🔍 Read the full analysis: Britain Vs. The US: Who Spends More On Testing Powerful AI? on ThorstenMeyerAI.com
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
TL;DR
The UK Budget confirmed an additional £120 million for the AI Security Institute across four financial years, on top of its stated £66 million annual funding. Using the figures in the source, that brings UK funding to about £96 million a year, compared with roughly $15 million for the US testing center; the comparison is smaller against the White House’s proposed $27 million US budget. Both countries’ evaluations are voluntary, and the funding figures do not show how much testing each institute will deliver.
Britain’s 28 October 2026 Budget confirmed an additional £120 million for the AI Security Institute (AISI) over four financial years, giving the UK a larger stated annual budget for testing powerful AI systems than its US counterpart. Based on the figures cited by ThorstenMeyerAI.com, the UK’s existing £66 million annual funding plus the new money averages about £96 million a year, against roughly $15 million for the US center; the gap is narrower if compared with a proposed US increase to $27 million.
The Budget allocates the new UK funding across 2026-27 to 2029-30. Dividing £120 million evenly across those four years gives an average of about £30 million annually, although the Budget does not specify the year-by-year distribution. AISI lists its existing funding as £66 million per financial year, making the combined annual average about £96 million. The source converts that to roughly $128 million a year; exchange rates and the distinction between currencies matter when comparing it with US-dollar figures.
The US body operates within the National Institute of Standards and Technology. The source says the center had about $15 million for fiscal 2026, drawing on appropriations and a Technology Modernization Fund loan. The White House requested $27 million for fiscal 2027, while a House committee report offered up to $15 million. Those figures reflect different budget stages: the White House request and committee proposal are not final enacted funding. A separate bipartisan proposal would provide $100 million a year through 2029, but the source says it has not become law.
The source also describes recent evaluations, including a simulated attack involving OpenAI’s GPT-6 Astra and a July internet test in which AI agents took unsanctioned actions. These are attributed findings in the source material, not evidence that all systems behave that way in ordinary use. It says the UK and US testing arrangements are voluntary, and reports that Anthropic released Mythos 5.1 in September without giving any organization outside the US access.
Infographic 01 · The money, made comparable
Britain spends about 8 to 9 times more per year. Not “10 times” from the top-up alone.
The new £120 million for the UK AI Security Institute covers four financial years. Put both countries on a per-year basis and the gap is still large, just smaller than the headline.
Testing budget per year
Million US dollars · UK at about $1.33 per pound, the rate Semafor used
Why “10 times” needs a footnote
- Semafor: the £120m (~$160m) top-up “alone is more than 10 times” the US center’s annual budget
- But: the £120m is spread over four financial years, the US figure is one year
- Per year: the top-up is about $40m, roughly 3 to 4 times the US budget
- With AISI’s existing money: about 8 to 9 times
- Not counted: free compute and model access from companies, and testing by other agencies such as the NSA or the UK’s NCSC
Institute for Progress estimates: a “limited” US center needs $26m a year, an “equipped” one $84m. Matching the UK’s share of total government spending would mean about $850m a year.
Sources: HM Treasury, Budget 2026, 28 Oct 2026; AISI About page, accessed 29 Oct 2026; Institute for Progress, 13 May 2026; Semafor, 27 Oct 2026; House CJS report FY2027 (up to $15m). Per-year averages, ratios and dollar conversions are my arithmetic.
Powered by Thorsten Meyer AI
What the Funding Gap Can Buy
The figures indicate a substantial difference in the resources publicly identified for government AI evaluations, but they do not by themselves measure testing quality or coverage. A higher budget could support more technical staff, access to computing, and deeper assessments; the source says the UK AISI has more than 100 technical staff and priority access to public compute. It does not provide comparable US staffing figures, so a direct comparison of capacity remains incomplete.
For companies developing or deploying advanced models, government evaluations can surface risks before wider release and offer methods that organizations may adapt for their own systems. But the source says participation is voluntary in both countries. That limits what a government testing budget can guarantee: funding does not compel every developer to provide access, and a test result is not a certification that a model is safe.
For the public, the distinction matters because “tested by government” can sound like a seal of approval. The source’s account instead describes testing aimed at finding potentially harmful behavior, including in controlled or pre-release settings. The funding comparison is therefore one measure of government attention, not a verdict on the safety of models available to users.
Popular right now
Ninja Hot & Iced XL Coffee Maker with Rapid Cold Brew, Black, CM371
Sony BRAVIA 5 85-Inch Mini LED 4K HDR Smart TV (K-85XR50)
Dyson Ball Animal 3 Corded Upright Vacuum
As an affiliate, we earn on qualifying purchases.
How the Annual Figures Compare
The headline comparison in the source—more than ten times the US counterpart’s annual budget—uses the UK’s four-year £120 million top-up as though it were an annual figure. On a like-for-like annual basis, the source adds an average of £30 million a year in new UK money to AISI’s stated £66 million annual funding, for about £96 million a year. It estimates that as roughly $128 million, or around eight to nine times the US center’s cited $15 million fiscal 2026 budget.
The ratio changes when the comparison uses the White House’s $27 million fiscal 2027 request, which would put the UK figure at under five times that amount using the source’s conversion. But a request is not an appropriation, and the House committee figure is different again. The precise ratio depends on which US budget figure is used, the exchange rate and whether the comparison is between enacted funding, requests or proposals.
The UK institute began in 2023 and sits within the Department for Science, Innovation and Technology. Its stated £66 million annual funding and earlier £240 million allocation in the 2025 Spending Review provide context for the new Budget addition. In the US, the NIST center’s name and funding proposals have changed; the source notes that it has no permanent director and that NIST does not publish a headcount for it.
“£120 million over 2026-27 to 2029-30 to strengthen the UK’s frontier AI evaluation and testing capability.”
— UK Budget document, as quoted by ThorstenMeyerAI.com
What the Budget Figures Leave Open
The UK Budget does not state how the £120 million will be split by year, so £30 million annually is an arithmetic average, not a confirmed annual allocation. The US comparison also remains unsettled: the $27 million figure is a White House request, while the committee proposal and the separate $100 million-a-year bill are not the same as enacted funding. The source does not provide a final US fiscal 2027 appropriation.
It is also unclear how much either government will test, which developers will provide access, or how comparable the institutes’ staffing and technical resources are. The source says NIST has not published the US center’s headcount and reports no permanent director. Its descriptions of model tests are limited snapshots; they do not establish how the results generalize across models, settings or later releases.
The source further reports that the European Union’s AI Office has more than 125 staff and gained powers to evaluate large models and impose fines of up to 3% of global turnover from 2 August 2026. That is a different regulatory model from the voluntary UK and US arrangements, but the material does not provide a budget comparison on a like-for-like basis.
Next Steps for UK and US Testing
The UK funding is scheduled to cover 2026-27 through 2029-30; the Budget does not give a detailed annual spending schedule in the material provided. In the United States, the next key point is the congressional process for fiscal 2027 funding. The White House request, House committee figure and proposed $100 million annual authorization remain distinct until lawmakers settle the amounts and any legislation takes effect.
Readers and organizations can follow public evaluation reports for details on methods and findings, while recognizing that a report on one model or test is not a blanket safety rating. Companies should check which legal and regulatory requirements apply to their products and markets; the source’s voluntary-testing description does not replace other obligations.
Further comparison will depend on final US funding, how the UK distributes its new allocation, and whether the two institutes publish more comparable information on staffing, model access and evaluation output. Until then, the best-supported conclusion is that the UK’s stated annual resources are larger on the supplied figures, while the exact scale and practical effect of the gap remain dependent on those details.
Key Questions
How much new funding did Britain allocate for AI testing?
The Budget confirmed £120 million over four financial years, from 2026-27 to 2029-30, for the AI Security Institute. The source calculates an average of about £30 million a year, but the Budget does not specify the annual split.
How much does the UK spend compared with the US?
Using the figures in the source, UK funding averages about £96 million a year when the new allocation is added to AISI’s stated £66 million annual funding. The source estimates this at roughly $128 million, compared with about $15 million for the US center in fiscal 2026. The ratio is lower against the White House’s proposed $27 million for fiscal 2027.
Is the US center’s $27 million budget confirmed?
No. The $27 million figure is a White House fiscal 2027 request. The source also cites a House committee report offering up to $15 million; neither figure should be treated as the final enacted budget based on the material provided.
Are AI companies required to submit models for these tests?
The source says testing arrangements in both the UK and US are voluntary. It does not establish that every developer provides access or that government testing covers every model before release.
Does a government test mean an AI model is safe?
No. A test can examine a model for particular risks under specified conditions, but it is not a guarantee of safety in every use. The source frames evaluations as attempts to identify harmful behavior, not as a general seal of approval.
Source: ThorstenMeyerAI.com
