#156 - Dr Tim Gregory - Why The "Energy Transition" Threatens Modern Civilisation

What It Covers

Nuclear chemist Dr. Tim Gregory joins Peter McCormack to argue that the UK's net zero strategy is physically undeliverable using wind, solar, and batteries alone. Gregory presents data on nuclear power's safety record, land efficiency, and carbon footprint, contrasts UK infrastructure failures with France's 56-reactor buildout, and explains why small modular reactors represent the most viable path forward for both energy security and AI data center demand.

Key Questions Answered

• •Grid Storage Reality: The UK's entire grid-scale battery capacity covers just over one hour of national electricity demand. On low-wind days, wind turbines operate at roughly 10% of their 30-gigawatt installed capacity — meaning 27 gigawatts simply disappears. Without gas as backup, and with no credible storage solution scaled to cover multi-day wind droughts, rolling blackouts by 2030 become a structural probability, not a fringe prediction. Policymakers have systematically underestimated this gap.
• •Nuclear Land Efficiency: A single Hinkley Point C station, covering roughly a quarter of a square mile, will supply 8–9% of UK electricity for approximately 60–80 years with the same carbon footprint as wind. Fourteen such stations would fully decarbonize the UK grid. France built 56 reactors between 1973 and 1999, generating 80% of national power at peak. The land, material, and mining requirements per terawatt-hour are lower for nuclear than any other energy source.
• •Biomass Accounting Fraud: Drax Power Station in Yorkshire is currently the UK's single largest source of CO₂ emissions — and it burns wood chips classified as renewable energy. The process involves felling North American forests, drying and chipping the wood, shipping it across the Atlantic on diesel vessels, and rail-freighting it to Yorkshire for combustion. This million-year-old technology receives renewable subsidies while producing more CO₂ per unit of energy than coal, representing a direct policy failure.
• •Nuclear Death Toll Calibration: Approximately 7 million people die annually from air pollution caused by burning fossil fuels and biomass — equivalent to the entire Chernobyl death toll every 30 minutes. Chernobyl's actual verified death toll sits between 200 and 500 people, primarily thyroid cancer cases in children exposed to radioactive iodine-131 through contaminated milk. The linear no-threshold radiation model, which generates estimates of tens of thousands of Chernobyl deaths, lacks statistical support at low radiation doses.
• •UAE Nuclear Benchmark: The UAE went from zero nuclear capacity in 2020 to 25% of national electricity from four Korean-built reactors by 2024, completing construction in eight years at a total cost of roughly $20 billion — approximately $5 billion per reactor. Hinkley Point C, a comparable two-reactor project, will likely exceed £50 billion by completion. The cost differential traces directly to regulatory frameworks, not reactor technology, as outlined in the 2024 Fingleton Nuclear Regulatory Task Force Review.
• •Fuel Cycle Economics: Current nuclear plants use an open fuel cycle — uranium is used once and stored. Closed-cycle reactor technology, demonstrated since the 1950s, recycles spent fuel and increases its energy yield by over 100 times. Using existing global nuclear waste stockpiles plus known uranium and thorium reserves in a closed cycle would provide approximately 1,000 years of energy supply. Development stalled when large uranium discoveries in the 1970s removed the economic incentive to recycle.
• •Small Modular Reactors and AI Demand: The International Energy Agency projects that by the mid-2030s, AI data centers will consume electricity annually equivalent to Japan's entire national usage — more than the whole African continent. Microsoft, Meta, and Amazon are already investing in small modular reactor technology specifically to power data centers. SMR units can be assembled in 12–18 months, fit within a data center footprint, and are manufactured on production lines rather than built on-site, opening nuclear deployment to private sector competition for the first time.