Japan’s National Institute for Quantum Science and Technology (QST) has accelerated its timeline for a prototype fusion reactor. This is in line with the Japanese government’s revised Fusion Energy Innovation Strategy (June 2025), which seeks to bring fusion power online decades earlier than the original 2050 goal.
To meet the 2038 target, QST has downsized the reactor’s design to be approximately the same size as ITER (30m in diameter and height) rather than 1.4 times larger as previously planned. The development is structured into three distinct phases:
- Phase 1: Testing integrated power generation between the fusion reactor and facility.
- Phase 2: Testing tritium breeding technology within the reactor.
- Phase 3: Achieving stable, long-term operations.
QST aims to generate power for several minutes with a net output of 10,000 kilowatts (10 MW) within one year of completion. The ultimate goal for this prototype is a steady-state output of 100,000 kilowatts (100 MW).
Construction is slated to begin in the late 2020s. The facility is expected to require at least 1 square kilometre of land. Both the Ibaraki and Aomori prefectural governments have expressed interest I in hosting the site.
The project is a pillar of Prime Minister Sanae Takaichi’s growth strategy, which labels fusion as one of 17 priority investment areas. It runs alongside the FAST project (Fusion by Advanced Superconducting Tokamak), a separate public-private venture aiming for its own demonstration by the end of the 2030s.
The JT-60SA (Super Advanced) tokamak in Naka, Japan, serves as the critical “testbed” for QST’s prototype reactor. As the world’s largest operational superconducting tokamak, JT-60SA bridges the gap between today’s experiments and future power-generating reactors such as ITER and the planned Japanese prototype.
JT-60SA is specifically designed to solve the physics hurdles that traditional experiments cannot reach due to their smaller scale or shorter pulse durations. These include:
- Steady-State Operation: Unlike previous reactors that run in short bursts, JT-60SA targets sustained plasma for 100 seconds. This is long enough to study “current diffusion,” a process essential for the continuous, 24/7 operation required by the 2038 prototype.
- High-Pressure Stability: The facility tests “high-beta” plasma regimes where the plasma pressure is extremely high relative to the magnetic field strength. Mastering this allows for the more compact and efficient reactor designs planned for the late 2020s construction.
- AI-Driven Control: In early 2026, QST began using artificial intelligence and novel computational techniques on JT-60SA to automate plasma start-up and stabilisation. This AI “autopilot” is intended to be a core feature of the 2038 reactor’s control system.
Following major upgrades throughout 2024 and 2025, JT-60SA entered “integrated commissioning” in late February. The device was recently fitted with high-speed control coils and a 4-tonne X-ray Imaging Crystal Spectrometer (XICS) from the US Princeton Plasma Physics Laboratory to provide ultra-precise measurements of plasma internal activity. Full-power plasma heating experiments are scheduled for mid-to-late 2026, aiming to reach temperatures of 200m degrees Celsius.
JT-60SA is the primary training ground for the “ITER Generation” of engineers. Moreover, the expertise gained in operating its superconducting magnets and complex cryoplants is being directly transferred to the team that will build and run the 2038 prototype.
