UK Government Slaps Down Green Call to Halt AI Data Centre Building

The Green party wants a freeze on large data centres. Labour says that would kill jobs and damage national security. Here is what the row is actually about.

AI2Day NewsdeskAI-assistedPublished Updated Editor: Lee Brown3 min read
Illustration: a matte-grey autonomous military ground vehicle parked on a concrete airfield at dusk, sensor array on top
Illustration made with AI. Not a photograph of the events described.
Share

Key points

  • Green party co-leader Zack Polanski called in 2025 for a moratorium, a full pause, on building large AI data centres in Britain.
  • Labour's government rejected the proposal, saying it "would be a disaster for jobs and national security".
  • Polanski described the facilities as "energy-guzzling, water-guzzling datacentres" that Britain should slam the brakes on.
  • The Green proposal would allow only small, locally focused data centres to continue.

Britain is building. Enormous warehouse-sized data centres, buildings crammed with specialised computers that run AI services, are going up across the country. Power, water, political heat: they consume all three. And now they're at the centre of a fight.

Zack Polanski, co-leader of the Green party, wants it stopped. He called for a moratorium on large AI data centres, arguing Britain should "slam the brakes on these energy-guzzling, water-guzzling datacentres". His proposal, first reported by The Guardian, would ban new large facilities but let small, community-scale ones continue.

We first covered the politics building around data centres in our 27 August story on why they've become a flashpoint.

What does Labour say?

Labour rejected the idea flat. A moratorium on large data centres, ministers said, "would be a disaster for jobs and national security".

The government's position is that AI infrastructure isn't a luxury. Data centres house the computing power behind hospital appointment systems, financial fraud checks, and the AI tools millions of people use daily. Pull the plug on new building and Britain falls behind rivals who keep constructing.

Why does it take so much energy?

Running AI takes a surprising amount of electricity. Training a model, teaching it to recognise patterns by processing billions of examples, can consume as much power as hundreds of homes use in a year. The servers, the specialised computers inside, need constant water cooling to avoid overheating.

That's the Greens' core concern. Britain has legally binding climate targets, and a wave of power-hungry buildings strains both the grid and the environment. A separate piece we ran on 27 August goes deep on how ghost data centre reservations are already jamming Britain's power grid.

What does this mean for ordinary people?

Right now, nothing changes. This is a policy argument, not a decision. But the outcome will matter.

More data centres means AI services stay fast and relatively cheap for UK users, with investment and jobs following. Tighter rules could raise the cost of running those services, pushing companies to build elsewhere instead. For people near proposed sites, the row's already real: local noise, grid pressure and planning battles aren't abstractions.

Both sides have a genuine point here, which is usually the moment when policy gets genuinely interesting. What I'd watch is whether Labour sets any conditions at all on future approvals, because a flat rejection of a freeze says nothing about what guardrails, if any, are coming.

Common questions

What is a data centre?

A data centre is a large building full of specialised computers that store information and run online services, including AI tools. They need huge amounts of electricity and water to operate.

Would a moratorium affect AI tools people already use?

Not immediately. Existing data centres would keep running. A freeze would slow future capacity, which could affect performance and costs for AI services over time if demand keeps growing.

© 2026 AI2Day