For many years, humans have pursued the goal of harnessing star power to produce electricity on Earth, with the achievement always seemingly on the horizon. Today, several startups are nearing that goal, racing to create fusion reactors capable of supplying power to the grid.
Fusion startups have attracted over $10 billion in investment, with more than a dozen securing more than $100 million each. Major funding rounds have closed recently, fueled by increasing energy demands from data centers and the progress fusion startups are making.
The essence of fusion power lies in utilizing the energy from atomic fusion to generate electricity. Humans have known how to fuse atoms for years, from the hydrogen bomb—an example of uncontrolled nuclear fusion—to the various fusion devices built in labs worldwide. Experimental fusion devices have managed to control nuclear fusion, with one achieving more energy output than input. However, none have produced a surplus sufficient for a power plant.
To address this, fusion startups are exploring various methods, with experts debating which might succeed. The industry is still nascent, so outcomes remain uncertain.
Here’s an overview of primary fusion power methods.
**Magnetic confinement**
Magnetic confinement is a widely used technique that employs strong magnetic fields to contain plasma, the core of a fusion device. These magnets are extremely powerful; Commonwealth Fusion Systems (CFS), for instance, is constructing magnets that can generate 20 tesla magnetic fields, significantly stronger than those in typical MRI machines. These magnets, made from high-temperature superconductors, require cooling to -253˚ C (-423˚ F) with liquid helium.
CFS is developing a demonstration device called Sparc on an accelerated timeline in Massachusetts, slated for activation in late 2026. If successful, the construction of Arc, a commercial-scale power plant in Virginia, will commence in 2027 or 2028.
There are two main types of magnetic confinement fusion devices: tokamaks and stellarators.
Tokamaks, theorized by Soviet scientists in the 1950s, have been extensively studied. They come in two shapes—a doughnut with a D-shaped profile and a spherical form with a central hole. The Joint European Torus (JET) and ITER are notable tokamaks; JET operated in the UK from 1983 to 2023, while ITER is scheduled to start in France in the late 2030s. The UK’s Tokamak Energy is developing a spherical tokamak, with its ST40 experimental machine undergoing upgrades.
Stellarators, similar to tokamaks in shape but with twists to accommodate plasma behavior, are the other principal magnetic confinement device. Wendelstein 7-X, a stellarator with modular superconducting coils, has operated in Germany since 2015. Several startups, including Proxima Fusion, Renaissance Fusion, Thea Energy, and Type One Energy, are also working on their stellarator designs.
**Inertial confinement**
The second major fusion approach is inertial confinement, which compresses fuel pellets until atomic fusion occurs. Most designs use laser pulses to compress these pellets, with multiple lasers converging from all angles simultaneously.
Inertial confinement is the only method that has surpassed scientific breakeven—where output exceeds input—achieved at the National Ignition Facility (NIF) at Lawrence Livermore National Laboratory in California. However, these measurements do not account for the electricity used by the facility itself.
Despite this, nearly a dozen startups are pursuing inertial confinement in their reactor designs. Notable companies include Focused Energy, Inertia Enterprises, Marvel Fusion, and Xcimer, all utilizing lasers. Two companies, however, are diverging from laser use: First Light Fusion is using pistons, while Pacific Fusion explores electromagnetic pulses.
**More to come**
These are the primary fusion power approaches, though not the only ones. More information on alternative designs such as magnetized target fusion, magnetic-electrostatic confinement, and muon-catalyzed fusion will be provided soon.