The Nuclear Regulatory Commission's proposed fusion framework awaits its first real test—no company has yet filed an application under the new structure. Meanwhile, technical progress accelerates: Helion closed a $465 million Series G, Xcimer brought its Phoenix laser online, Pacific Fusion validated pulsed-power modules delivering 440 gigawatts peak output, and Thea demonstrated full-field planar shaping coils. The mismatch is stark: hardware and physics are advancing faster than the regulatory infrastructure needed to turn experimental machines into licensed power plants.
Tennessee's regulatory framework remains unproven until Type One Energy submits its first license application—and that filing has yet to materialize.
The NRC's framework closed public comment on May 27, putting final publication on track for October 2026. The proposal regulates fusion under the existing byproduct material approach in Part 30, avoiding the heavier utilization-facility licensing regime that governs fission reactors. Commonwealth Fusion Systems filed a grid interconnection application with PJM for its proposed 400 MW ARC plant in Chesterfield County, Virginia—years ahead of construction completion—but the regulatory path to operate remains separate and still depends on finalization of the federal rule. The NRC process is progressing, but until the framework faces real-world application reviews, its practical sufficiency remains untested.
Helion Energy announced a $465 million Series G on June 4, led by Thrive Capital at a $15.5 billion post-money valuation, bringing lifetime capital raised to slightly over $1.5 billion. That makes Helion the second-largest privately funded fusion company, though still roughly half of Commonwealth Fusion Systems' nearly $3 billion total. The Series G is the seventh-largest single round in fusion history and the largest so far in 2026. Helion's Polaris prototype became the first privately funded machine to operate with deuterium-tritium fuel and exceeded 150 million degrees Celsius plasma temperature, while construction continues on Orion, the company's first fusion power plant in Malaga, Washington.
The funding underscores investor confidence in Helion's pulsed non-ignition magnetic target approach, but converting technical progress into regulatory approval and grid-connected operation remains the long pole.
Xcimer Energy announced on June 3 that its Phoenix laser system has started operations in Denver, the largest privately owned laser system in the world. Phoenix is proof-of-concept for Xcimer's krypton fluoride excimer laser architecture using Stimulated Brillouin Scattering to compress microsecond-long pulses into nanosecond timescales required for inertial confinement fusion. The system demonstrates integrated excimer amplification and SBS pulse compression, rebuilding industrial capabilities around large-scale electron-beam-pumped excimer lasers that largely disappeared after the Cold War. Xcimer recruited engineers, physicists, and pulsed-power specialists from national laboratories, the National Ignition Facility, and the U.S. Naval Research Laboratory to transfer critical technical knowledge to a new generation.
The company's roadmap targets Anvil, a 200-kilojoule commercial-scale amplifier, by 2028, followed by Vulcan (4 to 12 megajoules targeting wall-plug breakeven) in the early 2030s, and Athena, a commercial power plant, by the mid-2030s. Xcimer reportedly expanded its leadership team recently, creating roles including a Vice President of Defense, and has dozens of open positions across engineering, operations, and science.
Pacific Fusion completed its second set of technical milestones, demonstrating 440 gigawatts peak output from a scaled prototype of its Impedance-Matched Marx Generator—roughly one-third of a full module and the highest peak power single-step pulsed-power driver ever demonstrated. It delivered approximately 1.1 megavolts peak voltage in 80 nanoseconds with roughly 2-nanosecond timing jitter. Pacific Fusion's concept scales the Sandia Lab Z-Machine approach both up (156 simultaneous current pulses) and down (using smaller, off-the-shelf capacitors) to make the system more producible and more powerful. Each full module, shipping-container-sized, will deliver more than a terawatt of peak power in pulses lasting about 100 nanoseconds. The company validated a low-cost trigger system capable of synchronizing all stages into a single coherent pulse through more than 1,000 qualification shots.
Pacific Fusion is on track to achieve net facility gain—more energy output than initially stored in the machine—by 2030, with manufacturing operations scaling at facilities in San Leandro, California, and Los Lunas, New Mexico.
Thea Energy announced successful operation of its first full-size, full-current, full-field stellarator planar shaping coil, generating magnetic fields over 6 Tesla at 20 degrees Kelvin in line with specifications for the company's upcoming Eos integrated stellarator. The planar coil architecture addresses the traditional complexity of stellarator design, where non-axisymmetric three-dimensional coils are expensive to fabricate and difficult to tune. Thea's "software-controlled" magnet approach uses arrays of simple planar coils to create arbitrary magnetic fields, allowing the stellarator to produce stable confinement without custom 3D magnets. The company closed an oversubscribed $100 million Series B on May 27, led by Thomas Tull's US Innovative Technology Fund, to expand magnet manufacturing capacity and begin construction of Eos later this year.
Thea also received DOE certification of its Helios power plant preconceptual design milestone, the first awardee to earn that distinction. The company is already in discussions with more than a dozen power offtakers, hyperscalers, and utility partners ahead of constructing its first Helios commercial power plant before the end of the decade.
China's HL-3 tokamak research team published simulation results exploring negative triangularity plasma confinement, finding that the configuration's L-mode performance can match positive triangularity H-mode confinement levels under comparable plasma current, toroidal field, auxiliary heating, and line-averaged density. The OMFIT integrated simulation platform showed that negative triangularity achieves normalized beta 1.08 times that of positive triangularity, energy confinement time 1.18 times longer, confinement factor H98 of 1.35 times, and thermal stored energy 1.20 times higher. The primary mechanism is reduced total energy transport in the normalized minor radius range of approximately 0.2 to 0.6. Turbulent transport dominates neoclassical transport by one to two orders of magnitude, and negative triangularity suppresses turbulent transport compared to positive triangularity.
Linear instability analysis at normalized minor radius 0.32 shows positive triangularity exhibits electron temperature gradien