Humanity runs on about twenty trillion watts, more than at any time in its history.
The Sun delivers nearly nine thousand times that to the Earth, every second, for nothing.
We are not short of energy. We are short of ways to catch it, keep it, and stop wasting it.
In 2023 the world used 620 exajoules of primary energy, the most ever. Averaged over the year, that is a continuous flow of about 19.6 trillion watts, or roughly 2,400 watts for every person alive: the power of a kettle, running day and night.
Fossil fuels supplied 81.5% of it. That share has barely moved in decades, because clean sources have so far mostly met new demand instead of replacing old supply. Emissions from energy passed 40 billion tonnes of carbon dioxide for the first time.
Four times as much as in 1965, growing about 2% a year.
The left side is the world's energy supply in 2023. The right side is the uncomfortable part: burning a fuel to make motion or electricity throws most of the energy away as heat. The split shown is measured each year for the United States and is similar elsewhere. Electric motors and heat pumps waste far less, so a clean system needs less primary energy to do the same work.
Source: Energy Institute Statistical Review 2024 for the mix (fossil total confirmed; shares rounded); Lawrence Livermore National Laboratory energy flow charts for useful and rejected energy · cited from memory, to verify
Two technologies have fallen in price faster than almost anyone forecast. A lithium-ion battery pack cost about 1,436 dollars per kilowatt-hour in 2010 and 115 dollars in 2024, a fall of 92%. Solar panels followed the same pattern over a longer period: each doubling of the number built has cut the price by about a fifth.
Neither needed a scientific breakthrough. They got cheap because they are manufactured, and manufactured things improve with every unit made. That is the most hopeful fact in this field: the cost of clean energy is something we can drive down on purpose.
Twelve times cheaper in fourteen years.
Source: BloombergNEF, Lithium-Ion Battery Price Survey 2024 (2010 and 2024 figures confirmed)
Sun and wind do not arrive when they are needed. A system built on them has to move energy across hours, seasons and continents, and has to survive a still, dark week in winter. Storage, long transmission lines, and firm sources such as nuclear, geothermal and eventually fusion are what turn cheap energy into reliable energy.
The second problem is waste. In a typical industrial economy, about two-thirds of the primary energy that goes in never does useful work: it leaves as heat from engines, power stations and furnaces. Using energy better is as large a resource as making more.
Fusion deserves a plain statement. In 2022 a laser experiment released more energy from its fuel than the laser delivered to it, for the first time. That is a scientific milestone, and it is a long way from a power station.
Sunlight reaching the Earth is about 9,000 times what humanity uses. A Type I civilisation would use about 500 times today's supply.
Source: Derived: solar constant 1,361 W/m² over the Earth's cross-section; Sagan's Type I threshold of 10¹⁶ W; Energy Institute 2024 for current use · cited from memory, to verify
Each mark is a thousand times the one before. A person is a hundred-watt machine. Our whole civilisation is two hundred billion of those. And the sunlight falling on the planet is nine thousand civilisations more. The right-hand end of this ruler is where the Kardashev page begins.
Source: Derived from standard physical values; Energy Institute 2024 for humanity · cited from memory, to verify
We take the position that energy use should grow, not shrink, and that it should stop costing the climate. Almost every measure of human welfare rises with access to energy. The task is to make the clean kind so cheap that it wins on price everywhere.
Optimisation is the other half of the title for a reason. A grid with millions of generators, batteries and flexible loads is a control problem too large for people to run by hand. The same is true of a chemical plant or a city's heat. We intend to treat the planning and operation of energy systems as an open scientific problem, with shared models and shared data.
This is the page where the Kardashev scale stops being an abstraction. The road to Type I is this road.
The small white square is the area of solar panels that would supply all the energy humanity uses today: about 490,000 square kilometres, close to the area of Spain. It assumes panels deliver an average of 40 watts per square metre around the clock, which is typical of good sites today. Nobody would build it in one place. The point is that the land is not the constraint. Storage, transmission and manufacturing are.
Source: Derived: 19.6 TW ÷ 40 W/m². Areas from standard geographic references · cited from memory, to verify
You run an electricity system. Two of the four days are cloudy and the wind comes and goes. Choose how much solar, wind, firm power (nuclear, geothermal or fusion, which run whatever the weather) and storage to build, in multiples of average demand. Hatched hours are blackouts.
This is a statement of intent, not a record of achievement. Each step carries a label for how firm it is. Nothing marked proposed or speculative has been done yet.
From the physics of a solar cell to the operation of a continental grid, as working simulations.
Free, well-documented tools and data so that any country or city can plan a reliable low-carbon system.
Control software, proven in pilots, that operates storage and flexible demand better than rules of thumb.
Contribute to making advanced geothermal, next-generation fission and fusion buildable and affordable.
Clean supply several times today's, shared widely enough that energy poverty ends. The first rungs toward Type I.
Every claim on this page is open to challenge in the Agora, in the Energy & Earth room.
Argue it in the Agora ↗