Nuclear fusion, the ‘holy grail’ of power, was always 30 years away—now it’s a matter of when, not if, fusion comes online to power AI

The project’s principal designer, nuclear physicist Annie Kritcher, wasn’t content to keep the science in the lab after achieving what she deemed the “Wright brothers’ moment” for fusion. Kritcher cofounded Inertia Enterprises in August to bring the power to the actual grid. The potential promise of fusion is for consistent, clean power without radioactive waste, intermittency issues, or the dependence on foreign supply chains.

Inertia isn’t a lone startup promising hopes and dreams. There’s a group of companies now pursuing the commercialization of fusion within a decade—not some far-off timeline. The bottom line is many more scientists and business analysts are now convinced fusion energy powering our homes is just a matter of when, not if, even if the timeline estimates remain overly optimistic.

Roughly 60 years ago, pioneering Soviet physicist Lev Artsimovich said fusion power will be ready “when society needs it.” The combination of advances in science, technology—supercomputing and superconducting magnets—and, critically, money from AI hyperscalers and others makes fusion power a realistic option when the world is demanding much more electricity.

“Fusion is the holy grail of energy. It’s a clean, no-carbon, unlimited fuel source,” Kritcher told Fortune. “It’s powering hope for our generation and future generations to come.”

Whereas traditional nuclear fission energy creates power by splitting atoms, fusion uses heat to create energy by melding them together. In the simplest form, it fuses hydrogen found in water into an extremely hot, electrically charged state known as plasma to create helium—the same process that powers the sun. When executed properly, the process triggers endless reactions to make energy for electricity. But stars rely on overwhelming gravitational pressure to force their fusion. Here on Earth, creating and containing the pressure needed to force the reaction in a consistent, controlled way remains an engineering challenge.

“To power one person’s lifetime, it’s a bathtub of seawater and a laptop battery’s size of lithium,” Kritcher said. “It’s not a lot of materials, and there’s no [long-term] radioactive waste like we have with fission.”

Microsoft founder Bill Gates says the “coolest” things he’s working are fusion energy and next-generation nuclear fission projects.

“If you know how to build a fusion power plant, you can have unlimited energy anywhere and forever. It’s hard to overstate what a big deal that will be,” Gates said in an October 2 essay. “The availability and affordability of electricity is a huge limiting factor for virtually every sector of the economy today. Removing those limits could be as transformative as the invention of the steam engine before the Industrial Revolution.”

Gates’ Breakthrough Energy Ventures investment firm financially backs fusion companies such as industry leader Commonwealth Fusion Systems, Type One Energy, and Zap Energy. The biggest challenge, as with all technology, is building the first one, Gates wrote. “We’re on the cusp of massive breakthroughs, and it’s clearer now than even before: The future of energy is subatomic.”

Kritcher’s project uses the world’s largest laser system at its California lab. She started Inertia with Jeff Lawson, the cofounder and former CEO of the Twilio cloud communications company and the current owner of The Onion satirical news publication. As CEO, Lawson’s bullishness on fusion is no joke.

Inertia follows the proven science while other companies use different fusion approaches that haven’t yet worked at scale, Lawson said. “That’s why fusion has been so elusive for decades. That’s why the running joke was always that fusion is 30 years away, and it’s been 70 years,” Lawson said.

“Now the big challenge we have is we have to go build the world’s largest laser, so that’ll be interesting and fun,” he said, laughing. He is hoping to complete the first pilot plant in the mid-2030s.

Inertia isn’t the leader in the fusion clubhouse when it comes to funding or construction, but it’s one of several in the hunt to prove it has the most scalable and affordable approach worldwide.

There likely will be multiple eventual fusion winners, said Prakash Sharma, head of scenarios and technologies for the Wood Mackenzie energy research firm, which projects that global electricity demand will nearly double by 2050, requiring $18 trillion in new investments.

While fusion power might enter the grid in a decade, he said, it will be closer to 2050 or beyond when fusion can grow to claim a notable chunk of the grid.

“It’s a question of when fusion becomes available, rather than a question of if,” Sharma said. “The challenges are being overcome, especially given the momentum behind the technologies and the interest from a number of different players like Google and Microsoft.”

Race against time

The Gates-backed Commonwealth Fusion Systems (CFS) leads the fusion space in funding, contracts, and has the advantage of being founded earlier than most in 2018 through a spinoff from the Massachusetts Institute of Technology—well before the Livermore breakthrough.

CFS recently inked power purchase deals with Google and the Italian energy giant Eni for its first commercial fusion plant, ARC, slated to come online in the early 2030s just outside of Richmond, Virginia. If all goes as planned, which is no sure thing, the 400-megawatt plant would become the world’s first fusion plant providing steady power to the grid—enough to power about 300,000 homes. CFS is racing against competitors such as Helion—backed by OpenAI’s Sam Altman and SoftBank—which aims to build a fusion plant east of Seattle to power Microsoft data centers.

CFS’s pilot project, SPARC, is under construction outside of Boston and is expected to open by 2027.

“In the history of technology, getting there early is by far the most important thing,” CFS cofounder and CEO Bob Mumgaard told Fortune. “We need a power plant making power, and we need that is as soon as possible.”

“We build stuff in New England in a way that’s not quite the same as Silicon Valley,” he added with a laugh.

Whereas Inertia uses lasers at Livermore, CFS is the leader in the most common form of fusion tech—the oddly named “tokamak.” The tokamak—shortened from toroidal chamber magnetic—uses powerful magnets. The design essentially involves a massive, doughnut-shaped machine that traps the plasma in a high-temperature, superconducting magnetic field.

The world’s largest tokamak, the long-delayed, research-focused ITER (international thermonuclear experimental reactor)—backed by more than 30 nations—isn’t expected to come online until 2035 in France. CFS aims to beat that timeline with a smaller, more efficient project for the grid. Time will tell.

Even Mumgaard isn’t convinced the tokamak is the best, long-term solution for fusion power. But it’s the best right now, he argues, and scientists are the most knowledgeable about the approach.

“You need to get there and get product into the world that works. The tokamak works, and it has the strongest scientific basis,” he said. “ITER is a [$30] billion statement of conviction that it’s probably going to work. Does it have its flaws? Absolutely. But the key is you know what they are.”

And Mumgaard is particularly bullish that fusion power will scale up quickly in S-curve fashion once the first commercial plants come online and prove out the science and tech. “The world is pretty good at building things fast when those things can make money and when there’s not a lot of constraints on the materials,” he said.

He’s optimistic Google is just the beginning of its deals with Big Tech to buy fusion power for the AI and data center boom. Fusion startups have secured roughly $10 billion in private funding in recent years—about $3 billion has gone just to CFS—but they still need substantially more to build commercial facilities. CFS also counts Nvidia, Mitsubishi, and more among its supporters.

“For hyperscalers, you have a buildout of infrastructure that’s very energy hungry. They can afford to spend on new technology,” Mumgaard said. “They need a lot of power in a concentrated way. They need it all the time. The use case fits fusion very well. The mindset fits fusion very well.”

On the other side of the world in New Zealand, Ratu Mataira was amazed as a student at the Victoria University of Wellington when CFS leaders came to study his college’s superconductor magnet research.

Mataira, 33, founded competitor OpenStar Technologies in 2021, taking an “inside out” approach to the tokamak. With OpenStar’s levitated dipole tech, the plasma surrounds the magnet—and isn’t confined within it—in a controlled chamber environment. The setup creates a contained magnetosphere, similar in principle to the magnetosphere that surrounds the Earth, generated by its North and South poles.

If proven out, OpenStar could build fusion reactors that are smaller, cheaper, faster to build, and easier to maintain and operate, Mataira said.

OpenStar successfully created its first plasma in November—at 540,000 degrees Fahrenheit, hotter than the surface of the sun—although Mataira recognizes the technology remains young compared to competitors and still has a lot to prove.

“The tokamak is is the devil we know, but that is also its key weakness because it’s not a fundamental strength,” Mataira said.

And he’s confident that harnessing the power of the stars for humanity remains an inevitability, he said. Most of the science and technology are solved or nearing the finish line. What remains are the physical engineering at scale and the finances. “The argument is it’s now really an engineering problem, and humans are pretty good engineers. We know how to solve these kinds of things.”

Funding remains a very real issue for the industry, he said, but Mataira insists the nascent industry is taking the issue seriously. “Because fusion is pure technology and pure capital, the economies of scale and cost allow us to project being able to bring those costs down over time. Eventually, fusion will be the dominant energy source.”

“Fusion isn’t just a billion-dollar opportunity; it’s a trillion-dollar opportunity. It’s the kind of thing that shifts the geopolitical balance,” Mataira insists.

What comes next

The other Gates-backed fusion player that’s making big deals is Type One Energy, which has a non-binding agreement with the Tennessee Valley Authority utility to convert the Bull Run Fossil coal plant that was retired two years ago into fusion.

Type One’s Infinity Project includes the Infinity One pilot plant and, in September, announced plans to develop the first commercial plant at Bull Run—the 350-megawatt Infinity Two project.

CEO Chris Mowry sees Infinity Two coming online by the early 2030s, putting Type One firmly in contention for the first grid-scale plant. Mowry was recruited to Gates’ Breakthrough from the more traditional nuclear fission world that makes up less than 10% of the global grid.

“Nuclear fission has been kind of stuck that way for 30 years,” Mowry said. “The world obviously needs a ton more of energy that is sustainable and climate friendly in addition to reliable and resilient, and I think that’s fusion.”

Type One uses stellarator fusion technology, which is a literal twist on the tokamak design by adding external coils that create a twisting magnetic field to better control the flow of the plasma. The downside is the stellarator is more complex and expensive than the more standardized tokamak. Mowry contends the stellarator eliminates the tokamak’s instability issues that remain unresolved.

The doughnut-shaped magnetic technologies in stellarators and tokamaks have struggled with imperfections or so-called holes in their magnetic confinement systems, allowing particles or runaway electrons to escape, disrupting the sustained fusion reactions and undermining performance. Modern modeling and supercomputing are helping to predict and correct the flaws, but more progress remains.

Unlike other fusion developers, Type One only aims to design and manufacture equipment for the plants—not own and operate them—dramatically cutting down on its capital costs and speeding up Type One’s potential to ramp up.

Mowry touts fusion’s advantages from both the materials and political regulatory standpoints. For instance, fusion doesn’t require all the expensive, nuclear-grade concrete that creates a protective dome encasing traditional nuclear radiation.

And the U.S. government isn’t making fusion go through the same regulatory permitting hurdles as fission, speeding up the process potentially to months instead of years. Specifically, the Nuclear Regulatory Commission will license fusion projects under the limited scope of its existing byproduct materials framework.

“[Fusion] isn’t intertwined in nuclear nonproliferation regulations and export restrictions,” Mowry said. “You ought to be able to build one of these things in three to five years when you get good at it.”

Whereas the Trump administration is targeting and attacking renewable energy power, especially wind projects, the White House is promoting the fusion technologies developed at the country’s national labs.

U.S. Energy Secretary Chris Wright toured CFS’s SPARC facilities on Sept. 29, declaring that fusion will help “American energy dominance reach new heights.”

Jefferies is one of the only investment banking firms studying the fusion space, and energy transition analyst Charles Boakye sees fusion becoming a major part of the grid.

Boakye initially feared fusion would get caught up in the partisan “climate conversation” in the U.S. “I think now the conversation is an access to energy conversation, and that perhaps has broader political support and is less controversial.”

But it’s going to take another 15 or 20 years or so after the first commercial plants come online to truly make a dent—and that’s a rapid, optimistic timeline, he said. “It took solar power 25 years to reach its first terawatt [worldwide], and then it took two years to reach its second terawatt. Once you hit that inflection point, you start to see real gains,” Boakye said.

But the potential for fusion is much larger, he said. Fusion is so energy dense and potent it could become the “final energy source” when it’s finally optimized for the grid.

“Going from wood to coal to gas, we’ve always increased the energy density,” Boakye said. “Fusion would be the final source of energy density.”

And, as OpenStar’s Mataira says, the clock is ticking.

“If we’re working toward 2050 climate goals, and we’re not beginning the process until the late 2030s of actually scaling and deploying these systems, then it’s basically too late.”