UK AI Sovereignty: How will Telecoms Power the Future?
Since taking office in the summer of 2024, the UK’s Labour government has repeatedly reiterated its ambition to make the country a global AI superpower. Prime Minister Sir Keir Starmer has set out a clear vision:
“Artificial intelligence will drive incredible change in our country,” he says. “From teachers personalising lessons, to supporting small businesses with their record-keeping, to speeding up planning applications, it has the potential to transform the lives of working people.”
The Prime Minister’s comments came against a backdrop of high-profile announcements. During the state visit of US President Donald Trump, major American technology companies committed new investments to support the UK’s AI future.
While the financial boost is welcome, government leaders are equally focused on digital sovereignty, ensuring that Britain retains control over critical digital infrastructure, data and the deployment of advanced technologies.
Understanding digital sovereignty
Digital sovereignty refers to a nation’s control over essential digital infrastructure, including data centres and AI platforms. For Jonny Williams, Chief Digital Adviser for the UK at Red Hat, sovereignty is a prerequisite for long-term leadership in AI.
“Digital sovereignty isn’t a luxury – it’s an urgent, practical requirement if the UK is to lead in AI,” he explains. “AI sovereignty allows the UK to have greater control over who accesses local data and how it is used.”
He notes that sovereignty extends beyond hardware. Data ownership and stewardship are equally critical to securing national interests while enabling businesses and public services to innovate and thrive.
Open source as a pathway
One clear route to sovereignty lies in open source technologies. Red Hat’s collaboration with Great Wave AI is demonstrating how this approach can work in practice.
Jack Perschke, Co-Founder and CEO of Great Wave AI, compares large language models (LLMs) to raw commodities: “LLMs are like crude oil,” he says. “They are valuable, but they need refinement if they are to become usable, useful products.”
His company’s platform functions as that refinery, transforming base AI capabilities into tailored tools for organisations. Security is a central consideration: “In an era where prompts and responses can’t be encrypted inside the model itself, the safest option is often running it in your own environment,” Jack adds.
Jonny agrees. “Open source AI provides the flexibility and transparency for the UK to host, audit and scale AI entirely on-shore,” he says. “This is critical for establishing a secure and trusted AI ecosystem.”
The three-layer adoption challenge
According to Jack, AI adoption in the UK falls into three layers. At the top, enterprise systems, such as Salesforce, are embedding generative AI into their workflows. At the bottom, productivity tools such as Microsoft Copilot and legal assistants like Harvey are gaining traction.
However, the most significant opportunity sits in the middle layer: bespoke, organisation-specific, multi-agent data transformation.
Jack estimates adoption here “at a fraction of 1%,” despite its potential to deliver transformative value. He warns that without orchestration platforms, enterprises risk creating fragmented “AI sprawl” with little governance or standardisation.
Building homegrown capabilities
Foreign investment from companies such as NVIDIA, Microsoft and Google has been celebrated as a milestone. Yet both Jack and Jonny stress the importance of cultivating domestic capabilities and talent to deliver sovereign AI at scale.
“AI sovereignty requires an investment into homegrown talent and companies that can transition the UK from pilots to sovereign AI at scale,” says Jonny.
The UK Home Office’s £3 million (US$4 million) tender for discovery and design work highlights the scale of costs involved in building capabilities from scratch. Platforms such as Great Wave’s offer a proven alternative, aligning with regulatory demands while keeping data and decision-making on-shore.
Strategic considerations for organisations
For businesses at the start of their AI journey, Jack recommends centralised oversight: “Bring all AI initiatives under one working group and order them by complexity to secure early wins before progressing to use cases harder.” He highlights that “agent orchestration is inevitable” and advises firms to “decide early whether to build or buy it.”
Jonny concludes that success will mean combining “world-class infrastructure, open source capabilities and local expertise” to deliver AI innovation that is “not only powerful, but secure, transparent and trusted by the public.”
For the UK telecommunications sector, the message is clear. AI adoption and digital sovereignty are intertwined and rely on resilient, high-capacity networks. Telecom providers will be the unseen enablers of a sovereign AI future.
