Policy

UK Government's £400 Billion AI Economic Forecast Traced Back to Google-Commissioned Report

UK officials' widely-cited £400B AI growth projection came primarily from a Google-commissioned 2023 study, while the government's own estimate is less than a third that amount.

Last verified:

The Divergence Between Public Claim and Internal Assessment

The UK government’s promise of a £400 billion economic windfall from artificial intelligence adoption has anchored AI policy messaging for two years—until this week’s parliamentary disclosure exposed a stark gap between the publicly promoted figure and officials’ actual projections. According to Wired AI, UK AI Minister Kanishka Narayan acknowledged during a parliamentary question that the widely-circulated £400 billion projection originated from a Google-commissioned 2023 report by the British consultancy Public First, not from independent government modeling. The government’s own internal assessment, by contrast, forecasts AI will contribute between £55 billion and £140 billion to the UK economy by 2030, with a central estimate of £120 billion—less than one-third of the publicized figure.

How the Attribution Gap Emerged

For nearly two years under two successive UK administrations, the £400 billion figure served as official talking points during legislative debates and policy announcements, intended to offset public anxiety about AI-driven job displacement. Wired AI’s investigation found that government ministers held the more conservative estimate internally since at least November 2025, yet continued citing the higher projection publicly. Skills England, the government agency overseeing apprenticeships, published the £400 billion figure as recently as four months before Narayan’s admission, suggesting incomplete circulation of revised estimates across the civil service.

The “Hugely Uncertain” Core Assumption

Narayan characterized his own government’s projection as “hugely uncertain,” signaling how dependent such forecasts remain on volatile adoption assumptions. PF Global’s head of economics, Jonathan Dupont, attributed part of the divergence to differing views on implementation velocity. According to Wired AI, Dupont argued that the 2023 Google-sponsored model assumed faster AI adoption rates than the government’s current assessments, which he framed as reflecting a more cautious, realistic pace given the technology’s volatility. This explanation, however, does not address why the government continued promoting Google’s higher figure after acquiring lower-confidence internal projections.

Parliamentary Scrutiny and Transparency Concerns

Dame Chi Onwurah, the Labour MP chairing the Commons Science, Innovation and Technology Committee, told Wired AI that the episode highlights “one of the central barriers to delivering the digital transformation of public services”—the risk that technological hype undermines credibility with the public. She emphasized that while governments must promote technology’s potential, overstating benefits without acknowledging uncertainty “risks undermining the case for change.” Notably, Narayan’s parliamentary response did not withdraw the £400 billion figure or declare it incorrect; it simply placed both numbers on the record without guidance on which should inform future policy or public communication.

Why This Matters

The revelation exposes a structural credibility problem in how governments monetize AI’s economic promise. When a single vendor-commissioned study anchors multi-year policy narratives without disclosure, and internal estimates remain compartmentalized rather than reconciled publicly, citizens and policymakers lack the granularity needed to assess trade-offs between AI investment and workforce transition costs. For UK government departments deciding AI integration priorities and budget allocation through 2030, the question is no longer whether AI will add £400 billion or £120 billion—it is whether either projection should drive hiring, retraining, or regulatory policy absent independent peer review and explicitly stated confidence intervals. The gap between public and private estimates may widen further if actual adoption lags both the Google model and the government’s revised forecast.

Frequently Asked Questions

Where did the UK's £400 billion AI economic projection come from?

According to Wired AI, the figure originated from a 2023 report commissioned by Google and produced by the consultancy Public First (now PF Global), not from independent government analysis.

What is the UK government's own AI economic forecast?

UK AI Minister Kanishka Narayan stated the government projects AI will add between £55 billion and £140 billion to the economy by 2030, with a central estimate of £120 billion.

How long has the government used the £400 billion figure?

According to the source, government departments cited the £400 billion projection for nearly two years across two administrations, despite having access to lower estimates since at least November 2025.

#UK #economic-impact #government-forecast #AI-adoption #transparency