First Principles Analysis

Current State of Fusion Technology

As of Early 2026

What Has Been Demonstrated#

  • NIF ignition (Dec 2022): Q_sci ≈ 1.5 (laser input to fuel capsule vs fusion output). However, the laser system consumed ~300 MJ to deliver ~2 MJ to target — system-level energy gain is deeply negative. Best subsequent shot achieved Q_sci ≈ 4.13 (8.6 MJ from 2.08 MJ laser energy).
  • Best tokamak Q_sci: JET achieved Q = 0.67 in 1997 (D-T fuel). JT-60 reached Q_ext ≈ 1.25 on D-D fuel (extrapolated to D-T). These remain the records as of 2026.
  • CFS SPARC: ~60% complete in Massachusetts. Uses high-temperature superconducting (HTS) magnets. Commercial ARC facility planned for Virginia, early 2030s, under a 200 MW Google PPA.
  • Helion Orion: Under construction in Washington state. Targeting 50 MW delivery to Microsoft by 2028 under the world's first fusion PPA. Uses D-He3 fuel cycle with direct energy conversion (non-thermal).
  • Spherical tokamak scaling: PPPL's NSTX-U, UK's MAST-U, and Tokamak Energy's ST40 have confirmed that energy confinement scales more favorably in spherical tokamak designs — potentially important for commercial viability.
  • ITER: Expected first plasma 2033–2034. Design target Q_sci = 10 (500 MW fusion from 50 MW heating). Will not generate electricity. Massive cost and schedule overruns.

What Has NOT Been Demonstrated#

  • No fusion device has ever produced net electricity (engineering breakeven, Q_eng > 1)
  • No tritium breeding blanket has been tested in a reactor environment at scale
  • No plasma-facing materials have been validated under commercial duty-cycle neutron loads
  • No fusion system has run in steady-state for operationally relevant durations
  • The DOE's own 2025 roadmap acknowledges "critical science, materials, and technology gaps — particularly in breeding and handling of fusion fuels"

Industry Landscape (2025–2026)#

  • 45+ private fusion companies globally, $9.7+ billion in cumulative private investment
  • 35 out of 45 companies anticipate commercially viable pilot plants between 2030–2035 (company self-reported; treat with skepticism)
  • 28 companies expect grid connection in the same period
  • Only 5 companies target commercialization before 2030
  • DOE Fusion S&T Roadmap (Oct 2025) targets commercial fusion power to the grid by mid-2030s, contingent on future public-private partnerships and Congressional appropriations

The gap between what has been demonstrated and what the industry promises is striking. The best tokamak Q_sci record — 0.67 — was set nearly 30 years ago. The industry needs to achieve Q_sci values of 30–50 for subsidy-free commercial viability. That's a 45–75x improvement over the current record, and no device currently under construction is designed to reach those levels.

This analysis is part of a series examining fusion energy feasibility. Sources include DOE Fusion S&T Roadmap (2025), IAEA World Fusion Outlook 2025, Fusion Industry Association 2025 report, and peer-reviewed publications in Nuclear Fusion and Fusion Engineering and Design.