ecosystem to Scale AI Packaging and Helios Racks

AMD is committing over $10 billion across Taiwan's semiconductor ecosystem to scale advanced packaging, ramp 2nm Venice EPYC at TSMC, and push its Helios rack-scale platform into hyperscale AI deployments this year.

4 hours ago2 min read
AMD's Gorgon Halo Pushes 192GB Unified Memory, Runs 300B-Parameter Models on One Chip
Industry

AMD's Gorgon Halo Pushes 192GB Unified Memory, Runs 300B-Parameter Models on One Chip

AMD's Ryzen AI Max 400 'Gorgon Halo' refresh hits 192GB of unified LPDDR5X memory — enough to hold a 300B-parameter model in FP4 on a single client SoC, with a $3,999 dev box aimed straight at Nvidia's DGX Spark.

6 hours ago2 min read