- High energy costs are forcing AI workloads out of the UK
- Cheaper electricity becomes the deciding factor for AI deployment
- U.S. infrastructure advantage accelerates shift of AI workloads
UK businesses pay more than four times as much for electricity as their US counterparts, and the AI industry is taking note.
According to CUDO calculation20% of UK businesses have already moved their AI workloads out of the country due to high energy costs.
The gap between where businesses want to use AI and where they can actually execute it is growing rapidly.
Article continues below
Why UK AI companies are looking abroad for cheaper energy
“What we’re seeing is a growing tension between where companies want to use AI and where they actually can,” said Matt Hawkins, CEO of CUDO Compute.
“If it is cheaper or easier to manage workloads elsewhere, they will move, regardless of sovereignty ambitions.”
A third of UK organizations say energy costs limit their ability to scale their AI operations, according to a survey of more than 700 senior AI decision-makers.
When asked which markets seemed most attractive for new AI cluster capabilities, 72% of UK respondents cited the US.
India followed with 62%, Eastern Europe with 58% and China with 55%. Western Europe and the Nordic countries performed lower, at 45% and 44%, respectively.
The message is clear: cost and performance still trump sovereignty for 43% of organizations when deciding where to deploy their AI tools.
While 46% of UK organizations say geopolitical instability is pushing them to keep workloads in their home markets, the economic pressure to offshor is intense.
Nearly one in three UK businesses say they are actively considering moving their workloads overseas due to geopolitical pressures.
Nearly half say data sovereignty, regulatory compliance or national security concerns shape their AI deployment strategy.
Still, 32% of AI-focused companies say they would consider moving their workloads overseas due to energy costs, compared to 18% of traditional companies.
Companies running the most compute-intensive workloads are also the most likely to look beyond the UK as economic conditions tighten.
The UK has ambitions and policies towards AI sovereignty, but there is a clear disconnect between these goals and what is actually available.
Organizations want to build in the UK, but they need the infrastructure to do so. The countries that solve this problem first will shape the future of AI, and the UK still has room to take the lead, but it must act quickly.
The research exposes a harsh truth for UK policymakers. Talk of AI sovereignty means nothing without the power infrastructure to support it.
The US, with falling energy prices and aggressive construction of AI-ready data centers, is already reaping the rewards of the UK’s inaction.
With every week that passes without significant progress in energy costs and network capacity, more UK AI workloads will migrate overseas.
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds.
