Skip to content
Briefings are running a touch slower this week while we rebuild the foundations.See roadmap
UK Startups and Innovation
13MAY

Slough saturates, AI datacentres head north

4 min read
20:05UTC

Eighty per cent of UK AI datacentre capacity sits in London, and West London's grid is full. Scotland and the north are being handed the next wave almost by default.

TechnologyDeveloping
Key takeaway

Grid-queue reform, not electricity-bill discounts, is the binding constraint on UK AI capacity.

The Register reported on 20 April 2026 that London hosts 80% of UK AI datacentre capacity but Slough, the West London town that has been Europe's largest datacentre cluster since the 2010s, has reached saturation with 35 facilities and an exhausted grid. 1 UK commercial electricity runs at roughly four times US equivalents per International Energy Agency comparison data, and the AI Growth Zones programme is directing new capacity toward Scotland and northern England where wind generation is abundant and planning cycles are shorter.

Datacentres are the physical plant of the AI economy: buildings full of servers drawing 50-100MW each and needing dedicated grid connections, cooling water, and fibre backhaul. A single Nscale-scale build consumes as much electricity as a small town, and the National Grid connection queue for large industrial loads currently sits on planning horizons of a decade or more in the South East. The British Industrial Competitiveness Scheme, published in the Industrial Strategy Quarterly Update on 9 April, cuts electricity bills by up to 25% for qualifying manufacturers. 2 Hyperscalers are mostly excluded from that eligibility; the scheme is scoped to manufacturing, not compute.

Pulsant's CMO framed the new logic: "Start with the workload, the latency tolerance and the power profile, then choose the geography." AI inference tolerates roughly 20-millisecond latency where high-frequency trading does not; the workload's latency budget is what makes geographic dispersal physically possible. A 50MW facility in Scotland beats the same facility in Slough on unit economics before the lease is signed, provided the grid connection can be secured in commercial time.

The binding constraint is upstream of both the AI Growth Zones push and the competitiveness scheme's 25% discount. Nscale's $2bn build-out and SAIU's compute commitments both depend on grid connections the National Grid queue has not yet cleared, and transmission upgrades sitting on decade-long planning horizons. The state has named the companies and gestured toward the geography; whether the electrons arrive on time is a DESNZ and National Grid problem that policy on the electricity bill cannot solve. The first rejected datacentre planning application or refused grid connection in Slough will be the trigger moment that tests the AI Growth Zone thesis in practice.

Deep Analysis

In plain English

Most of Britain's AI computing sits in west London and a town called Slough nearby. But Slough has run out of electricity capacity; the grid simply cannot connect any more data centres there. The government wants new AI computing built in Scotland and northern England, where there is abundant wind power. The problem is that connecting new data centres to the electricity grid anywhere in the UK can take a decade because of a backlog at National Grid. So the limiting factor for British AI infrastructure is not money or ambition; it is the queue to get plugged into the power supply.

Deep Analysis
Root Causes

National Grid's transmission planning regime operates under a first-come-first-served queue that was designed for industrial loads with 25-30 year site horizons; AI data centres, which deploy in 18-36 months and are power-dense relative to footprint, do not fit the queue's amortisation logic. DESNZ's February 2026 consultation on transmission impact assessment reform is the mechanism that would fix this; but the consultation has not yet produced a statutory response.

UK commercial electricity is four times US equivalent prices (IEA, 2025) primarily because of network charges (TNUoS and DUoS) rather than wholesale power costs; Scotland's wind-generation surplus reduces wholesale cost but does not reduce network charges for facilities that are not directly co-located with generation assets, meaning the cost gap with the US is partially but not fully closed by geographic northward shift.

What could happen next?
  • Risk

    If the National Grid connection queue is not reformed before the first AI Growth Zone planning application is submitted in Scotland, the zone will receive designation but no new large-scale data centre connections before 2030, making the policy a geographic label without a functioning mechanism.

    Medium term · 0.65
  • Consequence

    Slough's 35 existing data centres face a five-year transition from AI inference growth to colocation consolidation as new capacity routes north; Berkshire councils dependent on data-centre business rates should model a 20-30% rates income reduction within five years.

    Medium term · 0.55
  • Opportunity

    Scottish data-centre operators co-located with wind generation assets; particularly those with direct wire connections to offshore wind farms in Orkney and Shetland; can bypass National Grid TNUoS charges entirely, achieving parity with US electricity costs without requiring any BICS subsidy eligibility.

    Long term · 0.5
First Reported In

Update #2 · Britain's innovation pipe leaks at both ends

The Register· 22 Apr 2026
Read original
Different Perspectives
Australian Department of Defence (AUKUS AI for Acoustics partner)
Australian Department of Defence (AUKUS AI for Acoustics partner)
Rowden Technologies holds active AUKUS AI for Acoustics contracts with the UK, US, and Australian defence establishments. The NWF's £25m investment in Rowden on 13 May brings UK sovereign capital directly into a trilateral programme, which from Canberra's perspective places additional UK government skin-in-the-game on a programme Australia co-funds and co-develops.
Sofinnova Partners (European VC co-investor in Cytospire Series A)
Sofinnova Partners (European VC co-investor in Cytospire Series A)
Sofinnova participated alongside the BBB in Cytospire's oversubscribed £61m Series A on 5 May, demonstrating that the BBB's expanded direct mandate is attracting established European specialist biotech funds rather than replacing them. European VCs see the BBB's cornerstone position as a signal reducing UK biotech execution risk rather than crowding out private capital.
Temasek (Singapore sovereign co-investor in Isomorphic Series B)
Temasek (Singapore sovereign co-investor in Isomorphic Series B)
Singapore's Temasek co-invested alongside the UK's SAIU in Isomorphic's $2.1bn round, treating the same Alphabet-majority company as an acceptable sovereign co-investment target. Temasek's participation normalises the structure: multiple sovereign wealth funds backed the same round, strengthening the precedent that UK-headquartered Alphabet subsidiaries qualify for state investment.
Alphabet / Google (majority Isomorphic shareholder, Mountain View)
Alphabet / Google (majority Isomorphic shareholder, Mountain View)
Alphabet co-invested via GV and CapitalG in the same Isomorphic Series B round that received UK sovereign backing, placing US corporate capital and UK public capital in the same syndicate without any governance asymmetry. SAIU's minority stake validates Isomorphic's strategic value without constraining Alphabet's control over IP, geography, or exit decisions.
DSIT / Liz Kendall, Secretary of State for Science
DSIT / Liz Kendall, Secretary of State for Science
DSIT framed the Isomorphic investment as backing a British-founded and headquartered company advancing UK AI capability, and described the nine-day sovereign deployment sprint as evidence the government's industrial strategy is operational. The department has not addressed the ownership question, the absence of eligibility criteria, or the pace-versus-doctrine tension in the BBB mandate.
Beauhurst / UK startup data analysts
Beauhurst / UK startup data analysts
Five sub-£50m rounds closed in nine days with zero VCT-backed angel networks on any cap table, confirming the post-cut investor map is forming fast in the £4m–£40m band. The gap is structural: 36.7% of university spinouts raised below £500,000 in 2025, a tier neither the SAIU nor the BBB direct mandate touches.