Revealed: landmark Scottish AI project has no prospect of meeting renewables promise

2026-07-06 · Source: AI (artificial intelligence) | The Guardian · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure · Depth: Fundamental Awareness, medium

Summary

A Guardian investigation reveals that a landmark £8.2bn AI datacentre complex in Lanarkshire, Scotland, developed by CoreWeave and DataVita, has no realistic prospect of fulfilling its promise to be powered entirely by on-site renewables by 2030. Initially announced in January with claims of up to 1GW of "new energy infrastructure", internal documents obtained via FoI requests show government and developers privately acknowledged a "power provision" "issue". DataVita's public claims of generating 1GW from 400MW solar and 800MW wind, requiring 40-100 sq km of land, contrast sharply with its current 25MW grid draw and planning applications covering only 2 sq km. While the government now states the site will connect to the grid, this faces an 8-10 year queue, raising doubts about the UK's ability to meet the extraordinary energy demands of its AI infrastructure ambitions.

Key takeaway

For policy makers evaluating large-scale AI infrastructure proposals, you must critically scrutinize claims of on-site renewable energy provision. Projects like Lanarkshire highlight a significant disconnect between public promises and the practical realities of land availability and grid connection timelines, which can extend 8-10 years. Insist on detailed technical audits and verified land-use plans to ensure project viability and avoid designating zones based on unfeasible promotional material.

Key insights

Ambitious AI infrastructure projects often misrepresent energy self-sufficiency due to grid and land realities.

Principles

• Large-scale AI infrastructure demands immense, reliable power.
• Public project claims require rigorous technical scrutiny.
• Grid connection timelines are a major development bottleneck.

In practice

• Evaluate land requirements for claimed renewable energy generation.
• Factor 8-10 year grid connection queues into project planning.
• Cross-reference public project claims with internal documentation.

Topics

• AI Datacentres
• Renewable Energy
• Grid Connection
• Infrastructure Development
• Project Scrutiny
• Scotland

Best for: CTO, VP of Engineering/Data, Director of AI/ML, Policy Maker, Investor, Tech Journalist

Related on AIssential

• The AI Data Center Delusion — The AI data center boom faces critical power grid and supply chain constraints, forcing reliance on fossil fuels despite "Net Zero" pledges.
• Plume raises €3.3M to cut years from renewable energy development timelines — Geospatial AI and document analysis can drastically accelerate renewable energy project development timelines.
• OpenAI halts UK stargate project amid regulatory and energy price concerns — Regulatory uncertainty and rising energy costs are impacting major AI development projects.
• AI Doesn’t Run on Vibes. It Runs on an Extension Cord. — AI's future hinges on robust energy infrastructure and regulatory oversight, not just software advancements.
• Officials hugely underestimated impact of AI datacentres on UK carbon emissions - The Guardian — UK government vastly underestimated AI datacentre carbon emissions, raising climate concerns.
• Elon Musk has given up on solar power (on Earth) — Elon Musk's xAI is pivoting to natural gas and space-based solar, diverging from Tesla's terrestrial clean energy vision due to perceived AI compute growth limits.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI (artificial intelligence) | The Guardian.