From ambition to infrastructure: what it will really take for the UK to lead in green AI
Last week at the Science Museum, UKAI convened leaders from government, energy, infrastructure, data centres and AI-enabled businesses for a focused, working discussion on the future of AI in the UK.
The occasion marked the launch of UKAI’s latest report on green AI infrastructure. The event brought together senior voices from across the ecosystem, including Mike Potter, former UK Government Chief Digital Officer; Alexandra Smith, CEO of Future Plus; Lord Drayson, Chairman of Locai Labs; and Laura Sandys CBE, Chair of the UK Energy Networks Innovation Taskforce.
Ahead of the public launch, UKAI hosted a closed, Chatham House roundtable with participants working at the sharp end of AI deployment. The discussion centred on the practical realities shaping the UK’s AI trajectory: power pricing, grid constraints, investment horizons, and the physical limits of existing infrastructure. Across sectors, a consistent view emerged. If the UK is to be a serious player in AI, success will not be determined by algorithms alone. It will depend on infrastructure, market design, and the signals sent to investors. This is the central argument of the report, and it was repeatedly reinforced throughout the day.
Panellists and roundtable participants alike were clear about the UK’s underlying strengths. The country benefits from world-class research institutions and a strong pipeline of technical talent. But these advantages are not currently translating into large-scale, economically viable and environmentally sustainable AI deployment. Instead, the UK faces a familiar set of constraints: high electricity prices, fragmented infrastructure planning, and long investment timelines. While these challenges are not novel, what has changed is the speed with which AI infrastructure is being deployed globally, and the risk that the UK falls behind as a result. The report argues that this is not a failure of intent, but of coordination.
Reflecting the report, energy emerged as a defining theme across the discussions. There was a shared realism that global investors will place compute where power is affordable, reliable and predictable over the long term. Without credible, stable price signals, the UK risks becoming a location for incremental deployments rather than transformational ones. The conversation focused on how we can improve market design and stability. The logic of competing on system efficiency and value per unit of energy rather than raw scale runs through the report.
Our evening session set out the report's core recommendations, reinforcing the same conclusion: the UK's AI future will be shaped less by aspiration and more by execution. If the UK is to lead responsibly in AI, it must do so by aligning infrastrucutre, energy policy, innovation and investment. The report argues that we should treat AI not as a standalone sector, but as part of a tightly coupled overarching system whose performance depends on coherence across these domains.
To explore the findings and recommendations in full, UKAI members can access the report now.
If you are not a member and would like to request access, please contact [email protected].