On September 15, media reported that Helion Energy had added $35 million to its Series G round launched in June, bringing the total to $500 million. For a fusion company, that is a remarkably strong report card. Its total staff has surpassed 800, its new Seattle office has just opened, and Orion, its first commercial fusion plant in Malaga, Washington, is under construction. CEO David Kirtley said on LinkedIn that the round added investors including crossover funds, pensions, and sovereign wealth funds, demonstrating growing demand for and confidence in fusion energy.
But just a week before the funding news, an Axios report revealed the other side: the company that once promised to supply Microsoft with fusion power in 2028 is quietly rewriting its milestones. Its website deleted the phrase “power generation in 2028”; executives publicly acknowledged that 50 MW full-power operation cannot be achieved in 2028; and the goal of an earlier prototype was downgraded from “net electricity” to “demonstrating electricity from fusion.”
On one side, capital enthusiasm keeps rising; on the other, technical uncertainty keeps intensifying—two forces are pulling Helion in opposite directions.
From Space Propulsion to Ground Power
To understand why Helion can attract both the smartest money in Silicon Valley and the most skepticism, one must return to its starting point.
Kirtley, one of Helion’s founders, earned a PhD in plasma physics from the University of Washington and then joined a company called MSNW. The company’s research direction sounded more like science fiction: using nuclear fusion technology to power spacecraft. It was there that Kirtley and his colleagues developed a bolder idea—if fusion can work in space, why not use it directly to generate electricity?
In 2013, Kirtley teamed up with three MSNW colleagues—John Slough, Chris Pihl, and George Votroubek—to found Helion Energy in Everett. From day one, they chose the FRC route, diverging from the mainstream, which destined every subsequent step to be accompanied by controversy.
A non-mainstream route plus an aggressive timeline allowed Helion to tell a more imaginative story than its peers. When almost all other fusion companies set their net electricity goals for the 2030s at the earliest, Helion once claimed that Polaris could achieve a net electricity demonstration as early as 2024. These bold promises attracted OpenAI founder Sam Altman—he personally holds about one-third of Helion’s equity and left the company’s board in March 2026, reportedly to clear obstacles for business synergies between Helion and OpenAI.
The Quietly Rewritten 2028
Helion’s most widely known promise is its power purchase agreement with Microsoft. In 2023, the company announced it would supply Microsoft with fusion power in 2028, with Constellation Energy serving as the power seller. The 50 MW Orion plant is sited in central Washington, and construction officially began in July 2025. At the time, Helion wrote in a press release that the company was “still on track to meet this goal.”
But change is happening—quietly.
In a September 8 report, Axios pointed out that Helion had this year removed from its website the statement that Orion would begin generating power in 2028. Today, the Orion project page only says the plant will “begin initial operations in 2028,” without defining “initial operations”—is it the device’s first ignition? Grid commissioning? Or already being able to export power? Outsiders have no way to know.
A more direct statement came from inside the company. This summer at the FusionX conference in Boston, Helion public affairs director Jackie Siebens admitted on stage: “No matter what, we cannot have 50 MW fully operational in 2028.” She also confirmed that the Microsoft contract requires Orion to be connected to the grid and begin operation by the end of 2028. But whether “connected to the grid and beginning operation” equals “supplying power to Microsoft” remains unclear. Axios quoted a Helion executive as saying full-power operation may have to wait until 2029 or even 2030.
Microsoft and Helion both declined to comment on the relevant questions.
This is not the first time Helion has adjusted its milestone wording. The earlier prototype Polaris was originally positioned as the world’s first fusion device to achieve “net electricity”—meaning electricity output exceeds the device’s own consumption. But now Helion’s website has quietly changed Polaris’s goal from “net electricity” to “demonstrating fusion power generation”—from “the electricity generated must exceed the electricity consumed” down to “as long as electricity is generated.”
A fusion energy analyst told Axios: “These are two quite different goals.”
Helion’s answers to some questions about Polaris
What is Polaris designed to do?
Polaris is designed to demonstrate electricity production from fusion and also to validate Helion’s ability to scale operations between different fuel mixes (D-D, D-T, D-He-3). This will be the first time a fusion machine has directly recovered and stored electricity from fusion-generated plasma expansion. Polaris is designed to demonstrate fusion power generation and to validate Helion’s ability to scale operations between different fuel mixes (deuterium-deuterium D-D, deuterium-tritium D-T, deuterium-helium-3 D-He-3). This will be the first time a fusion device directly recovers and stores electricity from plasma expansion generated by fusion.
This achievement will give us confidence in our ability to design and build fusion machines that can produce the fusion energy in the operating fuels required for a commercial fusion power plant. It will be an important milestone, not only for us, but for the fusion industry. This achievement will give us confidence to design and build fusion devices that can produce fusion energy under the operating fuel conditions required for a commercial fusion power plant. This will be an important milestone not only for us but for the entire fusion industry.
Is Polaris a net electricity machine?
In 2021, we used “net electricity” when discussing the goals of Polaris. In hindsight, we realized that the term “net” didn’t align with the commercial-driven goals of our 7th fusion prototype, and also wasn’t as clear as we intended. Today, we prefer to direct focus to the more concrete goal of Polaris: to demonstrate electricity from fusion, where we show we made fusion energy and converted a portion of it to electricity on the capacitor bank. This achievement will help us move toward commercial electricity production for the grid. In 2021, when discussing Polaris’s goals, we used the term “net electricity.” In hindsight, we found that the word “net” did not match the commercially oriented goals of our seventh-generation fusion prototype, nor was it as clear as we had intended. Today, we prefer to direct attention to Polaris’s more concrete goal: demonstrating fusion power generation—that is, proving that we produced fusion energy and converted part of it into electrical energy on the capacitor bank. This achievement will move us toward producing commercial electricity for the grid.
FRC Route: Pulse Fusion in an “Hourglass”
To understand these controversies, one must first understand what kind of machine Helion is actually building.
Helion has followed a non-mainstream route from the start. The company uses field-reversed configuration (FRC) and magneto-inertial fusion (MIF), a miniaturized, pulsed, non-thermal-equilibrium path under the broad category of magnetic confinement fusion. Its core idea is: first form an FRC plasma ring in a small reaction chamber, then pulse-compress it with magnetic fields, thereby reaching fusion conditions at a volume and operating timescale far smaller than those of traditional tokamaks. (See figure below.)
The company’s current core device, Polaris, is its seventh-generation prototype, about 60 feet (roughly 18 meters) long, and began operation in late 2024. Before that, Helion had iterated through six generations of prototypes, and Polaris was positioned as a “commercial viability validation device.” In early 2026, Polaris achieved two industry milestones: it became the first fusion device among private companies worldwide to achieve deuterium-tritium (DT) fusion, and it reached a plasma temperature of 150 million degrees Celsius. Helion is also the first US fusion company approved to possess and use tritium.
However, achieving deuterium-tritium fusion and achieving “net electricity” are two different things—the former only proves the device can trigger a fusion reaction, while the latter requires output electricity to exceed the device’s own consumption. This is exactly the core of outside skepticism after Helion changed Polaris’s goal from “net electricity” to “demonstrating electricity.”
Academic Skepticism
The non-mainstream route plus the unique design of pulsed compression has allowed Helion to attract capital while also facing growing skepticism from academia.
In 2026, two academic papers questioned the physical basis of this route from different angles.
• French plasma physicist Timothée Nicolas’s paper focuses on overall energy balance. His analysis argues that in the adiabatic compression process of a reactor like Helion Polaris, factors including fusion reactions, confinement time, radiation and transport losses, and magnetic energy utilization efficiency may combine to make it impossible for the device to produce enough fusion energy to offset system losses. Nicolas told Axios bluntly that some of Helion’s assumptions may be “off by several orders of magnitude,” though he added that on the question of whether all fusion companies can ultimately generate electricity, he is on the relatively pessimistic side.
• German plasma physicist Karl Lackner’s paper questioned another core assumption. Helion’s design relies on the premise that ions in the fusion fuel can remain far hotter than the surrounding electrons throughout the entire pulse cycle. But Lackner’s research points out that once actual collisions between ions and electrons are taken into account, this temperature difference may be difficult to maintain.
A venture capitalist focused on the energy sector told Axios that these two papers point to “some fairly basic issues—things you learn in your first year as a fusion graduate student.”
Helion did not stay silent. In its official FAQ, the company responded to the two scholars’ doubts, saying its design can recover electrical energy at an efficiency significantly higher than historical estimates, and that the device can complete the fusion reaction and energy recovery before the hot fuel ions cool.
But the company’s response remains at the textual level. Compared with competitors such as Commonwealth Fusion Systems (CFS), Helion discloses far less experimental data and research detail. CFS published five peer-reviewed papers in 2026 around its ARC commercial reactor, while Helion’s strategy is to keep its mouth shut. Kirtley has previously said the company is concerned about intellectual property being stolen by competitors (for example, China). (This statement comes from Axios reporting.)
Industry Expectations Shift Back Collectively
Of course, Helion’s goal adjustment is not an isolated case; the entire fusion industry’s commercialization expectations are shifting backward.
At the recent 2026 Fusion Energy Conference sub-forum “Capital Boosting Fusion Energy Development,” Zhong Wulv, chief engineer of China Fusion Energy Co., introduced the US Fusion Industry Association (FIA) “2026 Global Fusion Industry Report.” Data show that among surveyed companies, those expecting commercial power generation in 2030–2035 decreased from 21 last year to 16; those expecting 2040–2045 increased from 5 to 9; and those expecting 2045–2050 increased from 1 to 3. More companies have pushed their timelines back.
Zhong Wulv pointed out that for the magnetic confinement route to achieve commercial power generation, it still faces three major scientific problems and three engineering and technical challenges.
Against this broader backdrop, earlier this year a Financial Times report quoted an anonymous fusion company executive—when asked whether Helion could supply Microsoft by 2028, the person said bluntly that “the numbers just don’t work.” The remark quickly spread through the industry. Helion’s chief financial officer later responded to media: “We are on track.” But the slap in the face came too fast; only a few months later, the “plan” in “on track” itself had changed wording.
Helion has not canceled the 2028 node, but the definition of “initial operations” remains blank; Polaris achieved deuterium-tritium fusion and 150 million degrees Celsius, yet its goal was changed from “net electricity” to “demonstrating electricity”; two academic papers raised basic physics questions, and the company’s response remains on its FAQ page; behind the $15.5 billion valuation is the reality that it has not yet generated actual revenue.
When the most aggressive fusion company begins to redefine its milestones, and the entire industry’s commercialization timeline shifts back collectively, one question faces all observers: What exactly should we use to measure a fusion company’s real progress—one after another refreshed funding amounts and valuation numbers, or the data from device experiments that have not yet been fully disclosed? When “net electricity” is replaced by “demonstration,” and “2028 power supply” is replaced by “2028 initial operations,” is the company honestly calibrating expectations, or using looser language to maintain a story that is already difficult to deliver?
These questions currently have no answers. And there is only a little more than a year left until Orion’s promised 2028.
Reference links:
https://www.helionenergy.com/polaris
https://www.axios.com/pro/climate-deals/2026/09/08/helion-energy-fusion-net-electricity-date-year
https://www.powermag.com/helion-announces-465-million-funding-round-to-support-fusion-energy/
Recommended reading:
Helion Is Selecting a Site to Build Its First Commercial Nuclear Fusion Power Plant
Helion Energy: Leading a New Era of Zero-Carbon Fusion Energy
Powering Microsoft in 2028, Helion Energy Receives Conditional Use Permit from Local Government