2025 has emerged as a pivotal year for fusion energy, with two landmark developments—Alphabet’s 200-megawatt (MW) fusion power purchase and China’s breakthroughs at the Bao’an Eastern Superconducting Tokamak (BEST)—signaling a decisive shift from lab-scale experiments to commercial viability. For decades, fusion has been hailed as the “holy grail” of clean energy, promising limitless power with zero carbon emissions and minimal waste. This year, those aspirations are no longer distant: corporate demand and technical leaps are converging to turn fusion into a near-term solution for climate and energy security.

In July 2025, Alphabet made headlines with a historic agreement to buy 200MW of fusion power from Canada’s General Fusion, set to start delivery by 2030. Valued at over two hundred million United States dollars annually once operational, the deal will power Alphabet’s data centers across the U.S. Pacific Northwest—a critical step toward the tech giant’s 2027 goal of carbon-neutral operations. Unlike previous fusion agreements, which focused on small pilot projects (under 10MW), Alphabet’s commitment validates fusion as a reliable baseload energy source. General Fusion’s magnetized target fusion technology, which compresses plasma to fusion conditions using mechanical force, aligns with Alphabet’s need for continuous, low-cost power: simulations show the 200MW supply will cut the data centers’ emissions by 40% and reduce energy costs by 15% compared to natural gas-fired generators. “This isn’t just a procurement deal—it’s a vote of confidence in fusion’s ability to scale,” said Ruth Porat, Alphabet’s Chief Financial Officer, adding the company may expand purchases to 500MW by 2035.

While Alphabet drives demand, China’s BEST facility is advancing fusion’s technical backbone. In May 2025, researchers at the Bao’an Eastern Superconducting Tokamak—a fully superconducting fusion device—announced a breakthrough: sustaining a 150-million-degree Celsius plasma (ten times hotter than the Sun’s core) for 1,020 seconds. This far surpasses South Korea’s 2023 record of 403 seconds and addresses a key barrier to fusion: maintaining stable plasma long enough to generate net energy (more power out than in). BEST’s design, which uses niobium-titanium superconducting coils to contain plasma, also reduces energy loss by 30% compared to non-superconducting tokamaks like the U.S. DIII-D. Crucially, China has opened BEST’s data to international partners, including the ITER project (the world’s largest fusion experiment), fostering global collaboration. “BEST isn’t just a Chinese achievement—it’s a tool for the entire fusion community,” said Dr. Wang Wei, BEST’s lead scientist, noting plans to reach 3,000 seconds of plasma confinement by 2026.

These developments are reshaping the global fusion landscape. Alphabet’s deal has spurred copycat moves: Microsoft announced in August 2025 it is negotiating a 150MW supply with U.S.-based Commonwealth Fusion Systems, while Amazon Web Services is exploring fusion for its cloud infrastructure. Global fusion investment hit a record eighty billion United States dollars in 2025, up 60% from 2024, as governments and firms rush to capitalize on the momentum.

Challenges persist, however. Scaling fusion from BEST’s 1MW experimental output to 200MW commercial plants requires solving material hurdles—such as developing reactor walls that withstand high-energy plasma. Superconducting material supply chains are also tight: current niobium-titanium production can support only 10 large-scale fusion projects by 2030. Additionally, regulatory frameworks for fusion plants are still in the works, with the U.S. Nuclear Regulatory Commission aiming to finalize safety standards by 2028.

Yet 2025’s milestones are transformative. Alphabet’s 200MW purchase proves market demand exists, while BEST’s leaps show fusion is technically feasible. For a world grappling with climate change, these developments offer a glimpse of a future where clean, abundant fusion power runs cities, data centers, and industries—no carbon, no fuel shortages. This year may well be remembered as when fusion’s “holy grail” finally became attainable.

（Writer：Weink）