← Back to feed

Prime Minister Keir Starmer opened London Tech Week on June 8 with a sovereign compute announcement: £1.1 billion in AI hardware investment, a commitment to build one of the world’s most advanced AI supercomputers by 2030, and a joint statement co-signed by Google, Anthropic, Microsoft, and OpenAI.

The joint statement from all four frontier labs commits to “close collaboration with the UK government and frontier AI labs in support of evidence-based policymaking and responsible AI development.” It is the first time all four major US frontier labs have signed a single government AI cooperation document.

The £1.1bn Hardware Plan

The AI Hardware Plan has three components:

National AI supercomputer — £750m The government plans a heterogeneous mixed-chip system at “one of the most advanced in the world” specification when deployed in 2030. The system joins the existing Isambard-AI and Zenith systems in the UK AI Research Resource. £400m of the £750m is earmarked for chips directly: £150m for next-generation inference chips and £250m for specialized chips as the technology matures.

AI Hardware Innovation Programme — £120m Funding for UK companies to design, develop, and test new AI chips. Playground Global — a hardware venture fund backed by Pat Gelsinger, former Intel CEO — will open its first non-US office in the UK as part of the programme. It is the largest single fund investment the British Business Bank has made.

Next National Supercomputing Service (NNSS) — £750m The government is launching a tender for the NNSS at the University of Edinburgh, the successor to ARCHER2. This system is described as distinct from the AI Research Resource — it provides high-precision modelling and simulation capability for climate, materials science, and advanced research workloads.

Why Starmer Framed It as Sovereignty

The speech positioned Britain as occupying a “third path” between ignoring AI and removing all guardrails. Starmer: “Where British tech companies start here, scale here, and stay here, but the rewards of their success are felt in communities right across the country.”

The £400m chip purchase commitment is structured to specifically buy from UK-based chip companies, using government procurement as demand-side support to keep hardware startups from moving offshore to capture US or EU markets.

Technology Secretary Liz Kendall: “AI is the defining currency of economic and hard power in today’s world, and the countries that control the hardware behind it will hold the keys to the future. The UK is already a global leader in chip design, and I believe this is a race Britain can win.”

The Adoption Programme Running in Parallel

Separate from the hardware announcement, the government ran an AI Adoption Summit on the same day with a £200m package:

• £100m extension of the Bridge AI scheme matching UK companies with UK AI providers
• New AI Advisory Growth Labs for regulators and businesses to trial AI in live environments, starting with legal services
• 1.7 million AI skills courses completed under the government’s existing AI Skills Boost programme
• Cisco, IBM, Deloitte, and others committing to expand employer training

Context

The UK’s prior AI compute investments include the £500m Sovereign AI fund announced in April, which focuses on early-stage equity investment. Monday’s announcement is additive hardware spending, not a repackaging of the earlier fund.

The joint statement from all four frontier labs is notable partly because it comes from companies with conflicting interests: Google and Anthropic are directly competing at the frontier; OpenAI’s relationship with Microsoft is strained. Getting all four to sign the same document for any government is an unusual coordination outcome.

Britain is now running three parallel AI infrastructure tracks — venture funding (£500m), hardware investment (£1.1bn), and supercomputing procurement (£750m NNSS tender) — while seeking formal lab cooperation that no other government outside the US has secured at this level.