You’re out of free articles.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Sign In or Create an Account.
By continuing, you agree to the Terms of Service and acknowledge our Privacy Policy
Welcome to Heatmap
Thank you for registering with Heatmap. Climate change is one of the greatest challenges of our lives, a force reshaping our economy, our politics, and our culture. We hope to be your trusted, friendly, and insightful guide to that transformation. Please enjoy your free articles. You can check your profile here .
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Subscribe to get unlimited Access
Hey, you are out of free articles but you are only a few clicks away from full access. Subscribe below and take advantage of our introductory offer.
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Create Your Account
Please Enter Your Password
Forgot your password?
Please enter the email address you use for your account so we can send you a link to reset your password:
The unsung hero of the energy transition needs a little help.

Transformers are the unsung heroes of the energy transition. These bulky devices come in a range of forms, from canisters mounted on distribution lines to garage-sized industrial units at substations and metal boxes on concrete pads outside apartment buildings. But regardless of form, they all serve a single essential purpose: adjusting voltage — either up, which is more efficient for long-distance transmission, or down for safe delivery to end users.
“If you’re getting an EV charging station, you need this equipment. If you’re building a large industrial facility that uses electricity to run the process, you need this equipment. Solar, battery storage plants, and wind energy projects need these. Data centers need these,” Anirudh Reddy, CEO of the power grid equipment startup Ayr Energy, told me. “It’s omnipresent.”
But ever since Covid scrambled global supply chains, the world has been staring down a severe transformer shortage. Even as economywide electrification accelerated, with rising electric vehicle adoption and rapid renewables deployment, the system failed to recover. Lead times for high-voltage transformers can now exceed two, three, even four years, raising costs and slowing the deployment of renewables and electrification projects.
While many of the big traditional players like Hitachi, Siemens Energy, and GE Vernova are still working to expand production, there’s now a group of startups looking to address the issue from other angles.
Reddy’s company, Ayr, is tackling it head on by standardizing transformer component designs and tapping underutilized manufacturing capacity in India to cut delivery times from years to months. Then there’s the growing field of startups that includes DG Matrix, Heron Power, and Amperesand, which are commercializing solid-state transformers for powering data centers, charging EVs, and solar and storage projects. This novel tech could offer an appealing replacement for today’s low- and medium-voltage transformers, as it provides grid services beyond just passive voltage conversions in a more efficient, compact package.
Right now, the most acute bottleneck and longest lead times are for high-voltage transformers, which is where Ayr sees the largest opportunity. These units step up a power plant’s output to transmission-level voltage for long-distance delivery, before stepping it back down to medium voltage at substations. These transformers can be the size of garages or small houses. Even when supply chains were functioning normally, they still took around a year to manufacture.
To return to that timetable, Ayr is taking its cues from automotive companies, which Reddy told me rely on standardized components throughout most of the manufacturing process and defer vehicle customization — think special features, finishes, and trims — until the final stages. When applied to transformers, this means equipment that is designed “such that you can address specification for a variety of projects and a variety of customers with the same architecture,” Reddy told me, allowing the company to shorten the time from initial customer engagement to final product delivery.
It all raises the question, why haven’t these equipment manufacturers been building more standardized transformers all along? Reddy blames it on a fundamental disconnect between transformer manufacturers and project developers. “That dialog never really happened. It was a very linear information flow,” he told me, with customers typically submitting bespoke specifications at the outset. That forced manufactures to build around these requirements from day one, making the whole supply chain and manufacturing process needlessly customized and complex.
But many of those unique design elements aren’t really necessary, Reddy said. Ayr often asks developers to tweak their specifications to align with a standardized model, which he told me yields no difference in overall project performance. That lets Ayr place orders with its Indian manufacturing partners in advance based on expected demand.
Surplus capacity in India is key to this business model, of course. Unlike U.S.-based manufacturers or those that export transformers and components to the U.S. — such as Mexico, Canada, and China — Reddy said Indian factories stayed ahead of local demand, even as electrification in the country ramped up. Lower labor costs allows factories to operate below peak capacity, providing an opportunity for Ayr to swoop in and contract with them immediately — even though many had previously produced almost entirely for the Indian market.
“It’s a highly manual, labor intensive process when you’re building bespoke heavy electric equipment. So the overhead of a factory that’s not producing equipment is pretty high,” Reddy told me. “Doing so in the U.S. would kill the company. Doing so in India, not so much.”
Ayr only emerged from stealth about seven months ago, and in the past year alone, Reddy said it’s built a backlog of over half a billion dollars in signed equipment orders, representing over 20 gigawatts of projects. The company started deploying its transformers and other backlogged power infrastructure in the middle of last year, and Reddy said it’s expecting its first high-voltage transformer to come online this quarter. So while the company has received early backing from venture firm General Catalyst — $3.5 million according to Pitchbook — Reddy told me it has no need to raise additional capital at this time.
Other transformer startups pursuing solid-state technology are bringing in plenty of venture capital — DG Matrix raised a $140 million Series B round and Heron Power raised a $60 million Series A round, both in February, while Singapore-based Amperesand secured $80 million last November — though their technology is only just beginning to commercialize. While conventional transformers rely on copper coils and iron cores to magnetically adjust voltage, solid-state transformers use power semiconductors — often made from silicon carbide — to perform voltage conversions electronically, while also enabling capabilities such as bidirectional power flow, AC/DC conversion, real-time voltage regulation, and rapid response to power surges.
But while these companies can also help ease the transformer shortage, their value proposition is distinctly different and likely to take longer to materialize than Ayr’s. For one, current solid-state transformer designs do not scale practically or economically to the high voltages required for the transmission grid. Instead, this novel tech is a better candidate to replace medium-voltage transformers on the distribution grid or low-voltage transformers inside facilities. This includes converting the low-voltage DC power produced by solar panels and batteries into medium-voltage AC power for the grid, delivering medium-voltage power from the grid to data centers and EV charging systems, and transporting low-voltage power around the data centers themselves, such as at the server rack level.
DG Matrix, for one, is primarily focused on data center applications for its solid-state transformers, which it initially plans to use to manage power generated onsite. The startup’s differentiating feature is its ability to balance electricity from multiple sources simultaneously, regardless of whether they’re operating on AC or DC power. For example, the company’s so-called “multi-port” device can integrate electricity generated from solar panels, natural gas generators, batteries, and the grid to provide power to data centers “in any ratio we want,” the company’s CEO Haroon Inam told me. It can also provide power back to the grid as needed.
Without such a unit, managing all these different sources would require significantly more space and numerous separate electrical components — protection equipment, multiple transformers for stepping voltage up or down, and converters to switch between AC and DC power. DG Matrix’s device thus promises to cut costs while boosting efficiency and reliability, especially for microgrid applications.
Now, Inam told me, the startup is looking at a “multi-billion dollar pipeline” for supplying low-voltage power to data centers for distribution inside the facilities themselves. It’s working to scale production at its North Carolina manufacturing facility, aiming to reach full capacity by July before pursuing a further expansion.
Heron Power, by contrast, is targeting a broader slice of the energy infrastructure market — in addition to data centers, it’s also seeking partnerships with operators of utility-scale solar and battery projects. Whereas DG Matrix is focused on coordinating onsite power from multiple sources, Heron is primarily interfacing with the grid, designing medium-voltage solid-state transformers that can step down power for delivery to end users such as data centers, and step up low-voltage power from solar and batteries to feed into the distribution grid — all without needing separate inverters for AC/DC conversions.
“It does what a traditional transformer does, but it also does what switchgear does, what tap changers do, what capacitor banks do, and what a synchronous condenser does, in a single package, managed in real time by software,” Heron’s CEO Drew Baglino told me via email.
During Baglino’s prior tenure leading the development and deployment of Tesla's EV Superchargers and large-scale battery storage systems, he constantly found conventional transformers to be “a supply chain constraint, a deployment bottleneck, and a physical hazard,” leading him to believe there simply had to be a better way. But because solid-state designs are driven by software-controlled power electronics rather than fixed hardware, that fundamentally shifts the paradigm, he said. “A traditional transformer that's wrong for the job gets ripped out and replaced. A Heron Link gets a firmware update.”
The company has publicly announced just two customers to date, but they’re big ones — clean energy developer Intersect Power and data center developer Crusoe. Overall, Heron said it’s secured over 50 gigawatts worth of orders, and is now working to build out a U.S.-based factory capable of producing 40 gigawatts of transformers annually, with initial production beginning next year.
Even if the transformer shortage resolves sooner rather than later, Inam doesn’t expect it to dampen interest in DG Matrix’s technology or solid-state transformers overall. Their ability to integrate multiple energy sources, he told me, “provides a more economic path to distributed generation” — especially as microgrids become an increasingly common way to circumvent the interconnection queue and generate power onsite.
“The fastest area of demand growth for transformers is for new energy technologies like solar, batteries, and data centers,” Baglino told me. So while he declined to speculate on when the shortage will end, he too expects continued demand. As for Reddy, he thinks the supply crunch is unlikely to fully ease until after 2030, when he predicts leading manufacturers will be able to fully ramp up the new transformer facilities and expansions that they’ve announced. .
At any rate, the frenzied data center buildout certainly shows no signs of waning, with the International Energy Agency projecting that global data center electricity consumption will double by 2030 and more than double in the U.S., where it will rise by about 2.3x. And with recent volatility in fossil fuel prices likely to accelerate the electrification of certain markets, transformers are set to remain as ubiquitous and critical as ever.
To meet the moment, Baglino said, “What's needed is the same concentration of focus, urgency, and scale that transformed electric transportation, directed at the infrastructure that powers everything else.”
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
The move by University of Pennsylvania researcher Danny Cullenward intensifies a debate over integrity at the carbon accounting organization.
A well-known scientist has resigned from the independent oversight board of the Greenhouse Gas Protocol, renewing questions about the integrity of one of the world’s most important arbiters of carbon emissions standards.
Danny Cullenward, who is also an economist and lawyer, notified the organization’s leadership on Monday that he no longer has “any confidence in the Protocol’s governance structure,” according to his resignation letter, which he posted publicly. He had previously tried to sound alarms about the organization and its lack of transparency in a paper he published in April.
Cullenward’s resignation letter goes a step further, accusing the Protocol of covering up an internal complaint he and a fellow board member filed, and of handing the reins of at least one of the organization’s standards to “a secret, industry-dominated drafting process.”
The Greenhouse Gas Protocol declined to comment on Cullenward’s resignation or answer questions about his account of events leading up to it.
The Protocol launched in the late 1990s as a joint project of the World Resources Institute, an environmental group, and the World Business Council for Sustainable Development, an industry association. Today it is the world’s leading standard-setter for corporate carbon accounting. More than 22,000 businesses rely on its methodologies to calculate and report their emissions. While adhering to the Protocol’s standards is still mostly voluntary, it will soon become a requirement under European Union and California disclosure rules.
Cullenward’s accusations arrive in the middle of a major revamp at the organization that began in 2022, designed specifically to improve the integrity of its corporate accounting standards. As part of the overhaul, it also put in place a new governance structure to improve transparency and accountability. Technical working groups made up of external experts would develop proposals to revise the standards to more accurately capture companies’ full carbon footprints, and then an Independent Standards Board would review and ultimately approve them. The Protocol appointed Cullenward to the independent board as one of its inaugural members in September 2024.
Cullenward’s reasons for leaving, as described in his letter, center around the development of a forest accounting standard to be used by companies that manage forests or have wood in their supply chains. The technical working group assigned to develop the standard could not reach a consensus, and ultimately submitted two competing proposals to the Board. Members associated with landowner groups and the forest products industry authored one of them, while the group’s research scientists primarily wrote the other.
According to Cullenward’s letter, as well as memos written by the academic scientists in the working group reviewed by Heatmap, the industry proposal, known as the “managed land proxy” method, would enable companies to claim they were removing carbon from the atmosphere when they cut down trees or used virgin wood. “This is the opposite of what physically happens when a forest is cut down,” Cullenward writes.
The method produces this counterintuitive result by allowing companies to take credit for all the carbon sucked up by the forests they manage, or in some cases by all the forests in a region, even if the company had no part in boosting that sequestration. If companies were to apply this accounting method to their products, Cullenward adds, not only would making virgin paper appear to involve zero carbon emissions, it would also apparently help to restore the climate. It would also look much more advantageous to the climate than producing recycled paper.
His concern is not just with this proposal, but also with how the Protocol handled a complaint filed by a proponent for the managed land proxy approach that challenged the scientists’ expertise. In response, the organization quietly solicited opinions from additional outside scientists on the two proposals.
Cullenward’s letter asserts that this was a decision made solely by the board’s chair, Alexander Bassen, alongside Protocol staff and without the rest of the board’s input. He writes that when these external comments were later shared with him and his fellow board members, the authors were “presented as neutral arbiters of a contested scientific debate,” even though they had been specifically referenced in the complaint as supporters of the managed land proxy approach.
Cullenward says he tried to “pursue internal accountability” but faced retaliation. In February he and another board member, an Australian forest ecologist named Heather Keith, filed an official complaint. The Protocol enlisted an outside mediator to resolve their dispute, but Cullenward says the hired adjudicator failed even to read the full complaint before meeting with him. The mediator also did not review any of the recordings of key board meetings referenced in the complaint, and was barred from speaking to technical working group scientists.
Cullenward and Keith eventually received a response to their complaint from the mediator but were told they could not share it, and the matter was deemed closed. According to a spokesperson for the Greenhouse Gas Protocol, who reached out to me with an update on the matter in late May, an independent review found “some process shortcomings” but “no material breach” of the organization’s rules or of due process. They added that “recommendations to address process shortcomings and strengthen conflict resolution are being reviewed and implemented.”
I reached out to Keith, who told me in an email that she was “deeply concerned about Danny’s resignation.” She praised his “wide-ranging expertise” in carbon accounting, law, and governance, and his “extensive contributions” to the board’s discussions. “One of the most valuable assets in a Board member is his demonstrated independence in making judgements that is based on a sound knowledge of climate science,” she wrote. The board “should be encouraging more people with Danny’s expertise and motivation for climate action to benefit the global community, not losing such valuable people.”
Cullenward’s primary concern moving forward is a new partnership between the Greenhouse Gas Protocol and the International Organization for Standardization, which establishes technical specifications for a range of industries and purposes, to unify their emissions accounting rules. The two groups’ first joint undertaking is to develop a standard for assigning emissions to specific products, which will include forest carbon accounting.
While the Greenhouse Gas Protocol has publicly listed the members it assigned to the joint working group, the ISO is under no obligation to do so. Cullenward asserts in his letter that the new joint groups “operate with confidential membership that is heavily tilted in favor of industry interests.” He says a representative from the World Business Council for Sustainable Development told him that the group may draw on an existing ISO standard based on the managed land proxy approach.
Meanwhile, over a year after the corporate forest accounting technical working group submitted its proposals, the Independent Standards Board is now contemplating kicking off a seven-month public comment period on the recommendations, Cullenward writes. He concludes that this elongated comment period is just for show, and that the issues “have already been delegated” to the joint working group with the ISO.
I asked the Greenhouse Gas Protocol how it planned to ensure “transparency and accountability for its stakeholders,” as it has previously promised, when the membership and meeting minutes of the joint ISO working groups are not disclosed to the public. I also asked, for the second time, whether the organization plans to publish meeting minutes from Independent Standards Board meetings — a requirement under the board’s governing rules that it has not followed. The Protocol declined to answer.
The U.S. electric vehicle maker’s make-or-break model, the R2, is finally here — and it’s pretty fun to drive.
The attainable Rivian is here, and not a moment too soon.
It’s been nearly a decade since the U.S.-based startup revealed its prestige R1T pickup truck and R1S SUV, earning plenty of “the next Tesla” hype and becoming lots of people’s favorite electric car brand. But with those R1 vehicles starting around $70,000 — and with nicer versions hitting six digits — lots of would-be drivers have been waiting for R2, the scaled-down vehicle first announced in 2024 and meant to take Rivian to the masses.
Now the moment has arrived: On Tuesday, Rivian began shipping the first version of the R2. I had the privilege of test-driving the vehicle that will make or break the brand last week on the highways and mountain roads outside Park City, Utah. If my experience is any indication, R2 is up to the job of making Rivian mainstream.
“A word we used really heavily throughout the development of R1 was … inviting,” CEO and founder RJ Scaringe said to the journalists at last week’s event. “We use that in the sense of inviting people to use it, inviting people to get it dirty, inviting people to have new experiences and new adventures in it. But by virtue of it being a flagship product, its price wasn't as inviting as we wanted. And so R2 really in many ways is the culmination of the full brand promise.”
First, the facts: R2 looks at first glance like a smooshed version of Rivian’s big SUV, with the same signature headlights and basic shape. It’s a little shorter, a little narrower, and 2,000 pounds lighter than its big cousin, seating five people as opposed to the seven that can cram into R1S. Range from the 88-kilowatt-hour battery is in the high 200 miles and tops 330 miles for some editions.
The stat that matters most is price. The first R2s out of the gate will cost around $58,000, and gradually less expensive tiers will arrive later this year and into next, culminating in the $45,000 base version at a yet-to-be-determined date. No, an EV around 50-grand doesn’t sound like a car for the common man. But as Scaringe noted, that is now the average price for a new car in America, which certainly makes R2 attainable for millions more drivers.
It’s also a lot of car for that money. Thanks to its boxiness, R2 feels like it has loads of room on the inside. Because of an improved battery shape, there’s actually more legroom for the rear passengers compared to R1. Double gloveboxes and a pretty big frunk add to the available storage space. (Rivian even fixed a pet peeve of R1 owners who couldn’t fit their monstrous water bottles in a convenient spot.)
Yet R2 doesn’t drive big. It rides high and offers the driver a wide view, but it’s not a tank like R1, which I found difficult to park in compact spaces like the one at my home. Its 5,000-pound weight is still a lot of heft (a Tesla Model Y is more like 4,000 to 4,400 pounds), but the car still feels zippy. The mass is simply overwhelmed by electric power, especially in the higher-end versions Rivian let us drive in Utah.
As the engineers on site noted, developing the R2 was mostly an exercise in subtraction — not just shrinking the physical size from the R1, but also making R2 cheaper to build by removing miles of wiring (something the brand visualized at the event by showing off bundles of copper in the style of a rubber band ball, representing all that had been cleaved). But R2 needed its own bells and whistles so it would feel desirable on its own and not appear to be merely a discount Rivian.
Those additions include rear windows that go all the way down, unlike the halfway that’s common in most passenger cars; the rear windshield descends, too. A fun button up front marked with a “5” will lower all four passenger windows plus the back windshield at once. In response to complaints about every function running through the center touchscreen, Rivian put in some buttons — or, rather, some wheels. On each side of the steering wheel, reachable by a person’s thumbs, are haptic “halo” buttons that can be pushed side to side or spun. These are not at all the subtle, slight wheels you’d find a Tesla, but rather beefy spinners meant to feel rugged and easy to manipulate.
During testing, I struggled with how hard to push them and in what direction to enter the desired mode that could then be adjusted by spinning the wheel, be it climate, music, drive mode, or the positioning of the side mirrors. But something tells me Rivian will refine the haptic feedback as R2 owners put miles on their vehicles. And even complicated or layered menus become second nature once it’s your own car.
Many of these vehicles will never go off-road, but Rivian still had to prove the R2’s backcountry bona fides. This is the adventure EV brand, after all, and part of the pitch for R2 is how much more it can do than a Tesla Model Y or Chevy Equinox EV. Keen to prove the point, Rivian swapped us halfway through the test drive into R2s with their tire pressure halved to make them mountain-ready, then directed us onto the rutty, boulder-pitted roads of Wasatch Mountain State Park to wade through water crossings and up to the top of a plateau. Here the touchscreen becomes an adventurer’s dream, displaying the vehicle’s moment-by-moment elevation, pitch, compass direction, and much more. Tap into the camera system and it can bring up the close-up view of what’s right in front of the vehicle and shows both front wheels to help navigate around pointy rocks and cavernous ruts.
R2 never wavered or felt as if it had taken on too much. It has all the capability you’d need as a trail warrior, and more than enough for the affluent professional who yearns to become outdoorsy. After so many decades when the world’s truly rugged vehicles were also low-mile-per-gallon polluters, it feels like a breakthrough just getting this much can-do spirit out of an electric car.
More salient for the urban dweller is Rivian’s big push into autonomous driving. As we noted in December after the brand’s AI and Autonomy Day, R2 is the company’s big play in that race: It vastly ups the amount of road open to Rivian’s hand-free autonomous driving feature works, raising it to about 3.5 million miles in the U.S. The company also told us that by the end of the year it would introduce point-to-point service, where the vehicle really can drive itself for the duration of a trip, with more autonomous features potentially on the way. During the test drive, the hands-free tech felt steady and assured on twisty local roads.
Rivian has a long way to go here, given Tesla’s major head start in developing vehicle autonomy. One big asset it does have is the thousands of drivers who’ve bought R1s and who opted to share their driving data with the company, helping it build a dataset that maps and models the world. The less expensive R2 should get many more people into a Rivian vehicle and accelerate that learning curve. That, plus the eventual addition of a LIDAR sensor to some models, will allow that kind of full autonomy that R2 will use when it goes into service as an Uber robotaxi following the ride-sharing company’s $1.2 billion investment earlier this year.
It’s difficult to overstate the importance of this vehicle for Rivian, or for the electrification of the American car. For the brand, this must be its Model 3 moment, where it leaps from a niche brand selling luxe status symbols to one that builds a huge number of EVs — and in the process hopefully becomes financially stable after years in the startup “valley of death,” between promise and profitability. Billion-dollar investments from the likes of Volkswagen and Amazon buoyed Rivian during those years; now R2 has to deliver on them.
As for the U.S. EV market as a whole? It also needs the R2. New EV sales are sagging in America, even amid gasoline price shocks caused by the Iran War. A $50,000 Rivian isn’t exactly the solution to the auto industry’s affordability crisis, but Scaringe argued that U.S. buyers also lack great choices. The industry leaders — Tesla’s Model 3 and Model Y — have been on the market since 2018 and 2020, respectively, with subtle tweaks and update since then. New offerings from legacy carmakers like Chevy and Toyota are a welcome change. Still, they feel like a Chevy or Toyota that’s been electrified, not like a vehicle built from the ground up to deliver on the promise of what a great EV can be.
Yet even now, the learnings from the EV startup world that led to R2 — dramatically simplified manufacturing to bring down costs, advanced touchscreen infotainment with elegant interfaces, EVs built fully integrated from the ground up rather than adapted from existing gas cars — are already influencing the rest of the industry. Just look at what Ford’s skunkworks operation is up to as the Detroit giant tries to catch up in the EV race starting next year. A successful R2 would push the car industry further in this direction.
R2 succeeds in bringing the feeling of a lusted-after EV to the five-seat, fully capable SUV, which has become the de facto family car of this country. And for all of Rivian’s focus on catching up in the AI and autonomy race, R2 still feels like a car you’re supposed to love to drive yourself, whether that’s to work, to grandma’s, or to the top of a mountain. It is, indeed, inviting. With Tesla having publicly abdicated its role of building great EVs for humans to drive, Rivian is now primed to seize the position.
Current conditions: China has triggered emergency warnings across six provinces as heavy rainfall floods the countryside • A magnitude 7.8 earthquake struck the Philippines, leaving at least 32 dead and more than 100 injured in building collapses • Temperatures in Albuquerque, New Mexico, are rising near 100 degrees Fahrenheit.
On Tuesday, Tennessee is set to become the first state in the nation with its own regulatory framework for nuclear fusion plants. You may be wondering, why Tennessee? The two-word answer: Oak Ridge. The Volunteer State has operated as a hub for nuclear energy research and development for more than 60 years, feeding off both the Oak Ridge National Laboratory and the Tennessee Valley Authority’s capacity to help commercialize new technologies. Now state regulators are establishing the first dedicated rulebook for building future fusion plants. “Tennessee has been named the top state in the nation for nuclear energy industry growth, and for good reason,” David Salyers, the commissioner of the Tennessee Department of Environment and Conservation, said in a statement. “This latest step supercharges our reputation as the global hub for nuclear innovation and positions us as the most responsive state to new advanced nuclear companies clamoring to call Tennessee home.”
It’s not the only government betting that the various attempts to commercialize fusion as an energy source will pan out in the near future. On Monday, NucNet reported that the British government had drafted legislation to “create conditions” for deploying fusion technology.
Typically, the rule of thumb in journalism is that the answer to a question headline is almost always “no,” otherwise the headline would simply state the fact. But this one is a genuine open question that climate-tech investor Shanu Mathew raised Monday in a post on X: Could PJM Interconnection, the nation’s largest grid operator, break apart? The speculation traces back to a Bloomberg article from last week in which unnamed federal officials suggested that the operator, which runs the grid from the Illinois prairie to the Jersey Shore, could split up as data centers put strain on the 13-state system’s electricity supplies.
The talks are happening as two of the largest utilities in PJM, NextEra and Dominion, discuss a potential $420 billion megamerger that would create, among other things, a storage giant, as Heatmap’s Matthew Zeitlin reported. The discussions are also occurring against the backdrop of major artificial intelligence companies going public, with ChatGPT-maker OpenAI following Claude-developer Anthropic in filing a confidential S-1 with the Securities and Exchange Commission this week.

In the United States, you can’t build a single commercial nuclear reactor in a decade. In China, you can apparently double the size of your entire fleet in that time. Between 2016 and 2024, China’s nuclear generation capacity soared by 76%, according to a new Energy Information Administration analysis. That’s equal to 24 gigawatts. In 2025, China added another 1.1 gigawatts, followed by 2.2 gigawatts more this year just through May. The country has at least 36 other reactors under construction, accounting for nearly half of the world’s ongoing nuclear projects.
Sign up to receive Heatmap AM in your inbox every morning:
Just five years ago, the global aviation industry made a landmark pledge to achieve net zero emissions by 2050. Now the head of the industry’s global body says that goal is likely already out of reach. Willie Walsh, the director of the International Air Transport Association, told The Guardian that “hope was fading fast” and a new “realistic timeline” needed to be established. More than half of the planned decarbonization of air travel relied on the development of sustainable aviation fuels that remain nascent at best. Money is pouring into the technology, as Heatmap’s Katie Brigham reported. But uptake so far “is about 0.2% of fuel,” Nicole Cerulli, a research associate for transportation and logistics at the market research firm Cleantech Group, told her.
One cold autumn morning three years ago, I made my way across downtown Ulaanbaatar to an American-style diner called Millie’s Espresso to meet with a Mongolian mining executive who was thrilled about Western countries’ recent investments in his industry. Landlocked between Russia and China, the geographically huge but sparsely populated democracy hoped to shore up its sovereignty by forging deals with the U.S., Europe, South Korea, and Japan to satisfy soaring demand for minerals. Already Oyu Tolgoi, one of the world’s largest copper mines, was underway in the country’s Gobi desert south, and that year the French government inked a deal to start producing lithium and uranium in Mongolia. Now the uranium part of that agreement is moving forward. On Monday, World Nuclear News reported that the French state-backed nuclear fuel producer Orano had broken ground on its first mine in the Central Asian nation. The project raised some eyebrows among Mongolians who complained that Soviet-era Russian uranium mining left behind nasty pollution, and the terms of Ulaanbaatar’s deal with Rio Tinto over the new copper mine have been politically contentious. But the sprawling, smog-choked capital city — the only major urban development in the rural nation — is in need of more power.
Russia had promised to help meet that power by building Mongolia’s first nuclear power plant. A politically well-connected businessman from Ulaanbaatar, whom I caught up with last night over text to ask about the mood in the country, said Moscow’s bid had drawn more positive attention than France’s plans to mine fuel for their own reactors. “In Ulaanbaatar, we experienced electricity shortages last winter that caused apartment heating to stop during the winter. It was crazy,” the executive told me. While he’s typically a critic of the ruling Mongolian People’s Party, which formed out of the old Communist Party apparatus following the fall of the Soviet Union, the executive told me the government’s actions were “good and brave” steps to “diversify investment in Mongolia.”
I hate to close out on a bad note, but this one felt important to include: America’s screwworm problem is getting worse. On Monday, the U.S. Department of Agriculture confirmed the first case of the flesh-eating parasite in a dog in New Mexico, in addition to four cases in total in Texas. “This situation is evolving, and we expect new information to emerge as our investigation continues,” Dudley Hoskins, USDA’s under secretary for marketing and regulatory programs, said in a statement.