Fusion ⚛️ Sexy energy just got serious
Clean, limitless and finally within reach
At The Economist’s Fusion Fest in London this week, one thing was clear: the race is on. Across the world, over 45 startups and public projects are no longer talking theory. They’re building machines, testing fuel, and chasing the biggest energy breakthrough in history.
Because we’re going to need a lot more power. Not just to hit climate targets, but to lift billions out of poverty and feed AI’s exploding energy appetite. At the Energy Transition Summit last month, the message was blunt: old-school nuclear is too slow, too expensive, and no longer trusted. Renewables are just not enough.
Fusion offers a clean slate. No carbon. No meltdown risk. No thousand-year waste. Just raw power from the same reaction that lights the sun.
Put simply, fusion smashes hydrogen atoms together at extreme heat and pressure. If you get it right, you release more energy than you use. That’s the goal, and after decades of false starts, we’re finally edging over the line.
The aim isn’t just to build hundreds of brand new plants. It’s also to plug into what we’ve got, just like we did going from coal to gas. And eventually, fusion could get small. Think mobile reactors powering remote villages or entire factories.
But the real blocker? Investment. Governments still matter, but private capital is moving faster. And we need both: money, approvals, and the systems to connect fusion to the grid.
This special dispatch is your map of fusion at this moment: the players, the progress, the politics. Because fusion isn’t decades away anymore. It’s closer to ten. And when it lands, it won’t replace renewables. It’ll fuel them.
The next phase of the energy transition just stepped up. Let’s get to it. 👇🏻
This deep dive is 7,500+ words, so you can use the table of contents on the left to jump between topics (if you’re reading online), and if you download the Substack app, you can listen to a narrated version of this feature.
Millennial Masters is brought to you by Jolt ⚡️ The UK’s top web hosting service
1. Ignition gets real
Upgrade the grind, not just the lab
In 2022, the fusion world changed, but only slightly. Lawrence Livermore National Laboratory (LLNL) achieved ignition: a fusion reaction that produced more energy than it consumed. “The laser was built to produce about 1.8 million joules of laser energy. Over time, we’ve been able to raise that energy to over 2.2 million joules,” said Kimberly Budil, director of LLNL. “We’re trying to work our way back down… to get to the highest possible yield with the lowest possible laser energy.”
But that breakthrough wasn’t a repeatable blueprint. The real grind is scaling it, and upgrading outdated tech while doing it. “We brought the National Ignition Facility online in 2009. For a high-tech facility, that’s roughly 1,000 years ago in science years,” Budil said. “Many of the systems in that facility are somewhere between completely obsolete to run-to-failure.”
Support for refurbishing is coming, but performance is already dipping. “We normally do about 400 experiments… that’s been steadily coming down to maybe 350,” she added. And that brings its own scientific rewards. “We’ve done a lot of materials testing… we can do diffraction dynamically at these incredibly high pressures and produce very high-fidelity data on how materials change on billionth-of-a-second time scales,” she explained.
Government turbulence hasn’t helped. “It’s a very complicated time in the government in the United States right now,” Budil said. But she sees upside. “We’ve already worked with the Secretary of Energy to implement some important changes… that are really going to net benefits for the laboratories.”
The biggest collective challenge? Yield. “If we brought several companies together and pooled resources… we could really focus on that pathway to higher gain.”
And while other nations explore hybrid fusion, Budil’s focus remains sharp. “We’re firmly committed to sticking to this fusion path… it has many, many benefits beyond going toward fission-fusion hybrids.”
Upgrades are already underway. “We’re currently working… to install the rest of the amplifier slots, which we think will allow us to take the laser energy to at least 2.6 megajoules,” she said. “That will give us, for the next decade, just an incredibly capable machine.”
The science is moving. Slowly, but deliberately. The next breakthrough won’t be in headlines, it’ll be in repeated, reliable output. And the real race is who can get there first.
From running joke to power race
Fusion’s long been the punchline: always 30 years away. But Bob Mumgaard, CEO of Commonwealth Fusion Systems (CFS), says it’s time to kill the joke and build the future.
“We’re building a facility outside of Boston (MA) called the SPARC facility,” he said. “It’s based on tokamak magnetic confinement, and that’s just over 60% complete.” The goal: net energy gain of around 10. Power-up starts in 2026. Ramp-up in 2027. And the first commercial plant? Already lined up for Virginia.
Magnets are the magic, he explained. “We’ve doubled the magnetic field strength over what was previously available.” They’re using 20,000 kilometres of superconducting tape, enough to stretch from pole to pole, scaled up 40x in just a few years. Most of it’s already installed.
But one plant doesn’t cut it. “If you only build one, you haven’t really built a field,” Mumgaard said. “You need thousands.” That means supply chains, regulation, talent; not just physics, but infrastructure.
CFS isn’t building tech for others to license. It’s going full-stack. “We do the whole thing end-to-end,” Mumgaard said. “In the future we’d love to hand off pieces, but right now, there’s no one mature enough to take them.”
What does fusion most resemble? Not fission. Not fantasy. “It looks more and more like a gas turbine, and less and less like a pile of hot rocks,” he said. “You need systems engineers who can make trade-offs across the whole chain.”
So why believe the 30-year curse is over? “There’s this fallacy: if something’s been a long way off, it always will be,” Mumgaard said. “But that hides the massive progress we’ve made.”
And now, progress is compounding. Real ignition. Real timelines. Real facilities. Mumgaard says the pattern now looks less like a slow space shot, and more like early biotech.
By 2029, SPARC is expected to deliver 15 megajoules of output. Others might catch up: one or two, maybe. But for now, CFS is leading the breakaway.
Can fusion hit the grid in time?
Fusion is full of promise. Yet until it delivers actual power to actual grids, it’s just that, a promise. But the shift from lab to large-scale deployment is an infrastructure challenge, a regulatory challenge, and a people problem.
Shay Bahramirad, vice-president of electric transmission at Pacific Gas and Electric Company, said it bluntly: “We need to work together and collaborate, to take the technology from the laboratory, go through the standards and socialisations, and make sure that it gets to the market in a timely fashion.”
She pointed to a major shift in grid demand: “We always have been seeing a steady state consumption, and the growth wasn’t there. But now with AI and data centres, it’s a great opportunity for different types of technologies.” Fusion, she argued, could be co-located with high-energy users like data centres, if the standards and infrastructure catch up fast enough.
Ryan Ramsey, chief operating officer of First Light Fusion, said three things will determine fusion’s success: regulation, infrastructure, and people. “Have we the right policies to make this happen? Do we have the infrastructure to support fusion power plants? And are we thinking far enough ahead to make sure we’ve got all those skills in place?”
On timelines, he sees the early 2030s as the moment pilot plants begin feeding into the grid. “We don’t think we see commercial power plants until 2040s and 2050s,” he said. “By 2050, you’re looking at lots of commercial power plants worldwide.”
Sehila Gonzalez de Vicente, global director of fusion energy at the Clean Air Task Force, argued that demand will push fusion forward, whether we’re ready or not. “Energy demand is growing and growing,” she said. “Fusion is a way to fulfil the demands of the immediate future and the next future.”
The need is clearest in countries outside the OECD, she said, where rising living standards are creating entirely new layers of energy demand. Fusion, if commercialised, could ease that pressure while helping to phase out fossil fuels.
Stuart Codling, group director for global fusion power at Amentum, said the key obstacle is coordination and the structures around it. “We need the right commercial contracts,” he said. “We need frameworks for different organisations to genuinely collaborate, and to facilitate change as we move through the engineering programme.”
Education and skills were major themes, too. With an entire industry to build, everyone agreed the pipeline needs work. Ramsey pointed to the UK’s Fusion Skills Council as one step forward. But as Gonzalez de Vicente put it, it takes more than engineers. “Fusion is becoming an industry. We need lawyers, economists, all sorts of skills. Not just physicists and engineers.”
The message was clear: the technology is close. But turning fusion into a cornerstone of our energy mix will take much more than science. It will take systems, funding, policies, and people — starting now.
2. Barriers still rising
Tech, trust, and trillions needed
Fusion’s promise of limitless, clean energy feels closer than ever. But despite recent breakthroughs, achieving ignition (where fusion reactions become self-sustaining) is just the beginning. For fusion to transform from scientific marvel into commercial powerhouse, significant hurdles remain in supply chains, geopolitics, funding, and technology itself.
Brandon Sorbom, co-founder of Commonwealth Fusion Systems, says that comparing progress across fusion approaches comes down to a fundamental benchmark called the triple product: density, temperature, and confinement time. While it’s a common language to compare techniques, Sorbom said: “You really have to educate people about what’s going on… a big part of that is actually publishing your results and being very transparent.”
Transparency is vital, but equally essential is the infrastructure that turns fusion concepts into reality. Chris Martin, Chairman of Tokamak Energy, warned against isolation, stressing that collaboration and data-sharing are key: “We only will do that by running down this enormous amount of learning across the supply chain.”
The technical challenges are immense. Kimberly Budil, Director of Lawrence Livermore National Laboratory, home to the National Ignition Facility (NIF), noted key technological gaps still exist, especially around “materials that can survive in high radiation environments over long times” and “a scheme for a closed loop fuel cycle.”
Yet the path to commercial fusion isn’t purely technical. Chris Mowry, Chief Executive of Type One Energy, said fusion companies need to integrate directly into existing energy markets. “Do we really need to reinvent how to build a power plant? I don’t think so,” he argued. Instead, Mowry called for industry-wide thinking about “ignition from a market adoption perspective,” shifting the industry from “pushing the ball down the field” to being “pulled into the end market.”
Geopolitical considerations are increasingly shaping fusion’s landscape. Francesco Sciortino, CEO of Proxima Fusion, points out the reality of geopolitics: Europe’s energy policy failures, especially in Germany, highlight why the old continent needs its own robust fusion supply chains: “Germany and Europe really need to get their act together.”
Money is perhaps the most pressing challenge. Asked about funding needs for the next decade, Kimberly Budil suggests a targeted infusion: “about ten billion” from public programmes to tackle immediate technical barriers. Chris Martin frames the choice starkly: “Do we want two more fighter jet programmes, or do we want a clean, renewable, vast source of energy for the world?” Francesco Sciortino estimates the total investment needed might reach around “50 billion,” encompassing technology demonstrators, net-energy devices, and first-of-a-kind plants.
But will this investment actually produce viable fusion power plants before flagship international projects like ITER? Brandon Sorbom is optimistic: “I personally think there’s going to be a few concepts that make it to ignition.” Chris Mowry agreed, “If you qualified by saying ignition [as] putting electrons on the grid, I would say maybe two or three.”
Fusion may still be a field defined by optimism, but the growing consensus is the science is nearly there — now the industry must show it can translate ignition into reality.
3. Startups shift the game
Tiny reactors make a big play
Fusion reactors usually mean vast facilities and billion-dollar bets, but Avalanche Energy is pursuing the opposite. “We decided we need to do something small,” said Brian Riordan, co-founder and COO. Their current prototype measures just 12 centimetres across.
Small is strategic. Rapid iteration, learning quickly through practical tests, is core to Avalanche’s philosophy. Riordan explained, “First one, test it, learn from it, know how to build it. How did it fail? Where did your assumptions and your simulations go wrong? Iterate on it.” Recently, Avalanche held 300,000 volts in their compact device for five hours, precisely what’s required for effective fusion reactions.
Compact reactors have immediate uses beyond power generation, particularly in defence and space. “High density in a mobile package is very interesting to…the capability-driven defence industry,” Riordan noted, highlighting neutron imaging and remote sensing applications.
Yet scepticism persists. “Sometimes I’ll talk to people, and I feel like a rock star tackling this big problem. Sometimes I feel like a snake oil salesman,” Riordan admitted. But visiting Avalanche’s lab changes minds: “When you see 50 people that are building these things…you get a little bit more excited and say, ‘Alright, maybe I’m going to dig in a little deeper.’”
Lab to grid with new tech bets
General Fusion founder Michel Laberge is betting on a novel fusion design that mixes magnetic and inertial confinement. Their new demonstration machine, LM26, recently began operations in Canada. “We got our first plasma about a month ago,” Laberge said. The next stage: compressing plasma to achieve 10 million degrees Celsius this summer, then scaling to 100 million degrees next year, the crucial temperature for practical fusion.
General Fusion’s unique approach involves compressing plasma using pistons pushing liquid lithium. It avoids key engineering hurdles facing other fusion methods. “We don’t have superconducting coils, lasers, neutral beams, RF heating; none of that,” Laberge explained. “It’s much cheaper.”
Critically, the system also solves neutron-induced material damage, a major barrier for fusion reactors. The liquid lithium wall absorbs neutrons, removing the need for new radiation-resistant materials. “We fix that problem today,” Laberge says. “We don’t have to wait.”
Smaller, simpler, and faster to build, Laberge argues his method is well-suited for rapid commercialisation. General Fusion plans to demonstrate net-energy fusion with LM26 by late 2026, followed by an engineering demonstration plant around 2030 in the UK. The commercial plant, a compact, 150-megawatt reactor producing steady power, could be operational by 2035.
Challenges remain, from ensuring plasma stability to managing corrosion from liquid lithium. But Laberge is confident: “Everybody wants to make a tokamak. It’s not a good idea. We have a good idea.”
General Fusion’s technology might finally move fusion from experimental machines to practical, grid-ready power stations.
Fusion’s real moonshot
Fusion is often framed as the ultimate moonshot, but to make it real, it may need less NASA and more SpaceX. A new generation of engineers and founders say the real opportunity is in getting fusion working commercially, fast.
“Fusion is a very difficult problem to commercialise,” said Greg Piefer, founder and CEO of SHINE Technologies. “We should practice… like sell fusion today.” His company is building what he calls a “Gigafactory for isotopes,” a $500 million facility that uses fusion neutrons to produce medical isotopes. “It’s safer, cheaper, cleaner,” he said. “As we get practice at doing fusion, we’re helping tens of millions of humans per year.”
Richard Pearson, co-founder and chief innovator at Kyoto Fusioneering, echoed the urgency. “If you think about fusion as the star in the jar, we’re working on the jar,” he said. “We’re working on the systems that fuel the stars, take energy away from the star, recycle and so on.” His analogy extended to the wider ecosystem: “We’re working on fuel injectors, maybe the turbocharger… someone else will design the tyres, someone will make the chassis.” The goal isn’t just moonshots, it’s sustainable industry.
Bianca Cefalo, CEO and co-founder of Space DOTS, argued that fusion needs to think like a modern space startup: lean, fast, and hyper-customer focused. “I need 3,000 users yesterday, not in 10 years,” she said. Cefalo’s team includes ex-Formula One and nuclear engineers brought in for their ability to move at speed. “From paper to integration in three months, the Formula One engineer said, ‘I know how to do that.’”
For Cefalo, urgency is existential. “Someone smarter with more money than you tomorrow can do exactly what you thought you were going to do, better than anyone,” she said. “If you’re not commercial, you’re not going to find investors. You’ll lose your team, your investors, and eventually your business.”
Francesca Faedi, academic liaison for commercial and innovation at Space Park Leicester, called for stronger integration between researchers and operators. “You need that open, wide mindset where you have the technical knowledge and also the market and business knowledge together,” she said. “We want the same thing… but we approach it from completely different points of view.”
For fusion to cross the threshold from science to scale, the lesson is clear: it’s not about the moonshot. It’s about building the rocket company that gets you there — and doing it in months, not decades.
4. Tools to break through
AI hunts answers faster than teams
AI isn’t just writing emails and drawing cats in space, it’s getting ready to help crack one of the hardest scientific puzzles on Earth: fusion.
“We’re really excited about the possibilities for artificial intelligence to accelerate fusion to the grid,” said Kenji Takeda, director of research incubations at Microsoft Research. From modelling plasma dynamics to discovering new materials, AI is moving in on problems that have hobbled human minds for decades.
This isn’t abstract. Microsoft is already working with fusion giants like ITER. “They’re using Azure OpenAI technology… they’ve trained it on a million IT support tickets,” said Takeda. “They’ve built a chatbot that actually helps the entire organisation.” The same AI is helping ITER’s engineers write safer software, and that’s just the starting point.
The real leap comes from AI’s ability to work at the language level of nature: electrons, atoms, molecules, fields. “We’ve built AI models that can generate materials that satisfy the properties you need,” said Takeda.
But what does that mean for fusion? Shruti Rajurkar, senior research technical programme manager at Microsoft Research, laid it out: “These materials in our fusion reactor are subjected to extreme conditions… high energy particle bombardment… defects, transportation and a lot of property changes.”
Traditionally, scientists try to search for materials that can survive this. “What AI can do is put it as an inverse design problem,” she said. “If I have certain requirements, can you generate the material for me that satisfies these?”
She compared it to text-to-image generators, but here, the “text” is a list of material properties, and the “image” is a working, testable atomic structure. Then there’s simulation. “It’s very expensive to do these calculations,” said Rajurkar, referencing the problem of scaling numerical plasma models. Microsoft is working on physics-informed neural networks and world models that learn how systems behave, without being explicitly told the physics.
These world models are already learning the dynamics of games like Minecraft. The parallels are uncanny. In games, the goal is not falling off a cliff. In fusion, it’s sustaining a plasma pulse. “These technologies… can transfer to an area like fusion,” she said.
And that’s the point. AI won’t solve fusion on its own, but it’s already proving it can save time, surface hidden insights, and tackle the brute-force maths slowing progress.
Spinouts making billions already
Fusion startups are racing to build powerful superconducting magnets, but their innovations may deliver big wins far beyond energy generation.
Itxaso Ariza, Chief Technology Officer at Tokamak Energy, explained how her company’s advanced high-temperature superconducting (HTS) magnets are also opening doors in markets from aerospace to magnetic levitation. By cleverly managing imperfections in superconducting materials, Tokamak Energy has developed “compact, reliable magnets,” adaptable to real-world conditions.
At Renaissance Fusion, CEO Francesco Volpe is taking a completely different route, creating superconducting magnets by laser-engraving large cylinders rather than winding coils. This radically simplified approach makes the technology more versatile and affordable. According to Volpe, the opportunities extend far beyond fusion into medical imaging, renewable energy storage, and even futuristic robotic applications thanks to their precise control over liquid metal components.
Jonathan Toretta, Chief Executive of TAE Global, shared a business-driven perspective. TAE initially pursued spin-offs “by accident,” he admitted, but the strategy paid off. Their expertise in fusion-related power systems led to successful ventures in advanced power management and even cancer therapy. These spin-outs, he noted, have become profitable businesses in their own right, supporting TAE’s primary fusion mission.
Chris Good, Managing Partner at Pine Island New Energy Partners, made a compelling investor’s case for fusion spin-offs. He argued that while fusion itself may be years away from profitability, “fusion is already driving massive demand in supporting technologies,” from superconducting tapes to lasers and cryogenics. For Good, these near-term commercial applications present major investment opportunities that don’t depend solely on fusion’s success.
Antonio Pellecchia from ASG Superconductors explained that superconducting technology is already a billion-dollar market: think MRI machines and proton therapy. New superconducting innovations could transform power transmission, ships, and transport infrastructure. “Superconductivity today,” Pellecchia highlighted, “is already big business.”
Fusion magnets started as tools to hold plasma, but they’re quickly becoming engines for entirely new industries. Investors and entrepreneurs alike are realising these spin-offs might deliver serious profits well before fusion hits the grid.
When defence meets deep science
National security hinges on technology that works, especially when things get extreme. Whether it’s nuclear submarines or satellites, surviving intense radiation is non-negotiable. But ageing test facilities aren’t keeping up.
JC Btaiche, CEO of Fuse, sees a direct link between defence testing and fusion’s future. Fuse has built the world’s first impedance-matched Marx generator, a device delivering ultra-fast radiation pulses. “We recently fired it at full scale, one terawatt at nanosecond rise time,” Btaiche said. Fuse’s approach could modernise radiation testing, supporting critical US defence projects and commercial fusion.
Admiral Charles A. Richard, former US STRATCOM commander, insisted rigorous radiation testing is essential. “Given the threats we face, every bit of our defence capability has to have confidence,” he stressed. Missile defence, satellite resilience, and power grids all rely on systems proven under extreme conditions. “We just haven’t had to think about this in a long time,” Richard admitted.
Fuse isn’t waiting. It’s already under contract with the US government and global corporations, providing commercial radiation testing. The aim? Accelerate fusion while ensuring strategic systems keep working. Richard, now a Fuse advisor, sees the parallel clearly: “Plenty of paper fusion plants exist. Turning them into reality takes real-world engineering.”
5. The new power race
Fusion reshapes global strategy
Fusion has moved from the margins of science to the centre of geopolitics. What used to be a laboratory challenge is now a race with major consequences for alliances, economies and energy security.
At the heart of this shift is ITER, a massive experimental fusion reactor under construction in southern France. It’s backed by 35 countries, including China, the US, India, Japan, Russia and the EU. The goal is to prove that fusion power is not just scientifically possible, but commercially viable.
Laban Coblentz, Director of Communication at ITER, explained the scale: “We’re the big dog in the room.” He described it as one of the only projects where countries that “normally are just fighting” — including the US, Russia and China — are working together.
But this international collaboration hasn’t stopped the race. Tone Langengen from the Tony Blair Institute pointed out that China is already going full speed. “They’re really doubling down on expanding their efforts,” she said. “This could become another case like solar or EVs, where one country runs ahead and ends up owning the whole supply chain.”
Europe, despite hosting ITER and having the scientific firepower, is falling behind. Milena Roveda, Chair of the European Fusion Association, didn’t hold back: “Europe doesn’t have a place at the table. If you don’t sit at the table, you’re on the menu.” She called for more than reports and committees: Europe needs to build.
Her proposal: a coalition of countries to develop fusion power the way they once built fighter jets together. “Not the whole European Commission, that’s too slow. But four or five countries coming together with a plan? That could work.”
Langengen added that deep-pocketed public finance is what separates China from the rest. “That’s the biggest problem we have,” she said. “We need to innovate how we fund this.” Fusion won’t be cheap, but the real cost is in delay.
Behind the scenes, artificial intelligence is starting to accelerate progress. At ITER, Coblentz said they’re using custom tools powered by Microsoft to pull insights from decades of research and build digital twins of the reactor systems. “We call her Lucy,” he said. “She’s faster than our oldest or youngest engineers at retrieving information.”
Fusion isn’t just a clean energy bet. It’s a driver of new technologies, a test of industrial strategy, and a marker of global influence. Countries that get there first won’t just power their grids — they’ll shape the future energy system the rest of the world depends on.
Britain makes its £400m move
Ian Chapman opened with a deadpan confession. “I told my daughter I was speaking here. She said, ‘Dad, everybody is so bored of listening to you.’” It’s the kind of comment only your kid could deliver — and it captured the mood perfectly: serious science, yes, but time for some straight talk.
Chapman is chief executive of the UK Atomic Energy Authority, currently the largest fusion organisation in the world with nearly 3,000 people. But he made it clear — success would mean that’s no longer true. “It means the industry is going somewhere.”
He set the stakes immediately: “2024, the hottest year on record. It is a crisis.” Fusion, he argued, isn’t some long-term fantasy; it’s central to solving today’s energy dilemma. Countries are caught in a bind: decarbonise fast or protect economic stability. “You don’t have to choose between doing good and doing well,” Chapman said, quoting Apple CEO Tim Cook. “It’s a false choice today, more than ever.”
Britain is trying to lead on both fronts. It’s building a fusion prototype plant (STEP) in West Burton, Yorkshire, and has just announced a slew of partnerships to keep momentum up. These include a tritium facility with Italian energy giant Eni, an AI growth zone at a fusion site, and a government-backed venture fund for fusion-scaleups called StarMaker One. “All in the last eight weeks,” Chapman said. “All very symbiotic.”
He laid out the UK’s industrial strategy like a pyramid: a top-level “integrator” building the plant (that’s Industrial Fusion Solutions, a government-backed company); followed by a tier of system providers: magnets, robotics, heat exchangers; beneath that, startups and SMEs building enabling technologies; and at the base, a foundation of research, skills, and innovation.
But he was blunt about where things often fail. “Technology — absolutely no good to anybody. You have to think about how you’re actually taking that to market. No good saying, ‘I’ve got technology now, but no supply chain and no people.’”
Chapman warned that fusion risks the same fate as fission: viable designs nobody will build. “Almost the worst thing you can do as a tech company is develop a product that nobody wants.” In fusion, he said, too many people are focused on building something that works, without asking if anyone would buy it. “There are lots of fission plants that would work but have never been built.”
To avoid that trap, he said the industry must prove both long-term viability and short-term commercial value. “You must be revenue generating. You must actually generate some returns to your investors.” Areas like magnets, robotics, computing and simulation already offer adjacent commercial revenue. “Not in 20 years, today.”
In a personal moment, Chapman reflected on his own transition, leaving fusion after a “half-life” to run UK Research and Innovation. “It’s quite a hard thing for me to do. I’ve committed a lot of my life to fusion.” He compared the effort to a football comeback. “My half-life in fusion tells me that most of the time it feels like being three-nil down at half-time.” But, as he reminded the audience, those matches are sometimes won.
His final message? Get in the game. “This is a crisis,” he said. “Do good and do well. Focus on how you’re going to make money out of it. Commercialise it.” Then he quoted (now Canada’s PM) Mark Carney: “Those who are part of the solution will be rewarded. Those who are part of the problem will be punished.” He urged: “Make your choice, folks.”
Old nukes still hold key lessons
Author and astronautical engineer Robert Zubrin made a full-throated case for nuclear energy, not just as a clean energy source, but as humanity’s best bet for lifting billions out of poverty.
Zubrin didn’t mince his words: “If we do want to relieve poverty, we’re going to have to increase world energy consumption five times over… and it’s going to take nuclear energy to do it, both fission and fusion.” The problem, as he sees it, isn’t just climate. It’s scarcity. And the only way to truly tackle both is by scaling abundant energy.
His core argument is that energy isn’t just about technology. It’s civilisation. “There’s a direct correlation between per capita energy consumption and living standards,” he explained, showing how global energy use has doubled every 30 years, even after global warming became a political priority. “People don’t want to be poor,” he said. “Any programme that increases the cost of energy is denying a decent life to the world’s poor.”
Zubrin took aim at the phrase natural resources. In his view, it’s a myth. “There’s no such thing as a natural resource,” he argued. “There are only natural raw materials. It is human creativity that transforms materials into resources.” Oil, uranium, and deuterium were once worthless until the right technologies emerged. Fusion, he says, is about doing that again, with water, with rock, and with the atom itself.
He warned fusion advocates to learn from fission’s history. “It’s been hyper-regulated by people who are hostile to the technology,” he said. “And this is something we absolutely have to watch for in fusion.” Public fear, bad regulation, and bureaucratic delays, he argued, have crippled what should be the world’s leading energy platform. He urged fusion developers to stay ahead of the framing before safety debates are hijacked.
Zubrin also sees fusion as essential for space exploration. Martian water contains far more deuterium than Earth’s, and fusion could eventually power rockets at 7% the speed of light. “It’s not just another way to make electricity. It’s a new kind of energy that can do new kinds of things.”
But his biggest warning wasn’t technical. It was philosophical. He pointed to a dangerous idea: the belief that there isn’t enough to go around. “That was the idea behind the catastrophes of the 20th century,” he said. “This notion that there’s only so much, and if they get more, we get less.”
Fusion, he argued, is the ultimate rebuttal. “If humans are creators of resources, then the human race is engaged in a joint project to expand the availability of everything for everyone. And fusion is how we’re going to prove that.” Mic dropped.
6. Story makes the science
The chemical chain reaction
The chemical industry runs on heat. Lots of it. Today, that heat mostly comes from fossil fuels. But if fusion delivers what it promises, a future of abundant, affordable electricity — everything changes.
Bjorn Theis, head of foresight at Evonik (German chemicals company), has one job: look 10 to 20 years ahead and spot the next big disruption. He’s convinced fusion is it. “Electricity could become the most important resource in the world,” he said.
The challenge is scale. “If we really want to electrify all our processes, we’d need the same amount of electricity that Germany consumes today — just for the chemical industry,” Theis said.
Fusion could meet that demand. It could also revive older technologies like electrochemical manufacturing, which once lost out to fossil-fuel heating. From power-to-heat steam crackers to power-to-chemicals and even power-to-food, Theis sees the makings of a new industrial era.
He pointed to desalination, air-to-water tech, CO₂ conversion, and synthetic biology, all of which demand serious energy. “We don’t want to say ‘cheap’ electricity,” he said, “but affordable electricity would change the game.”
Fusion isn’t just about energy. It’s a foundation for solving water shortages, scaling sustainable food, and driving the next wave of computing, AI, and manufacturing. As Theis put it, “We need a platform technology to pay the electricity price for the future we want.”
Startups join forces to scale fast
If fusion is going to power the future, it won’t get there alone. Across the energy sector, the call for collaboration is growing louder, and in the fusion world, that means rewiring how public and private players work together.
“As a startup or as a company, focus is one of your core aspects,” said Heike Freund, chief operating officer at Marvel Fusion. That’s why her team has partnered with Siemens Energy and Colorado State University to share the heavy lift of building fusion test facilities. “Finding partners to work on topics that are not in your core focus area is super important.”
Rather than building standalone machines that risk becoming orphans of innovation, Marvel is embedding its tech in university facilities, a move that keeps students and researchers engaged long after version one has launched.
That kind of thinking is key, said Erik Fernandez of Fusion Europe. The old tech-transfer model (lab-to-industry handoffs) no longer fits the pace or scale of fusion innovation. Today’s ecosystem demands real co-development, with supply chains and industry stakeholders integrated early. “We have between seven and nine billion in the market from private money,” he said. “This is a real game changing.”
Fusion Europe was set up to be a common platform, a place where early-stage companies, scientists, and corporates can align. “Now the association is being established… but establishing an association is not enough. Now we need to do real useful activities.”
That means getting deep into the tech. Vittorio Badalassi, distinguished R&D staff at Oak Ridge National Laboratory, focused on fusion blankets, a critical but unproven technology. “Nobody has ever built a blanket yet,” he said. “There are a lot of concepts… but nobody knows how good they are, how well they are working.” His team is working on high-fidelity digital twin simulations, aiming to avoid the costly trap of real-world trial and error. They’ve also patented a new blanket design that could bypass the need for lithium-6 enrichment, a material currently cornered by China.
At CERN, the world’s leading particle physics lab, the shift towards fusion is already underway. “We are very resilient to the process of taking low TRL objects from crazy ideas to something that works,” said Luca Bottura, who leads CERN’s fusion tech coordination. The same high-temperature superconducting magnets built for particle accelerators are now being tested for use in fusion.
But it’s not just about tech, it’s also about how teams function. Freund highlighted the importance of integrated collaboration. In traditional customer-supplier setups, requirements are static. In fusion, change is the only constant. “Being very open and transparent on both sides, that things change over time… that they need to be integrated.”
And while the “valley of death” between lab prototypes and commercial products still looms, fusion seems to be breaking the mould. “There is this novelty where the private sector seems keen to invest also on low TRL technologies,” said Badalassi. “I haven’t seen this before.”
Whether that leads to a Manhattan Project-style coordination remains to be seen. But for now, the diversity of approaches is fuelling rapid innovation. “Currently, it’s good for fusion we have different approaches,” said Freund. “We are really pushing innovation in the field.”
The energy transition needs answers fast. These kinds of collaborations might be messy, but they’re exactly what fusion needs.
Busting myths, winning hearts
Outside of the hype bubble, fusion isn’t a silver bullet, and it’s not “just like fission but better.” It’s a fundamentally different energy source with its own risks, realities, and potential. The challenge? Making sure the public gets that.
“Fusion, it is cleaner, it is greener, but it’s not a silver bullet,” said Laura Berzak Hopkins, deputy chief research officer at Princeton Plasma Physics Laboratory. “We need to be very careful that we’re not greenwashing fusion.”
That starts with ditching jargon and talking like real people. Tristram Denton, UK director at the Fusion Industry Association, put it bluntly: “Outside the room, nobody cares.” The obsession over tokamaks versus stellarators or magnetic versus inertial confinement? “We can go right into the amazing discussion… but we need to talk about clean energy, because that’s what we’re here for.”
Fusion may sound like science fiction, but it’s already part of our reality. And it’s evolving fast. As Jin Zhang, assistant professor in microwave electronics at Queen Mary University of London, explained: “There will be some radioactive waste due to the bombardment of the neutron from the infrastructure… but we don’t have a meltdown that happens in Fukushima. Even if something goes wrong, we don’t have that bad consequence.”
Still, public trust will make or break fusion. And the lessons from nuclear fission are clear: overpromise, underdeliver, and communities will turn their backs. “Spending 70 years being later and more expensive than you thought you were going to be hasn’t engendered trust,” said Denton. “It’s about informed consent.”
That means focusing on real-world concerns. “They want to talk about lorry movements… job creation, probably above all else… the quality of the jobs,” said Denton, reflecting on his experience walking communities through site selection. “You’re a clean project, which means people are okay with you. You’re a safe project, which means communities are happy to host you. And then after that, you’re a project.”
Commercial viability is no small hurdle. But Hopkins was clear: “It would be a huge failing if we get the science and we get the technology ready and we don’t have the regulatory infrastructure, and we don’t have those market forces sorted out.”
Zhang’s team is already working on solutions to bring costs down: “We’re developing a completely new version of something called magnetron… If we can reduce the cost to 10% of what’s currently available, we can pave the way to commercialisation.”
As for the classic myth that fusion is “always 30 years away”? Hopkins gave it the context it deserves: “When I started in the field, fusion was 50 years away… If we continue this level of investment and focus now, then I think it is a myth. But it’s also within our power to make that very true.” And Denton had a personal pledge: “You’ll never hear me again say 30 years.”
Smart bets will decide the race
The path to fusion won’t be won with optimism or effort, it’ll take systems thinking, technical realism, and strategic funding bets. That was the message from Ross Koningstein, Director Emeritus at Google and founder of its advanced nuclear energy R&D group, speaking at Fusion Fest.
His team (known internally as NERD) kicked off over a decade ago to explore what was then a long-shot sector. Today, the story’s changed. “Fusion is not just likely, but going to happen,” said Koningstein. “We see things happening that are taking us towards these points where we can say, yes, fusion is on our near horizon.”
But he was clear: the journey from lab to grid doesn’t hinge on breakthroughs alone. It’s about how you build the whole ecosystem, and resist the instinct to back a single horse.
“We don’t want to pick a winner. That was not the success pattern of the last iteration,” he said. “We want to see a horse race… not a tonne of horses, but a few; so you have investment going into a number of different approaches, and at the same time, you get the discipline of private capital.”
Koningstein recounted an early meeting with then–US Energy Secretary Ernie Moniz where Google’s map of advanced nuclear companies caught the Department of Energy off guard. Moniz asked which one should win. “And we were like, well… we don’t want to pick a winner,” Koningstein said.
Instead, he sees the role of policy as enabling multiple shots on goal. That means a mix of public funding, private capital, and enough flexibility for startups to move fast.
“In a company, you’re always faced with the urgent,” he said. “But there’s always this tension between the urgent and the important… and at some critical juncture, the important should either become your or somebody else’s urgent.”
It’s a lesson that’s shaped Google’s own fusion contributions. His team developed Bayesian models and machine learning tools to help fusion startups analyse plasma behaviour and improve machine design. Beyond his own group’s work, DeepMind has applied AI to control tokamak reactions and optimise plasma shapes, contributions that are already shaping the commercial toolkit.
But Koningstein returned to the bigger picture: government strategy, investor discipline, open communication, and cross-sector input. “There needs to be a fair amount of government and private capital,” he said. No grandstanding. No silver bullets. Just a push for better questions, clearer bets, and an ecosystem wide enough to win.
Time to stop doubting, start building
Fusion energy faces waves of scepticism, yet turning doubt into excitement is precisely Philippe Larochelle’s mission. As Partner at Breakthrough Energy Ventures, backed by Bill Gates, Larochelle sees fusion not as a distant dream but as an immediate opportunity to unlock vast, sustainable energy.
Larochelle directly challenged the common doubts thrown at fusion, starting with the most basic: impossibility. “Fusion is not impossible,” he said bluntly. “99% of all the practical energy you’ve ever used comes from fusion.” Fusion powers the stars, and plasma (the superheated fuel of fusion) is surprisingly accessible. To prove the point, he held up a handheld plasma device: “Plasmas really aren’t that complicated,” he joked, explaining that scaling up known technologies is the core challenge, not breaking new scientific ground.
But why fusion over other energy sources? Larochelle pointed to fusion’s extraordinary energy density. Transitioning from chemical fuels to fusion would unlock “a factor of a million” increase in energy density, like “trying to imagine all the good ways the world would change at the dawn of the steam engine.”
Economically, Larochelle argued that fusion is potentially cheaper than many assume. “Fuel is infinite and practically free,” he said. Early techno-economic assessments suggest fusion could eventually produce electricity at under five cents per kilowatt-hour — competitive with almost any energy source. He added that fusion plants might even require fewer raw materials per megawatt generated than solar, wind, or conventional nuclear power.
Finally, he addressed the looming competition with China, highlighting the vast sums China is investing in fusion. Larochelle believes Western countries don’t need to outspend China but must invest strategically, following models like NASA’s partnership with SpaceX, where private innovation was amplified by focused government backing.
The scientific hurdles have largely been overcome, he concluded. The industry now faces engineering, manufacturing, and regulatory challenges, but these are exactly the challenges best solved by smart investment and scaling.
“Perhaps we’re at the end of the beginning,” Larochelle ended, echoing Churchill. Fusion is no longer just science, it’s now firmly entering its commercial phase.
7. A test of delivery, not just discovery
The science is here. Now comes the hard part.
Fusion machines are firing across the globe. Magnets are stronger. Fuel cycles are closing. Even the sceptics are softening. But scaling it? That’s the real test and the next great engineering challenge of our time.
This isn’t just physics anymore. It’s a full-stack systems problem. The breakthroughs are coming from materials labs, supply chains, AI models, and regulatory war rooms. Fusion is becoming a proving ground for how we actually build the future, and whether we can do it fast enough.
What it needs now is pace, coordination, and cash. Not spray-and-pray investment, but smart capital. Strategic bets. Deployment plans that start now, not in 2040. The tech is maturing. But the grid, the rules, the skills: that’s where the bottlenecks live.
The new wave of founders and operators get this. They’ve built machines. Scaled companies. Navigated markets. They’re not dreaming about fusion in 2100. they’re shipping it in the 2030s.
Fusion won’t fix everything. But it will change everything around it. The moment it hits the grid, energy stops being the bottleneck and starts being the multiplier.
The world’s not waiting for perfect. It’s begging for power. And fusion is ready to answer — if we move fast enough.
Now the only question is: who gets there first?
Millennial Masters covers what’s next, from AI to energy to the edge of space:
Great coverage of the event, thanks!
Fascinating article. If we could just get democracy going a little coupled with removal of that childish profit motive thing, we'll have the job cooked and served in no time 🤓