China's Artificial Sun: EAST Tokamak Records and the Fusion Race 2026

On the outskirts of Hefei, capital of Anhui province, a donut-shaped machine quietly does something no other device on Earth has done: it holds a plasma at 100 million degrees Celsius — seven times hotter than the core of the Sun — for more than a quarter of an hour. The machine is EAST, the Experimental Advanced Superconducting Tokamak, nicknamed the "artificial sun" (人造太阳), and in 2025-2026 it has set a string of world records that have reframed the global fusion race. China's fusion program has moved from participant to front-runner, and the question of when — not whether — fusion power arrives increasingly runs through Hefei.

What EAST Just Did: The 1,066-Second Record

The headline result came in January 2025, when EAST sustained a steady-state, high-confinement plasma for 1,066 seconds — over 17 minutes — at an ion temperature above 100 million degrees Celsius, more than doubling its previous record of 403 seconds set in 2023. In 2026, follow-up campaigns have pushed the machine further: extended high-performance runs above the 1,000-second mark, improved control of the "H-mode" high-confinement regime, and record performance in the internal transport barrier modes that fusion researchers believe are the path to steady-state reactors.

The numbers matter because fusion's hard problem is duration. Making fusion happen in a laboratory is easy — the reaction was first demonstrated in 1932; the hard part is sustaining it: holding a 100-million-degree plasma stable and confined, without touching the reactor walls, long enough to produce net energy. Tokamaks (a Russian acronym for "toroidal chamber with magnetic coils") do this with magnetic fields, and every decade of research has been a slow climb in the product of plasma density, temperature, and confinement time — the "triple product" that determines whether a fusion device is viable. EAST's achievement is the best demonstration yet that the steady-state regime required for a power plant — running for hours, not seconds — is physically reachable.

The Machine: Hefei's Superconducting Workhorse

EAST is a medium-sized tokamak that entered operation in 2006, built by the Institute of Plasma Physics at the Chinese Academy of Sciences (ASIPP) in Hefei. Its signature feature is full superconducting magnets — both toroidal and poloidal field coils — making it the first fully superconducting tokamak ever built, a design that allowed the long-pulse operation no copper-coil machine can sustain. The machine's name is an acronym (Experimental Advanced Superconducting Tokamak) with an apt meaning: "east" is the direction of the sunrise, and Chinese scientists have leaned into the metaphor of an artificial sun rising.

EAST's role in the international fusion ecosystem is dual. It is a national flagship, training the cohort of Chinese plasma physicists who now staff a dozen newer machines, and it is an international testbed: EAST has hosted collaborations with ITER partners worldwide, and its results inform the design of ITER, the 35-nation megaproject in France that aims to demonstrate net-positive fusion energy by the late 2030s. China is ITER's largest single-country contributor after the European Union, manufacturing key components — including the toroidal field magnet systems — in domestic factories.

The New Machines: From EAST to the Burning Plasma Era

EAST is not the end of China's ambitions; it is the rehearsal. The next step is CFETR, the China Fusion Engineering Test Reactor, a much larger device (roughly the scale of ITER) designed to bridge the gap between experimental machines and a demonstration power plant. CFETR has been in design for years; in 2025-2026, site selection and engineering design progressed with reports pointing to a start of construction before 2030 and operation in the mid-2030s, targeting a demonstration of net electrical output — the final milestone before commercialization.

The broader Chinese program has also gone industrial. The "artificial sun" narrative has attracted serious private capital: at least two domestic startups — Energy Singularity (能量奇点) and Startorus Fusion — have raised hundreds of millions of dollars to develop alternative designs (a spherical tokamak and a high-field approach respectively), and provincial governments are competing to host fusion research parks. China's state strategy, embedded in the 15th Five-Year Plan deliberations, treats fusion as the long-term endgame of the energy transition: the country that masters fusion holds a license to effectively unlimited clean energy.

The Global Race: Who Is Actually Ahead?

The fusion race in 2026 has four main players. The United States leads in private fusion investment and in the "burning plasma" milestone: the National Ignition Facility (NIF) at Lawrence Livermore achieved ignition — more energy out than in — via laser fusion in December 2022 and has repeated the feat several times since, though inertial fusion's route to power plants remains longer and costlier than magnetic confinement. Europe hosts ITER, the largest public project, but ITER has been plagued by delays and cost overruns (budget estimates now exceed $22 billion), with first plasma pushed toward 2034-2035. Japan operates JT-60SA, the world's largest superconducting tokamak before CFETR, and hosts ITER's sister project.

China's position is distinctive: unmatched public funding discipline, a young and expanding workforce (ASIPP alone trains hundreds of PhDs per year), and — critically — the world's fastest industrial learning curve, the same advantage that drove its solar and EV industries. International observers in 2026 increasingly describe the fusion race as "China vs. everyone else on cost and schedule": Chinese analysts project CFETR-class machines built at a fraction of ITER's cost and timeline, and there is little in the historical record to suggest the projection is fantasy.

Why Fusion Matters: The Energy Endgame

Fusion's appeal is almost unfair. The fuel — deuterium and tritium, isotopes of hydrogen — is abundant: deuterium is extracted from seawater, and tritium can be bred from lithium inside the reactor itself. A single gram of fusion fuel yields roughly the energy of 8 tons of coal, with no carbon emissions, no long-lived radioactive waste (the structural materials become radioactive but manageable for ~100 years rather than millennia), and no risk of a runaway reaction — a fusion plasma that loses confinement simply extinguishes. It is the only energy source that could plausibly replace fossil fuels at civilization scale.

For China specifically, fusion is the capstone of a strategy that is already the most aggressive clean-energy buildout in history. Beijing's climate pledge — peak carbon before 2030, neutrality by 2060 — cannot be met by wind and solar alone, because their intermittency requires either massive storage or a firm low-carbon backbone; nuclear fission supplies that backbone today, and fusion would supply it without fission's waste and siting politics. The state's framing is explicit: fission is the bridge, fusion is the destination.

The Skeptics' Case: Fusion Is Always 30 Years Away

The counterargument is as old as fusion research: commercial fusion has been "thirty years away" since the 1950s, and the joke remains the most cited line in the field. The honest assessment in 2026 is that the remaining challenges are real: materials that survive 14 MeV neutron bombardment (the fusion reaction's neutron flux is far more damaging than fission's), tritium breeding at scale, the heat-exhaust problem (handling power densities that would melt any solid surface), and the economic question — whether a fusion plant can ever beat solar-plus-storage on cost, the technology it is racing against. Even optimistic roadmaps put the first commercial plants in the 2040s at the earliest, and fusion's history argues for adding a decade to any timeline.

But the physics community's tone has shifted. The NIF ignition results and EAST's steady-state records have changed the argument from whether to when, and China's industrial model — state coordination, serial manufacturing, compressed schedules — is precisely the model that has repeatedly closed "impossible" gaps in clean technology. The artificial sun in Hefei is no longer a curiosity; it is the most visible symbol of a conviction now held across the world's energy establishment: the fusion race is real, and China intends to win it.