AI for Business

Brookfield Targets £20 Billion AI Data Center at Drax Power Station Site

Brookfield is planning a large-scale artificial intelligence data center on the Drax Power Station estate in North Yorkshire, with the energy company Drax Group announcing Tuesday that it has...

Share:

Brookfield is planning a large-scale artificial intelligence data center on the Drax Power Station estate in North Yorkshire, with the energy company Drax Group announcing Tuesday that it has signed an agreement with the asset manager and its neocloud platform Radiant to develop the project jointly. Shares of London-listed Drax rose 8.8% in early trading following the news.

The first phase aims for up to 100 megawatts of capacity running from the first quarter of 2028, with another 500 MW added between 2028 and 2031. The partners then intend to secure more than 1 gigawatt at the site, though no timeline was given for that target.

Brookfield estimates total investment, semiconductors included, could pass £20 billion ($26.8 billion), with £2.5 billion earmarked for phase one. Radiant CEO Mahdi Yahya called it "a unique opportunity for the UK's AI ecosystem to build at scale," saying British researchers, startups and enterprises would gain access to frontier compute at home.

The initial phase will draw power from the grid and target commercial customers from early 2028. It still needs planning, regulatory and network approvals, a tenant must be secured, and final investment decisions have not been made, Drax said.

The announcement comes as Europe works to catch up with the U.S. on AI infrastructure, with the UK held back by high energy costs and long grid connection waits. Microsoft reportedly withdrew from a deal to anchor an Nscale data center outside London amid such delays, while Nordic countries with abundant power and land have drawn major commitments, including Google's planned €13 billion ($15.1 billion) investment in Finland.

Source: CNBC

Ready to Modernize Your Business?

Get your AI automation roadmap in minutes, not months.

Analyze Your Workflows →