AI investments: UK government faces scrutiny over phantom AI claims

Published: 10 March 2026. The English Chronicle Desk. The English Chronicle Online.

The UK’s multibillion-pound AI drive is facing growing scrutiny over phantom investments. Since 2024, the government has highlighted large-scale deals aiming to establish datacentres, supercomputers, and thousands of jobs across the country. These announcements have been central to promises of “mainlining AI into the veins” of the economy, promoting the idea that Britain could become a global AI leader. The focus keyword “AI investments” appears to underpin the entire plan, yet investigations reveal inconsistencies between public claims and the reality of projects.

Two major firms, London-based Nscale and US-based CoreWeave, have been at the centre of the UK’s AI infrastructure ambitions. Nscale recently appointed former UK deputy prime minister Sir Nick Clegg and former Meta chief operating officer Sheryl Sandberg to its board, alongside announcing a $2bn funding round. This propelled the company’s valuation to $14.6bn, generating optimism for job creation and technological advancement. However, independent investigation shows that the supercomputer site near Loughton remains undeveloped, with the location still functioning as a scaffolding yard.

The Department for Science, Innovation and Technology declined to answer detailed questions regarding the supposed AI investments, asserting only that it “rejected these assertions.” Government statements claimed that over £100bn had flowed into the AI sector, which reportedly grew 23 times faster than the wider economy last year. Despite these claims, officials acknowledged they do not actively audit or contractually verify the investment commitments made by private firms.

CoreWeave’s investment announcements in 2024 included £1bn to establish two new datacentres in London Docklands and Crawley, promising thousands of jobs. Publicly, the project was framed as physically building new facilities, but planning records indicate no new construction took place at either location. Instead, CoreWeave rented space within pre-existing datacentres, one built in 2002 and another in 2015, both already hosting major clients such as Google and Fujitsu. Effectively, CoreWeave’s spending largely involved relocating computer chips produced overseas and establishing an office in Southwark, rather than creating significant new infrastructure.

Cecilia Rikap, an economics professor at University College London, explained that such practices are increasingly common, with tech firms inflating investment figures to satisfy government expectations. She noted that equipment purchases and acquisitions are often presented as fresh investment, a strategy that allows governments to highlight supposed growth in technology infrastructure without substantial economic impact. In the case of CoreWeave, the UK government clarified that reported figures were generated by the company itself, with no verification or oversight performed.

Nscale’s flagship project, announced in January 2025, also illustrates discrepancies between promise and delivery. The government highlighted a $2.5bn investment to construct the UK’s largest sovereign AI datacentre, with a supercomputer to be operational by 2026. Yet Nscale only submitted planning permission at the end of February 2026, and land records suggest the company has not formally registered ownership of the site. Government responses indicate the $2.5bn figure represents an intention to commit capital, rather than a signed contract or completed work.

Despite these issues, both Nscale and CoreWeave are pursuing additional initiatives. Nscale, in collaboration with Microsoft and OpenAI, is establishing Stargate UK, a project designed to develop domestic AI facilities across multiple sites. CoreWeave is promoting an AI growth zone in Lanarkshire, projected to deliver 3,400 construction jobs and incorporate 1GW of on-site renewable energy. However, energy supply projections appear unrealistic, as the site currently has only 24MW of electricity capacity, representing a small fraction of the claimed renewable output. Experts warn that ambitious renewable plans and job creation figures may be overstated to secure government approval and positive publicity.

The pattern of announcements and delayed or minimal delivery has raised concerns internationally about the credibility of high-profile AI investments. Governments and corporations worldwide pledged over £500bn in AI-related projects during 2025 alone, reflecting optimism about AI’s potential to transform economies. In the UK, Prime Minister Keir Starmer has argued that fully embracing AI could contribute £47bn annually, a claim which now faces scrutiny as evidence suggests some pledges may primarily benefit US-based corporations rather than directly supporting the domestic economy.

The scrutiny highlights broader questions about oversight and transparency in the rapidly expanding AI sector. The government maintains that datacentres are critical for delivering AI benefits, including boosting productivity and enhancing public services. Yet critics argue that the lack of active auditing and reliance on self-reported figures from multinational companies undermines the reliability of reported economic impacts. Without thorough oversight, the promised jobs and infrastructure may largely enrich foreign stakeholders rather than deliver tangible advantages for UK residents.

For CoreWeave, much of the responsibility for reporting job creation and infrastructure progress lies with its partners, including DataVita, who manage energy and construction planning. Government officials have acknowledged that they do not monitor these commitments directly, further complicating efforts to quantify the real impact of investments. This reliance on corporate self-reporting fuels concerns that claims of economic growth may be exaggerated or misrepresented to attract further private and public support.

The debate over phantom investments underscores a tension between ambition and accountability in the UK’s AI strategy. While the government has marketed AI initiatives as transformative for the national economy, evidence suggests many high-profile projects rely on flexible interpretations of investment, leasing, and infrastructure development. Critics, including academic experts and independent investigators, argue that these strategies risk prioritising corporate prestige and headline figures over genuine economic and technological progress.

As the UK continues to pursue AI expansion, transparency and verification of investment claims will be essential to ensure public trust. Without careful oversight, there is a risk that promised benefits, including thousands of jobs and substantial technological advancement, will remain largely aspirational. Experts stress that credible data, realistic timelines, and verifiable outcomes are necessary to transform bold investment rhetoric into tangible economic growth.

The unfolding situation also provides a cautionary tale for other nations embracing AI at scale. Globally, countries are racing to attract investment and demonstrate technological leadership, yet the UK case illustrates how announcements can outpace delivery. As governments highlight AI projects in press releases and official statements, citizens and industry stakeholders are increasingly seeking clarity about actual economic and technological impact. The UK’s experience underscores the importance of combining ambition with robust accountability mechanisms to realise the full potential of AI for national prosperity.

Check our latest news