The UK’s approach to AI is entering a new – if uncertain – phase. As new leadership looks to reshape the agenda through the appointment of Kanishka Narayan as AI Minister and dissolution of DSIT, their focus needs to shift from ambition to delivery, something the past administration struggled with. Structural changes and ministers’ matter, but they are only the starting point. For businesses deciding where to invest, innovate and scale, the pressing challenge is what comes next, as the success of the UK’s AI ambitions will depend on whether the government can provide a credible roadmap that gives organisations the confidence and opportunity to capitalise on AI.
The UK already has world-class research institutions, a strong technology ecosystem and a growing appetite for AI adoption across industries. Translating those factors into sustained economic growth requires a sharper focus on establishing practical foundations that enable organisations to move from merely experimenting with AI to implementing it deeply in their operations.
Ambition has never been the problem
The UK has been bullish about its aim to become a global AI powerhouse. Recent years have produced no shortage of strategies, frameworks and announcements designed to accelerate AI adoption, but progress cannot be measured by the number of plans published. It’s hard not to think back to the AI Opportunities Action Plan from last year as a case study – after an initial six-month update, we’re heard precious little since about progress.
As the government develops the next phase of its AI agenda, the focus must shift from setting direction to actual delivery. Businesses need confidence that the foundations for adoption are being put in place. This means expanding access to AI skills, accelerating investment in digital infrastructure, establishing clear standards, and providing organisations with long-term policy certainty. Without that, even the most ambitious AI vision will struggle to translate into lasting economic impact.
The DSIT question deserves a straight answer
While the AI Opportunities Action Plan’s impact was questionable, it was still the government’s central AI programme yet was led by the now-defunct DSIT. Anyone considering a multi-year AI investment in the UK will ask the same question – what’s going to happen to those commitments DSIT was supposed to own? For businesses making long-term investment decisions, clarity is as important as the plan itself. Machinery-of-government changes are normal, so what matters is ensuring long-term programmes retain clear ownership, accountability and momentum throughout periods of transition. That remains unclear in this case.
For a finance director weighing up whether to establish AI infrastructure in the UK or in a market with a steadier delivery record, that uncertainty is reason enough to hold back investment. OpenAI has already made that call by pausing its £31bn Stargate UK project this past April, citing the nation’s regulatory uncertainty as a chief reason – and that was well before the current shake-up. This project was originally touted as a centrepiece of the government’s AI ambitions, and its pause has cost the UK: thousands of GPUs, new data centre capacity and a real marker of sovereign compute strength now lost.
The government’s most important contribution will be creating conditions that allow businesses to invest with confidence – both among foreign firms placing their bets on the UK, and local enterprises who could still be going further to embed AI in more processes but can’t yet do so in the current environment. That means providing policy stability, clear regulatory direction and access to the talent needed to support long-term adoption.
Put businesses to work
Procurement reform also needs to be on the agenda. Empowering UK firms to innovate and solve some of the country’s biggest problems can’t only happen within the context of supporting AI sovereignty. If anything, a focus on independence misses the point.
Giving local companies the opportunity to fix public services, for example, cannot only be a knee-jerk reaction when sentiment towards big tech sours, like what we saw when MPs opposed Palantir’s NHS contract (rightly or wrongly). Companies cannot be expected to build brilliant things in a sandbox – they need real opportunities to get their hands dirty and direct their best thinking and technologies towards actual use-cases.
To do this effectively, local players need clear, responsible pathways to leverage models from the likes of OpenAI and Anthropic while sovereign alternatives are built in tandem. Leaders cannot promote sovereignty based on principle, when the fact is we’re still at least five years away from matching the capabilities that exist abroad. While the country embarks on that crucial journey, companies need access to the advanced models in the meantime to propel their growth and tackle real use-cases of consequence.
This aligns perfectly with the Prime Minister’s goal to decentralise power and the economy from London. There is impressive work coming out of places like Oxford and Manchester to name a few, and the more the government recognises the importance of giving local enterprises a shot at lucrative public-sector contracts, the more these firms will grow and the quicker they will scale their own capabilities to match those present overseas.
Judge on outcomes, not intention
The most meaningful success metrics are whether UK organisations are able to consistently move AI initiatives from pilot to production, and whether those investments deliver measurable operational and commercial value.
The UK does not need another plan, framework or statement of intent – we already know where we need to be. It needs delivery. That means clear ownership of existing commitments, practical procurement and adoption pathways that organisations can use, and a policy environment stable enough for businesses to invest with confidence.
Get that right, and the UK’s AI ambitions will start actually translating into business and economic value. Keep repeating the mistakes of the past by focusing on intent over action and we’ll be having this same conversation in another twelve months.
