Lloyd Rushton-Jones, General Manager & Regional VP, Northern Europe at Denodo, commented:

“The Government’s additional £140 million commitment to sovereign AI capabilities and its new Data Centre Covenant are welcome steps towards strengthening Britain’s AI infrastructure. But building more domestic capacity does not automatically give organisations meaningful freedom of action.

“Genuine AI sovereignty goes beyond where data is stored or technology is hosted. It also means retaining authority over how information is accessed and used, the policies that govern it, and the business context AI systems rely on to make decisions. Organisations need to be able to change technology providers without losing that operational understanding or becoming locked into a single ecosystem.

“As the UK invests in its AI infrastructure, the real test will be whether businesses and public services gain greater authority over their information, more control over how AI operates, and greater freedom to choose and change technology providers. Data residency matters, but it should not be mistaken for the complete measure of sovereignty.

“The UK could become better equipped for AI without necessarily becoming more productive because of it. The Government’s investment in local AI adoption and its new AI Challenge Prize is certainly positive, but access to more powerful technology does not automatically translate into better business performance or public services.

“Organisations need to move beyond adoption rates and successful pilots to understand whether AI is helping people make better decisions, reducing unnecessary work and improving how services are delivered. The additional funding for the AI Economics Institute recognises the importance of understanding AI’s economic impact. That assessment must also consider the full costs of making AI work, including data integration, running systems and human oversight, to establish whether the benefits justify the investment.

“Even the most advanced AI will struggle to deliver meaningful results without an understanding of an organisation’s data, policies and business context. Ultimately, success shouldn’t be measured by how much AI we deploy, but by what people and organisations can actually achieve because of it.”