Back
LAS VEGAS — On Monday night in a subterranean hall under the Las Vegas Convention Center, Tesla released an upgraded version of its grid-battery product that will allow developers to build bigger energy-storage projects faster. That kind of acceleration is sorely needed as the storage industry positions itself to meet historic grid demand in the next few years.
While better known for its pioneering electric-car business, and the polarizing antics of CEO Elon Musk, the company is also a pacesetter in the fast-growing U.S. energy-storage industry.
Tesla’s white boxy Megapack product, which stitches together lithium-ion batteries inside a large container, has been a top competitor for years. Around the U.S., Megapacks play a crucial role in keeping the lights on: In Oahu, they enabled the safe retirement of Hawaii’s last coal plant; in Oxnard, California, they allowed the city to avoid building a gas plant on its coastline; across Texas, they’re helping lower electricity prices and avoid shortfalls during record heat waves, as are batteries from companies like Fluence and Wärtsilä.
But the storage industry is still young, with plenty of room to streamline operations and bring down costs. That’s what Tesla hopes to do with the new Megablock, which packages four Megapacks around one transformer.
One of these blocks holds 20 megawatt-hours of power, which can be discharged for up to four hours at peak capacity. Scaled up for a large project, 248 megawatt-hours can fit into an acre (for comparison, the Oxnard project packed in about 200 megawatt-hours per acre using earlier-generation Megapacks back in 2021).
Tesla is taking orders for Megablocks now and expects to ship starting in late 2026. The firm will manufacture them near Houston, with lithium-iron-phosphate batteries from multiple sources, including a 7-gigawatt-hour-per-year manufacturing line planned to be completed at the company’s Nevada Gigafactory in early 2026.
The timing is no coincidence. Tesla’s new announcement comes as AI computing gobbles up electricity in unfathomable quantities. The rapid construction of new power sources has emerged as a defining imperative for America’s tech industry as it races to achieve what it sees as the transformative benefits of advanced AI. President Donald Trump has claimed that global AI supremacy is a key priority, even as his administration has taken aggressive steps to choke off development of the nation’s fastest-growing sources of energy.
Tesla says the Megablock design will allow developers to deploy 1 gigawatt-hour’s worth of storage in just 20 business days. That’s an astonishing rate, if borne out in real-world conditions. The firm has made bold claims in the past that have failed to materialize on time if at all, like its visions of a widely adopted solar roof or a huge autonomous taxi fleet. But the Megablock doesn’t hinge on a fundamentally new product; it’s another step in the steady evolution of a flagship technology.
The Megablock’s main innovations are that it reduces the amount of electrical work required in the field while also packing in battery cells as densely as possible without going over the weight limit that triggers expensive specialized shipping protocols.
“For us, one of the key metrics was, what’s the maximum percent cell mass you can get?” said Mike Snyder, Tesla’s VP of energy and charging, in an interview after his presentation. “Because the cells are what matters. So we made sure we increased that. It’s an 86,000-pound box, and 75% of that is cell.”
A giant battery installation requires thousands of perfectly executed electrical connections, and mishaps can cause major problems. Megapacks come with their batteries pre-wired, allowing for factory-grade quality controls. But currently, each pack then needs to be connected to a medium-voltage transformer to ship power in and out, which takes up to 24 individual connections per pack.
“That’s just a lot of labor in the field, and it’s a lot of places where something can go wrong,” Snyder said. “One of those bolts, one of those cables, it causes downtime and you have to go fix it.”
The new Megablock, in contrast, needs just three connections per pack.
Tesla timed the unveiling for the opening night of RE+, the bustling solar and clean energy industry conference, but hosted it on Musk-affiliated turf: a station for his side project boring holes under the Las Vegas Convention Center. Attendees were invited to experience the thrill of being chauffeured in a manually driven Tesla through a one-lane tunnel — not necessarily a harbinger of the future of transport, but it was lit by colorful LEDs.
Musk himself was not on hand, but Snyder addressed the crowd from a stage that was also lit by LEDs and flanked by Cybertrucks and Tesla’s new humanoid automatons. The screen behind him lit up with sharply produced drone footage of massive Megapack installations in lush locales.
The spectacle offered an implicit riposte to the Trump Department of Energy, which a few days earlier had tweeted, “Wind and solar energy infrastructure is essentially worthless when it is dark outside, and the wind is not blowing.” The claim reflected either a remarkable ignorance of energy storage, a longtime research and deployment priority of that very same department, or a desire to pretend batteries don’t exist.
Batteries accounted for 23% of new grid-scale capacity built in the U.S. last year, compared to just 4% of new capacity that came from the fossil-gas plants much admired by the Trump administration.
While Tesla’s CEO spent hundreds of millions getting Trump elected and a few months slashing the federal civil service, Tesla’s engineers kept hacking away at the problem of making better batteries.
The attention to detail goes down to the paint job.
If you look at enough photos of grid battery projects, they blur into beige anonymity. But seeing the Megablock up close, the coat of white paint held more allure than it does from afar, more of a pearlescent sheen. A Tesla tour guide told me the shade was selected to maximize reflection of heat from the sun, thus reducing the energy needed to keep the batteries cool. The central chamber of the Megapack features a supercharged version of a Model Y heat pump, borrowed from colleagues on the automotive side of the company to chill liquid cooling streams that keep Megapack batteries and inverters safe.
Also taking in the sight was Tyler Norris, a Duke University doctoral student and leading researcher on how the U.S. might power its data-center boom. He noted that speed to market has become the premium that large energy customers are clamoring for.
“The U.S. is just in a major capacity crunch right now,” Norris said. “We’re going to need all sources of peaking capacity that we can get, and battery storage and the Megapack solutions are a critical option.”
The U.S. is expected to build 18 GW of batteries this year, per federal data, up from 13 GW last year, with California and Texas continuing to lead the way. Trump’s actions have injected deeper uncertainty into the market in 2025, with tariff fluctuations and new anti-China regulations. But the big tax-and-spending law passed by Republicans in July kept energy-storage tax incentives in place for years to come, even as it gutted wind and solar tax credits, meaning the outlook for storage is less muted than it is for the renewables it’s often paired with.
Meanwhile, data centers haven’t yet become big installers of on-site batteries, Norris added, but that could start to happen in the next couple of years. The typical four-hour duration of commercially available batteries today doesn’t lend itself to a round-the-clock power supply, so data-center developers are still figuring out how best to slot batteries into their energy portfolios.
Also viewing the big white box was Jesse Peltan, known for his spirited and data-rich defenses of clean energy on X, the Musk-owned social media platform.
“I think Megapack is the most underrated product that Tesla has by far, and I think Megablock is going to make it easier, cheaper, faster to interconnect Megapacks into the grid,” he said.
The cost of keeping California’s power grid up and running is skyrocketing, and in turn, so are households’ energy bills. Virtual power plants, which harness the combined power of lots of rooftop solar systems, home batteries, EVs, and smart-home appliances, can help — especially if utilities use them to relieve pressure at counterintuitive “sweet spots” on the grid.
So finds a new report that examines how the state’s utilities can spend less on new infrastructure by occasionally paying homes and businesses to reduce power use or to inject energy into the system — a concept known as “load flexibility.” Think tank GridLab published the study in collaboration with Kevala, a grid-focused data analytics startup.
One of the main reasons utilities’ expenses are rising is that the companies are putting more money toward their distribution grids — the poles, wires, and transformers that deliver power from electrical substations to homes.
Spending on distribution grids has grown rapidly in the past decade, and made up 44% of total utility spending in 2023, according to data from Lawrence Berkeley National Laboratory. Most of that cash is going toward replacing aging equipment and keeping up with booming demand for electricity.
The distribution grid is an even greater expense in California, according to Ric O’Connell, founding executive director of GridLab. Utilities there must invest heavily in wildfire-prevention measures, and the state’s ambitious decarbonization goals mean the power system needs to support the rapid electrification of homes and vehicles.
If California can defer upgrades to its distribution system, it can produce savings for customers, O’Connell said.
“That’s where the money is,” he said. All things being equal, “deferring the greatest number of highest-cost grid upgrades will save the most money.”
And according to GridLab’s new study, the best way to defer the most upgrades is to find those grid sweet spots — specifically, the areas with circuits, transformers, and substations that are least strained — and rapidly scale up virtual power plant programs to serve them.
Kevala, the startup that partnered with GridLab on the study, has a decent idea of where those sweet spots might be, based on its past analyses of distribution grids in California and nationwide.
The new study looks at the ideal way to deploy the 3.5 gigawatts of “load shift” capacity that California hopes to add to its grid by 2030.
For the research, Kevala compiled data on every feeder line, substation transformer, and substation of California’s three biggest utilities from today through 2030. It then ran three scenarios for using that 3.5 GW of load flexibility to relieve strain on that infrastructure: spreading the VPP effort equally across the grid, targeting the most overloaded parts of the grid first, and prioritizing the least overloaded parts.
That last technique was by far the most cost-effective, the analysis showed. Putting it into practice could reduce grid costs passed on to utility customers by a total of $13.7 billion through 2030 — about $10 billion more than the alternative approaches.
The reason? Taking on the least overloaded circuits first allows the same amount of load flexibility to defer new investments across a wider swath of the low-voltage grid, O’Connell said. The strategy also happens to target more urban areas, where much of the grid is buried underground, making it more expensive and difficult to upgrade.
That result came as something of a surprise.
“At first, we thought you’re going to start with the most heavily overloaded circuits and allocate flexibility to those, and then work your way down,” O’Connell said. “But we found you basically exhaust your flexibility on a handful of circuits — and you’re basically not saving a lot of money.” For those instances, “maybe it makes sense to spend real money on poles and wires.”
VPPs may also struggle to meet the challenge of deferring investments in the most strained parts of the grid, he noted. The history of these efforts appears to bear that out.
For more than a decade, utilities and regulators have been working on so-called “non-wires alternatives” projects — using batteries, energy efficiency, and grid-responsive devices to defer the need for big grid upgrades. Since 2014, California state policy has required regulators and utilities to work toward building these “distributed energy resources” — DERs for short — into their multibillion-dollar annual spending plans.
But beyond some showcase projects like New York utility Con Edison’s Brooklyn-Queens Demand Management initiative, relatively few proposals have moved past the planning phase. In California, despite programs launched over the past decade, “nothing’s really happened,” O’Connell said. Critics say the lack of progress is largely because utilities have proposed grid projects that DERs couldn’t possibly solve within the timeframes and cost restrictions provided.
On the other hand, “there are many circuits that are overloaded on a few hours of very hot days. I just need a little bit of DERs to solve that,” O’Connell said. “If we have a limited amount of valuable load flexibility, we should sprinkle a little bit of it across these lightly overloaded circuits.”
Targeting the least overloaded circuits could also minimize the risk of VPPs falling short of the job, he said. Slightly overloaded transformers and power lines can undergo overload conditions for short periods of time without blowing up or breaking down.
Larger-scale non-wires alternative projects like those that have been targeted in the past have a slimmer margin of error, he said. Utilities have traditionally demanded that any DERs being deployed to solve those grid constraints be made available for that purpose to the exclusion of any other use.
That’s a tough sell for customers of the companies putting VPPs together. Most consumers buy batteries for emergency backup power or to store surplus solar power — not to turn them over completely to utility control.
Customers willing to enroll their EV chargers, air conditioners, water heaters, and other appliances in flexibility programs would likely balk at the idea of being unable to use their devices when they really need to. Past VPP initiatives show that customers are far less likely to stick with them if they aren’t able to “opt out” of particular dispatches when circumstances demand it — say, when they need to charge their EV quickly after work to take their kid to soccer practice, or keep the house cool when elderly relatives are visiting.
With less-overloaded parts of the grid, by contrast, “maybe we can get the utilities a little bit more relaxed about it,” O’Connell said. “They’re always worried about, ‘What if the DERs don’t show up?’”
There’s a big catch when it comes to putting insights like these into action, however, said Kevala CEO Aram Shumavon. Utilities in California and elsewhere haven’t yet built VPPs and DERs into how they plan investments. That makes it much harder for the companies to consider them as options — which means they wind up choosing the traditional grid upgrade instead.
That’s the safer tried-and-true choice — and utilities, with their “extreme aversion to quantify risk, struggle with making innovative decisions,” he said. “But we’re spending a lot of time right now on what feels like baby steps, compared to how this market as a whole will need to function.”
It’s taken years for California utilities to start using the inherent flexibility of these technologies to help with grid operation and planning. But now, after some experimentation, they’re starting to prove that EV charging hubs, distributed solar installations, and utility-scale batteries can operate to fit within the hour-by-hour constraints on the grids they’re connected to. Similar efforts are now underway with customer-owned batteries and home energy control systems.
Still, VPP and DER programs are simply not expanding fast enough to meet California’s needs, Shumavon contends. “Once you move it into a program or procurement that requires a larger amount of situational awareness, we are woefully behind where we should be as an industry.”
Even getting the grid data needed for VPP providers to know where their solar-charged batteries or controllable household loads could do the most good has been a challenge. State legislators recently killed a bill provision that would have required California’s three major utilities to share data to inform how VPPs can reduce grid costs.
But Shumavon thinks that utilities in California are coming around to the need to move faster. The “non-wires alternatives” concept arose decades ago, when electricity demand was largely flat across most of the United States, and utilities had little incentive to support an alternative to investing more in their grids, which is how they earn guaranteed profits.
But that situation has radically changed in the past few years. The AI boom requires grids to handle gigawatts of new power, and utility rates are rising across the country. “The risk they’re facing is that they can’t do the rate increases, and they still have to deploy more capital, which has an upward pressure on rates,” Shumavon said. “That’s the point at which politicians get angry.”
O’Connell agreed that “utilities are much more interested in doing this now. They’re seeing rate pressure being a much bigger deal for them now. Anything they can do, it means that billions less in capital spend will show up.”
But the recent study by GridLab and Kevala “wasn’t going to get into how you design the program and how you pay them,” he said. “It’s more like, ‘You can do this — let’s figure it out.’”
AI data centers could flood the overtaxed U.S. power grid with demand and further drive up energy costs for consumers. Or, they could simply agree to use less electricity during the handful of hours per year when the grid is under the greatest stress, making it possible for tech companies to get the power they need without straining the system.
It sounds like an easy fix, but in reality it’s complicated to modulate the demand of a data center that can use as much power as a small city. That’s why it’s rarely done today. In fact, a Department of Energy report last year “identified no examples of grid-aware flexible operation at data centers” in the U.S., with one exception — Google.
Now, the tech giant is taking its flexibility efforts one step further and applying the concept to the machine learning operations that underpin its large language models, the technology driving the current boom in AI development.
This month, Michael Terrell, Google’s head of advanced energy, announced agreements with Indiana Michigan Power (I&M) and Tennessee Valley Authority (TVA), two utilities facing a lot of data center demand, that “represent the first time we’re delivering data center demand response by targeting machine learning workloads.”
Google’s new announcements are a really big deal, said Tyler Norris, a Duke University doctoral fellow and former solar developer and special adviser at the Department of Energy. That’s because they’re the first example of the kind of collaboration between data centers and utilities that needs to happen to keep costs from spiraling out of control.
Estimates of power demand from the AI race are all over the map and hard to trust, but at a minimum, most experts agree that data center demand for electricity outstrips supply. Consumer advocates and state lawmakers are increasingly worried that this dynamic is going to cause electricity rates to surge, as utilities incur the costs of building the power plants and grid infrastructure to serve data centers, and potentially push those costs onto customer bills.
That supply-demand imbalance is also a problem for firms like Google and its competitors, which are locked in a multibillion-dollar race to build the best possible AI system — a heated competition in which who gets electrons first could be a deciding factor.
Over the past five years or so, through its “carbon-intelligent computing” program, Google has been actively shifting nonurgent computing workloads — like processing YouTube videos — to prioritize clean power and avoid dirtier energy. It has also shifted computing load to help utilities manage grid emergencies.
Until recently, it hadn’t messed with power demand for machine learning. This sort of flexibility is new territory for utilities and grid operators, too: It’s not standard practice to allow large customers to come online only if they agree to curtail their power use, Norris pointed out.
“We’ve never planned loads this way, essentially for the entire history of the electric utility industry,” Norris said. But without this kind of approach, “it’s effectively impossible to see how some of these load forecasts can be met with purely physical infrastructure building.”
Flexible data centers like Google’s may have a significant advantage in getting those all-important electrons in the near term. A more rigid project may have to wait years to come online as grid infrastructure and power plants are built; a flexible data center, meanwhile, could be fast-tracked for interconnection using the grid capacity that’s already available.
The solution has its limitations, Terrell told Canary Media in an interview, but where it makes sense, it can be a powerful tool.
“We can’t do it everywhere. Some of our loads can’t be curtailed,” Terrell said. But where Google is able to do it, “there’s value to being able to secure capacity without having to wait for new infrastructure.”
The grid is crowded — but there’s plenty of room for data centers that can be flexible. That’s what Norris and a team of researchers at Duke University concluded in a February report.
The analysis found nearly 100 gigawatts of existing capacity on U.S. grids for data centers that can commit to 0.5% “annual average load curtailment.” That equates to being able to curtail less than half of their total power use for about two hours at a time during peak demand events that happen about 100 hours of the year.
There’s a simple explanation for this spare space: Grids and power plants are overbuilt to meet peak demands, or “worst-case conditions,” as Norris put it. But if data centers agree to avoid using power during those moments, it can obviate the need to expand the grid further to serve new, higher peaks.
It’s not a new idea in principle. Utilities have paid customers to reduce power use during peak demand for decades. But existing “demand response” programs tap current customers to help prevent emergencies for the grid as it is built today, Norris explained.
Google’s new deals with I&M and TVA, by contrast, are aimed at managing growing demand “in the form of a definitive long-term contract the utility can use for planning purposes,” he said. In other words, instead of using demand response to manage existing power needs, the utilities and Google are now wielding this approach to allow new users to come online. “That’s what sets it apart.”
Norris isn’t aware of any other data center-utility projects that are taking this longer-term planning view. In fact, “most data center developers won’t even release the nameplate megawatt scale of the facility,” he said — a feature of the highly competitive AI race.
In part because of this secrecy, it’s unclear how data center growth will play out in the real world. Most projections are based on speculative requests from developers seeking power in multiple locations for projects that may or may not end up being built.
But forecasts of data center growth generally indicate that they’re set to overwhelm the grid.
A December report from consultancy Grid Strategies found that five-year growth forecasts for U.S. utilities and grid operators have quintupled between 2022 and 2024, with data center hot spots such as Virginia, Georgia, Texas, and swaths of the Midwest particularly impacted. The past month has seen utilities in California, Colorado, New Jersey, and Pennsylvania report gigawatts of new data center requests.
That’s going to drive up utility rates, which are already rising due to a number of factors, including expensive investments in grid maintenance and expansion. While it can take time for the costs of accommodating new data centers to arrive on customers’ utility bills, the sheer scale of that expansion means that “the affordability concerns here are being put into stark focus,” Norris said.
Those future costs are starting to pile up.
Georgia Power won regulatory approval earlier this year to move ahead with plans for a controversial and unprecedentedly rapid buildout of power plants, almost entirely based on huge, uncertain forecasts of data center growth. The company has filed a more than $15 billion proposal revealing that much of its new infrastructure will be gas-fired. Virginia utility Dominion Energy is pushing for a similarly massive investment in fossil-fueled power to serve the world’s highest concentration of data centers. And Louisiana regulators last week approved utility Entergy’s plan to spend billions of dollars on gas-fired power plants and grid investments to serve a $10 billion data center from Meta.
In some states, customers are already paying more for energy because of data centers. PJM Interconnection, the grid operator serving Washington, D.C., and 13 states from Illinois to Virginia, has seen prices for capacity to maintain its grid skyrocket in the past year. PJM’s inability to bring new generation online is a chief culprit. But its ballooning future demand, another important part of the equation, is “almost entirely due to existing and projected data center load additions,” according to PJM’s independent market monitor.
Norris argued that utilities, regulators, and grid operators must start demanding that would-be data centers commit to some level of flexibility to receive grid interconnection. “If you’re planning for all new loads to be inflexible and serving them with firm at all hours of the year, that’s going to be extraordinarily expensive,” he said.
While data centers could build their own power supplies, “we also don’t want them to be running the backup [diesel generators] 200 hours a year to get online faster,” Norris said. Reliance on polluting on-site generators is already a problem for communities in Memphis, Tennessee, which are protesting the use of hundreds of megawatts of gas-fired turbines at a data center built by Elon Musk’s xAI.
Google’s approach of managing its data center power use to reduce carbon emissions represents a much cleaner alternative. “The capabilities we developed to do load shifting for carbon, we use the same capabilities to do demand response,” Terrell said.
Google’s agreement with TVA applies to existing data centers “north of Nashville and in North Alabama. We need to grow, but [TVA was] not in a position to serve us” in the short term, Terrell said. TVA has not released details of its agreement with Google, and Terrell declined to provide more specifics.
Google’s agreement with I&M centers on the tech giant’s $2 billion data center in Fort Wayne, Indiana, which started operations late last year but expects to ramp up its power needs over time, Terrell said.
In broad terms, the plan states that Google will commit both to restraining its use of power at its Fort Wayne data center during critical hours and to transferring credits for a portion of carbon-free energy it has contracted for in the region to I&M to help it meet its capacity requirements. “We need to be bringing new resources onto the system,” Terrell said.
Many of the details of I&M and Google’s plan filed with state regulators last month have been redacted for confidentiality reasons, which has raised concerns from consumer advocates. But the proposal does appear to align with new regulations aimed at controlling data center costs.
Earlier this year, Indiana utility regulators approved a settlement between I&M, data center developers, and consumer advocates that set new requirements for “large loads” — namely data centers — to commit to covering a significant portion of the costs they incur. The goal is to avoid forcing customers to pay higher bills for decades due to investments made to meet data centers’ needs.
More such rules are coming. Ohio regulators in July approved a similar settlement agreement, and a broader energy law passed in Texas this year will require large data centers to reduce power use during grid emergencies. PJM launched a fast-tracked effort this month to create new large-load interconnection rules, and Southwest Power Pool, a grid operator serving 14 Midwest and Great Plains states, plans to streamline connection for data centers and other big facilities that can commit to flexible operations or to providing their own power.
Data center operators have traditionally shied away from altering operations to save or shift energy, said Astrid Atkinson, CEO of grid-software startup Camus Energy. That makes sense, given the high value of the computing they’re doing — something Atkinson dealt with as former lead of the Google teams that maintain reliable computing at data centers providing web services and social media.
But data centers training AI models have more flexibility than those providing time-sensitive or business-critical services like processing financial transactions, she said. “They’ll be running large training jobs that use up a lot of their nameplate capacity for a period of time, but then they may sit idle for a long period of time,” she said. “You can potentially move them around in time a little bit.”
Camus Energy is already working on projects to enable flexible EV charging, but much of its recent work with utilities centers on managing new data centers, she said. “If it makes the difference in being able to build or expand a site now, or having to wait five years, that makes it worth doing.”
Indeed, some other utilities and data center operators are exploring grid flexibility. The Electric Power Research Institute, a largely utility-funded nonprofit, last year launched its DCFlex initiative, a collaboration that’s testing flexible computing at sites including an Oracle data center in Arizona and a Google data center in North Carolina.
And although most regulated utilities lack incentives to work on flexible interconnection — they earn guaranteed profits based on how much money they spend on grids and new power plants — Norris thinks the surge in demand could change their calculus. There’s only so much cost regulated utilities can put on their customers before regulators are forced to intervene, and the AI boom is testing those limits.
“The more they’re looking at big upgrades and infrastructure investments, the more they need to balance that with affordability,” Norris said.
Terrell said that Google is “having more conversations with utilities” about data center flexibility, though he declined to provide further details. “It’s an advantage for our business to be able to go to utilities and offer this.”
For decades, electrical engineers have dreamed of a device that can seamlessly connect solar panels, battery systems, and on-site generators to high-powered equipment like EV chargers or data center servers, without loads of expensive hardware to make it all work together.
Now, these devices, called solid-state transformers, are actually starting to hit the market — and they couldn’t be coming at a more opportune time.
That’s because the technology could be key to dealing with the torrent of power demand from data centers, factories, and electric-vehicle charging hubs that threatens to overwhelm the grid and cause utilities to burn more planet-warming fossil fuels.
Right now, these large electricity customers are clamoring for more power than the U.S. grid can easily supply. In theory, this problem could be solved by allowing them to install their own solar arrays, batteries, and generators on site — ideally as a microgrid — but that seemingly simple solution is actually complicated and costly to execute.
Every solar array, battery, fuel cell, generator, or other source of on-site power requires multiple pieces of equipment — electrical protection gear, isolation transformers, step-up and step-down transformers, power converters — to safely turn direct current into alternating current or vice versa, and to raise or lower voltages to match the needs of different loads within a building.
Solid-state transformers can do all that from a single device, controlling electricity as nimbly as routers control the flow of data. That’s particularly valuable when it comes to managing equipment with high power needs, like EV chargers, or with extremely sensitive requirements for power quality, like the server racks populating data centers.
So says Haroon Inam, CEO and cofounder of DG Matrix, one of a handful of companies starting to get solid-state transformers into real-world applications. DG Matrix raised $20 million in March and is building a factory in North Carolina, set to open late this year, that will be capable of producing up to 1,000 units annually, he said.
“We’re hitting the massive underserved commercial and industrial microgrid market,” he said. “People haven’t done it because it costs so damn much to build individual snowflake microgrids.”
DG Matrix is not the only firm working on this. Heron Power, a startup founded by Tesla alum Drew Baglino, has raised $43 million in funding with the goal of building its first solid-state transformers in 2027. Amperesand raised $12.5 million last year to continue developing solid-state transformers being tested on Singapore’s power grid.
Major electronics companies are interested. Electrical equipment giant Eaton last month agreed to acquire Resilient Power Systems, which raised $5 million in 2021 to build and deploy its power-conversion devices for EV charging hubs and other energy-hungry settings. Eaton will spend $55 million on the company on closing; additional payments based on Resilient Power’s financial and technological performance in the coming years could total another $95 million.
“People have been working at this technology for well over a decade,” said Aidan Graham, senior vice president and general manager of Eaton’s critical power solutions business. But now, following several key engineering advances, the technology may finally be ready for primetime — and utilities and others are starting to test it out.
Eaton has been working on solid-state transformers for years. The company isn’t saying how it intends to scale up manufacturing and deployment of Resilient Power’s technology. But “there are a couple of branches we’re chasing,” Graham said, including EV charging and integrating batteries into data centers and other critical environments, “where people‘s lives are on the line, or a lot of money is on the line, if the power goes out for even a fraction of a second.”
Michael Wood III, DG Matrix’s chief of staff, said the company is testing its devices with companies including electrical-equipment manufacturing giant ABB, North Carolina-based utility Duke Energy, and PowerSecure, a major microgrid and data-center power system developer owned by utility Southern Co.
“The best way to get the next gigawatt of energy is to build distributed systems,” Wood said. “Today, you need all of this gear to make those projects work. DG Matrix eliminates all that balance of systems and boils it down to a single system.”
Using a DG Matrix solid-state transformer can cost half as much as using the standard mix of multiple technologies to connect the components of a typical on-site microgrid, Inam said. It also makes it a lot simpler to quickly mix and match devices or to change up the configuration of systems at data centers, EV charging hubs, and other potential microgrid sites.
So if solid-state transformers are such a useful technology, why are they just now getting into the field?
There are good reasons why it’s taken so long, said Vlatko Vlatkovic, a veteran of General Electric’s industrial electrification business and a partner at DG Matrix investor Clean Energy Ventures, who joined the startup’s board of directors this year.
Much of the power grid relies on electromechanical devices that operate in relatively simple ways that haven’t changed much in a century. Despite recent advances that have enabled things like solar inverters or electric-vehicle drivetrains, the same kind of semiconductors that make modern computing possible have yet to be applied widely to the power grid.
“It’s always been a big challenge to move that industry towards using more power electronics,” Vlatkovic said, particularly at the higher voltages of electricity on the grid. Until relatively recently, the underlying technology “wasn’t big enough, wasn’t reliable enough. There were technical issues.”
Similar challenges have dogged solid-state transformers in higher-voltage industrial applications, said Neal Dikeman, a partner at Resilient Power investor Energy Transition Ventures. Consistent advances in silicon carbide semiconductors have helped, as have strides in the computing ability required to make them effective at power conversion, he said. “But that doesn’t make it easy.”
Inam, who served as chief technology officer at grid power-controls provider Smart Wires before joining DG Matrix in 2023, noted several key challenges that the startup had to solve to get to this point.
Dissipating the heat created by converting alternating current and direct current at high voltages is tricky, for one. So is dealing with “electromagnetic noise,” or interference caused by that same high-frequency electrical switching. “If you don’t understand how to critically mitigate that noise, it gets into everything. It causes overheating, blow-ups, and misperformance,” Inam said.
Solving those challenges has its rewards, however. “We’re at the point where the technology is mature enough and good enough so that we can introduce reliable devices,” Vlatkovic said.
The timing couldn’t be better.
“Everything’s being electrified, from cars to industry to housing,” Vlatkovic said. “If you look at the projections for what the grid needs to deliver over the next 10 to 20 years, at a minimum we have to double the capacity of the grid. Some projections say we need to triple what we have.”
Meeting power demand from data centers is a particularly big opportunity, Inam said.
Tech giants’ AI ambitions are taxing the grid capacity of utilities in data center hot spots like Virginia, Georgia, and Texas. That’s led data center developers to explore ways to reduce the stress they put on grids — including the potential for generators and batteries built nearby or on site.
“The three big problems are speed to power — customers can’t get power fast enough — cost of power, and the ability to aggregate multiple resources to reach flexibility,” Inam said. “We talk to enterprise customers with hundreds or thousands of sites. Their biggest challenge is having to design every single one from scratch. They’re looking for a turnkey solution to the challenge of not having to deploy one, but having to deploy 1,000.”
Solid-state transformers can help meet those needs, Vlatkovic said. “You go from complex installations and multiple companies to one company doing everything.”
Packing more capabilities into a smaller “power-dense” package also saves valuable space in tight environments like data centers and EV charging sites, Eaton’s Graham said. And solid-state transformers can be made en masse in factories, reducing the cost and time spent on electrical labor on job sites. “You’ve pulled that back into a controlled manufacturing environment,” Graham said.
Plus, having a single device that can perform multiple tasks simplifies engineering needs, Dikeman said.
“If you’re using off-the-shelf components and designing a complex system, the mismatching of” different devices that don’t perfectly match the needs of the system “drives up costs and drives down efficiency,” he said. “You can get around that by building custom stuff — but that’s more expensive and more risky. When you get to solar and storage and data centers and people who need to go fast and need things that are reliable and cheap, all of that breaks down.”
All of these potential benefits have led PowerSecure, the microgrid developer, to launch pilots of at least two solid-state transformer technologies, including its tests with DG Matrix, said Joaquin Aguerre, the company’s director of strategic portfolio development. “We’re trying to be in front on this technology.”
PowerSecure has designed and installed more than 2.4 gigawatts of microgrid capacity for customers ranging from big-box retailers and hospitals to utilities and data centers. It’s particularly interested in solid-state transformers to integrate power-efficient “hybrid microgrids” that combine “solar, energy storage, natural-gas generators, fuel cells, EV charging, you name it,” Aguerre said.
“There is starting to be a real market need,” he said. At the same time, “the majority of these companies are still in the early stages. … The next logical step is doing proper pilot programs, to see real customer use cases at a smaller scale” and to test the durability and reliability of the technologies in question.
After all, whatever their drawbacks compared to cutting-edge power electronics, traditional transformers “don’t fail that often,” Aguerre pointed out. “Everyone’s going to expect the same reliability for whatever solid-state transformer they’re looking at.”
A clarification was made on Aug. 19, 2025: This story originally stated that Eaton will pay $55 million to acquire Resilient Power. The piece has been updated to clarify that the deal also includes additional payments contingent on Resilient Power’s performance over the next couple of years.
The Trump administration is expected to use a controversial Energy Department report to justify keeping costly fossil-fueled power plants online past their retirement dates. But nine state attorneys general and several clean-energy industry groups are demanding the agency fix the document’s heavily criticized methodology.
The report, which found that the country will face a hundredfold increase in grid blackout risks absent extraordinary federal intervention, was blasted by experts upon its release in July. The DOE’s analysis ignores hundreds of gigawatts of new energy resources likely to come online in the coming years, the vast majority of it solar, batteries, and wind power, and it overstates power plant closures expected over the next five years.
“DOE’s assumptions unreasonably presume that the market, grid operators, and state regulators will take no action in the next five years to address load growth or reliability issues, and that no alternative other than preserving aging coal and gas power plants will ensure grid reliability,” the state attorneys general wrote in their joint request for rehearing filed earlier this month.
The DOE hasn’t yet cited the analysis to support any stay-open orders. But the attorneys general of Arizona, Colorado, Connecticut, Illinois, Maryland, Michigan, Minnesota, New York, and Washington wrote that an April executive order from President Donald Trump and “subsequent statements by DOE make clear that the report will be used to justify Section 202(c) orders going forward.”
Already, before the report was issued, the DOE had used Section 202(c) of the Federal Power Act to order the J.H. Campbell coal plant in Michigan and the Eddystone oil- and gas-burning plant in Pennsylvania to keep running on the eve of their planned closures, at a steep cost to consumers. Forcing more such plants to stay open would drive up electricity costs further and scramble long-running plans from utilities and energy developers to build resources to replace the shuttered facilities.
Doing so would also be illegal, the attorneys general state. “The Report is arbitrary, capricious, contrary to law, and unsupported by substantial evidence in violation of the Administrative Procedure Act and Federal Power Act because it suffers from numerous analytical, mathematical, and empirical flaws.”
The DOE wrote the July report to comply with April’s executive order that seeks to give the agency unilateral authority to force power plants to keep running, even when utilities, state regulators, grid operators, and other experts say it’s safe — and economically prudent — to close them down. The DOE did not respond to Canary Media’s requests for comment.
The report’s flaws were reiterated in a separate rehearing request filed this month by the American Clean Power Association, American Council on Renewable Energy, and Advanced Energy United. The trade groups argue that the DOE’s analysis “fails to take account of (or simply mischaracterizes) major developments that will affect resource adequacy in the next half-decade and beyond,” including “the pace of new resource development, the retirement of existing resources, and the well-established regulatory and market mechanisms that connect these threads.”
In a webinar earlier this month, executives of these trade groups said they also fear that the DOE will use these flawed assumptions to justify ordering fossil-fueled power plants across the country to keep running.
“DOE’s analysis takes a series of outlier assumptions and applies them all in one study as the only future scenario, and the result is that we’re getting predictions of blackouts,” said Caitlin Marquis, managing director at Advanced Energy United. “When it’s applied as directed in the executive order to resource-retention decisions, there will be real-world consequences to those actions.”
Indeed, forcing fossil-fueled plants to stay open could “inflict significant harm on our states,” the attorneys general wrote in their rehearing request. In Colorado and Washington state, coal plants set to close in 2025 could be forced to keep running, despite their closure plans being “thoroughly vetted by state and regional authorities and approved only following an extensive examination of cost considerations and reliability impacts.”
States that are part of regional power markets must still share in the expenses of power plants ordered to stay open, as is the case for the J.H. Campbell plant. Keeping that facility running between late May and the end of June cost $29 million, and the total could surpass $100 million by the expiration of the DOE’s 90-day stay-open order this week. That price tag is being spread across the states that are part of the Midcontinent Independent System Operator’s north and central regions, which include Michigan.
The financial toll could rise dramatically if the DOE uses its authority under Section 202(c) to prevent any fossil-fuel plants nationwide from closing on schedule in the coming years. An analysis from consultancy Grid Strategies found those costs could add up to $3 billion to nearly $6 billion per year by 2028.
This month’s filings aren’t the first challenges to the DOE’s use of Section 202(c) authority.
State regulators and environmental groups filed rehearing requests to the DOE’s stay-open orders in Michigan and Pennsylvania, on the grounds that they violate the agency’s legitimate use of Section 202(c) to deal with near-term emergencies. The DOE did not respond to those requests, which prompted Michigan’s Attorney General Dana Nessel and environmental organizations, including Earthjustice and Sierra Club, to file petitions for review asking the federal D.C. Circuit Court of Appeals to open a case allowing the groups to fight the DOE’s decisions in court. Those petitions for review are pending.
It’s not clear if the agency will respond to these new challenges either, which could prompt lawsuits from the states or the industry. The offices of the nine attorneys general that are seeking a rehearing on the report did not immediately return Canary Media’s requests for comment.
China’s dominance of the battery supply chain is uncontested. Many U.S. storage companies have tried to catch up by replicating the technologies already in mass production there. But a smaller cohort is taking a different tack: building factories for next-generation batteries that could give American manufacturers more of a competitive edge.
Peak Energy is one of the newest members of that cohort. The startup, which appeared on the scene in 2023, took a big step this summer when it shipped its first sodium-based grid-battery system for installation in the field. The 875-kilowatt/3.5-megawatt-hour battery is now being completed in Watkins, Colorado, at a testing facility known as the Solar Technology Acceleration Center.
In fairness, the battery cells were imported from China, but Peak designed and built a new enclosure for them in Burlingame, California. Since the sodium batteries are especially rugged, Peak could forgo the temperature-control equipment needed for the current favorite chemistry for grid storage, lithium ferrous phosphate (LFP). If this first installation works well and the cost savings are as consequential as promised, Peak plans to build U.S. manufacturing for the whole package, cells and all.
The installation is a rare bright spot as the storage industry at large grapples with the impacts of Trump administration energy policy. President Donald Trump’s unpredictably shifting tariffs on China have raised the costs of imported batteries and made it hard to plan. The White House’s signature budget law ripped up some — but not all — tax credits meant to support domestic manufacturing of batteries, and added dense new bureaucratic requirements around components from China. New investment in domestic clean-energy manufacturing has plummeted since Trump took office.
But the power sector still wants to build grid batteries at record pace, especially as supersized data centers clamor for electricity supply as soon as possible.
Upstart battery-makers often jockey over how much energy density they can pack into their cells, or how they can reduce the fire risks that follow from squeezing so much energy into a tight footprint. Peak Energy brags more about what its technology doesn’t need: heavy-duty climate control.
“If you think about it, an LFP [energy storage system] is essentially a giant refrigerator that has to operate flawlessly for 20 years in the desert,” said Cameron Dales, Peak’s chief commercial officer and cofounder. That’s because that particular chemistry ideally needs to stay within a few degrees of 25 degrees Celsius (77 degrees Fahrenheit) to preserve its useful life; serious deviations from that safe zone could lead to declining performance or even dangerous failures. A handful of dramatic battery fires has already inspired community pushback against storage plants, making safety a crucial part of the industry’s social license. Indeed, this week U.S. Environmental Protection Agency Administrator Lee Zeldin pledged to support communities resisting nearby battery installations.
The sodium-ion cells that Peak favors — technically called sodium iron pyrophosphate or NFPP — can withstand a much broader range of temperatures, from minus 20 degrees C (minus 4 degrees F) to 45 degrees C (113 degrees F). Peak’s engineers thus dispensed with the usual battery-cooling systems, relying instead on what Dales calls “clever engineering” around how the cells fit into the broader package. “There’s no moving parts, no fans, no liquids, no pumps, no nothing,” he said. The container does include a solid-state heater to ensure the cells never get too cold to charge.
This saves money by reducing the cost of materials and cutting auxiliary power usage up to 90% over the life of the project. But axing the conventional safety equipment brings one more major benefit, because that hardware has paradoxically caused several of the recent high-profile grid-battery fires (by, for example, erroneously spraying water on live batteries, which can make a fire where there wasn’t one).
Plenty of cleantech startups have pitched themselves as safer alternatives to dominant strains of lithium-ion batteries, only to be crushed mercilessly by the lithium-ion manufacturing juggernaut. Overwhelming scale and a wealth of industrial expertise keep pushing mainstream batteries to lower prices and superior performance. However, the up-front costs of the batteries themselves are now just a small piece of the overall bill.
“What has not really been addressed is the construction and installation of a project, and then, even more importantly, the long-term operating costs associated with running that power plant,” Dales said.
According to Dales’ calculations, the energy savings from the passive cooling of Peak Energy’s battery enclosure over a period of 20 years more than cover the initial cost of the battery cells. That’s one way to lure customers to a type of battery they haven’t seen before.
“How can a startup, who’s just getting up to speed and their costs are high and volume is not there yet, compete and win on a project like that?” he said. “It’s because these project economics are so good that even today, we can win on cost relative to … a Chinese LFP system.”
Flipping the switch on the Colorado project is just the start. Then Peak Energy needs to find paying customers interested in much bigger versions of the technology. But the startup has an innovative plan for that next step.
The founders of many battery startups focus on a technology that they find interesting (maybe they chose it for their doctoral research years ago), and then at some point have to convince customers to buy it. This typically leads to what Dales identified as “a classic failure mode, to get piloted to death.” The eager startup spends its precious time developing insignificant yet money-losing pilot installations with lukewarm customers, who try it for a few years and decline to make a follow-up purchase. Then the startup runs out of cash and collapses.
Peak Energy’s founders decided on a different strategy: develop a product in conversation with prospective customers, so they actually want to buy it when it’s ready.
The Colorado project, paid for by Peak, will be scrutinized by a consortium of nine utilities and independent power producers, who have signed on to receive exclusive performance data. If the project meets agreed-upon metrics, these companies will buy Peak’s product for their own use.
“If we do what we say we’re going to do, and the economics are what we think they are, then you should sign up for doing a real project, because it actually makes sense for you,” Dales explained. “That’s how these companies have entered the program, and now we’re in the ‘proof is in the pudding’ phase.”
Some of those consortium members have requested batteries for demonstration projects in 2026, in the storage range of 10 MWh to 50 MWh, Dales said. One large power developer is working on a 2027 project that would deploy nearly one gigawatt-hour of Peak’s sodium batteries to support a hyperscaler data center.
The path from initial installation to giga-scale projects always takes longer than battery startups initially pledge. In fact, only lithium-ion batteries have crossed that threshold, while more unusual variants languish in the minor leagues.
But Peak doesn’t have to invent the core technology — it’s piggybacking off an emerging field of China’s battery industry — and it’s coming to market at a time of propulsive growth in grid storage demand. Its task might not be quite so daunting as it has been for other battery innovators.
California’s premier “virtual power plant” program is already reducing the state’s reliance on polluting, costly fossil-fueled power plants. And that’s just the start of what the scattered network of solar and batteries could do to stymie rising utility costs — if the state Legislature can stave off funding cuts to the program, that is.
So finds a new analysis from consultancy The Brattle Group on the potential of the statewide Demand Side Grid Support (DSGS) program to help California’s stressed-out grid keep up with growing electricity demand. The program pays households and businesses that already own solar panels and batteries to send their stored-up clean power back to the grid during times of peak demand, like hot summer evenings.
Continuing the program’s payments to those customers to make their stored energy available could save all California utility customers anywhere from $28 million to $206 million over the next four years, the report found.
The findings come as state lawmakers attempt to rescue the DSGS program from a new round of funding cuts. Last year, California lawmakers slashed DSGS spending to deal with an unexpected budget shortfall. The situation is still troubled this year, and Democratic Gov. Gavin Newsom has proposed defunding the program further, leaving it little money to pay participants beyond this year.
But the program could regain its financial footing if newly introduced legislation becomes law.
This week, California Assemblymember Jacqui Irwin, a Democrat, released draft legislation that would allocate money to DSGS from the state’s much-contested Greenhouse Gas Reduction Fund, which is supported by payments from polluting companies. That draft legislation calls for depositing 5% of revenue collected by electric utilities for that fund into a new account to finance DSGS from 2026 to 2034. Lawmakers don’t have much time to move the proposal forward, with the state’s legislative session ending Sept. 12.
Saving the program would be a win for reducing the state’s sky-high utility costs, according to Ryan Hledik, a principal at Brattle and coauthor of the report. “It’s cheaper to pay customers to provide grid resources from technology they’ve already adopted than it is to go invest capital in new stuff,” he said, including the fossil-fueled generators now used to meet peak grid needs.
California has already committed billions of dollars on emergency backup generators and on keeping aging fossil-gas-fired power plants open past their planned closure dates, he noted. The high end of the savings DSGS could provide is based on the assumption that it “would be a substitute for spending money on more expensive emergency resources,” he said.
At the same time, DSGS could also bring down the “resource adequacy” payments shelled out by California utilities, community choice aggregators, and other power providers to secure enough grid resources to meet peak demand in future years. Those costs have been rising in California, though not as drastically as they have in other parts of the country.
Since its launch in 2023, the battery program Brattle analyzed, which is one of the four options for customers to participate in DSGS, has grown to a collective 700 megawatts of capacity. The report forecasts the program could nearly double its current capacity to reach 1.3 gigawatts by 2028, covering roughly half the total residential distributed-battery capacity expected to be online in the state by then.
That won’t happen without state funding for the program, however — and though some state lawmakers are attempting to save DSGS’s funding, it remains unclear if the money will be there for future years.
If Irwin’s proposed provision becomes law, it would supply roughly $70 million to $90 million per year to DSGS over the next five years, said Brad Heavner, executive director of the California Solar and Storage Association. DSGS needs at least $75 million this year to operate in 2026, according to a letter sent to California lawmakers on Tuesday by 35 companies, trade groups, and advocacy organizations active in solar, batteries, on-site generators, and demand response, including Heavner’s group.
The amount of funding dedicated under the proposed legislation “won’t be enough for all the program activity we expect — but it will be enough to have a core program,” he said.
DSGS’s cost-effectiveness, demonstrated by the Brattle analysis, should give lawmakers confidence that the money isn’t being wasted, Heavner said. “It’s great that the Brattle study finds there’s a two-for-one benefit — every dollar spent here saves two dollars” for utility customers across the state, he said.
Brattle’s research was funded by Sunrun and Tesla, two companies with longtime programs that sign up customers to make their excess battery capacity available for grid services. Both firms benefit from initiatives that boost the value of the rooftop solar and battery systems they sell to households in California and beyond.
But the study also matches broader research on how virtual power plants can reduce blackout risks and electricity price spikes on U.S. grids. VPPs are collections of homes and businesses with smart thermostats, grid-responsive EV chargers, water heaters, and other appliances that can reduce how much power they’re using, as well as rooftop solar-charged batteries or generators that can push power back to the grid as needed.
Under the Biden administration, the U.S. Department of Energy found that the hundreds of billions of dollars that consumers spend on EVs, rooftop solar systems, batteries, smart thermostats, and other appliances could provide 80 to 160 gigawatts of VPP capacity by 2030, enough to meet 10% to 20% of U.S. peak grid needs and save about $10 billion in annual utility costs. (The Trump administration has removed this DOE report from the internet, but archived versions are available.)
VPPs also pass the eye test: They’ve helped avoid blackouts in Puerto Rico, New England, and California this summer. States including Colorado and Virginia have passed laws or created regulations requiring utilities to expand VPPs.
DSGS, for its part, has “scaled in a way that folks can no longer poke holes in its reliability,” said Lauren Nevitt, Sunrun’s senior director of public policy. Sunrun has dispatched hundreds of megawatts from its customers’ batteries in California so far this summer, all during hours of the evening when wholesale electricity prices spike above $200 per megawatt-hour.
In a two-hour experiment last month, Sunrun and Tesla dispatched 535 megawatts of battery power to the grid in what utility Pacific Gas & Electric called “the largest test of its kind ever done in California — and maybe the world.”
Lining up a steady source of funding for years to come would give these participating companies confidence that their investments in DSGS won’t be left stranded by future budget cuts, Heavner said — and encourage even more investment going forward.
Pressure to curb energy costs is particularly acute in California, where residential customers of the state’s three major utilities now pay roughly twice the national average for their power and where rates rose 47% from 2019 to 2023.
It is also among the best-positioned states to take advantage of VPPs to rein in those costs. California leads the country in rooftop solar, backup battery, and EV adoption, and a 2024 Brattle analysis found that VPPs could provide more than 15% of the state’s peak grid demand by 2035, delivering $550 million in annual utility customer savings.
DSGS is only one of a number of VPP options available in California. But advocates say it’s by far the most successful in a state that’s seen mixed progress on VPPs to date. In the past five years, stop-and-start policies from the California Public Utilities Commission have reduced overall capacity from demand-response programs that pay utility customers to turn down their electricity use to relieve grid stress.
DSGS, which is run by the California Energy Commission, has grown rapidly due to a combination of factors, said Edson Perez, who leads California legislative and political engagement for clean-energy trade group Advanced Energy United. It’s available to residents across the state, rather than being limited to individual utility territories and programs. It also has relatively simple enrollment and participation rules compared to many other programs, he said.
It can be tricky to quantify the costs and benefits of these kinds of programs compared to traditional utility investments in power plants or large-scale solar and battery systems. But Brattle’s new report is the “first analysis of what its value is out in the field,” he said, and the results show “it’s very cost-effective.”
Solar-charged batteries are also much less polluting than the state’s other emergency grid-relief resources, he said. DSGS is one of a set of emergency programs launched after California experienced rolling blackouts during summer heat waves in 2020 and more heat-wave-driven grid emergencies in 2022.
But most of the billions of dollars in emergency funding have gone to fossil-fueled generators. California had spent about $443 million on state-managed generators that burn fossil gas or diesel fuel as of December 2024, and has committed about $1.2 billion to keep fossil-gas-fired “peaker” plants in Southern California open until 2026, well past their scheduled 2020 closure date.
“We’re in a statewide affordability crisis,” Perez said. “Leveraging existing resources out there drives down costs for everyone.”
Electricity costs are going up in the U.S. — and the Trump administration’s attempts to choke off clean energy development are only going to make matters worse.
The average price of electricity for residential consumers is set to hit 17 cents per kilowatt-hour this year and could climb to 18 cents per kilowatt-hour in 2026, per a new report from the U.S. Energy Information Administration.
Electricity prices are rising at more than twice the rate of inflation. Just five years ago, in 2020, average U.S. power prices were only 13.15 cents per kilowatt-hour — 23% lower than they are today.
The difference may seem small, but even one additional cent would tack on roughly $108 to the average U.S. home’s expenses each year. It’s taking a toll on people’s wallets: A survey conducted this spring found that three in four Americans said they’re worried about rising utility bills.
Republican leaders — most recently U.S. Energy Secretary Chris Wright — have tried to blame the trend on the large amounts of clean energy hitting the grid, but that’s not the problem. Solar, wind, and batteries are the cheapest form of power, and a 2024 report from research group Energy Innovation found no correlation between renewable energy adoption and utility rate increases.
Numerous reports and studies reveal that the core drivers of rising prices include an aging distribution grid that requires expensive repairs, and damage to the system from the wildfires and storms exacerbated by climate change. Then there’s the volatile price of natural gas, which produces about 40% of U.S. electricity. Skyrocketing demand for power is also increasingly a factor, as people electrify their homes, businesses, and cars, and in particular as data-center developers snap up as much energy as they can to support their AI ambitions.
In January, President Donald Trump took office promising a great many things — including to make energy more affordable. But since then, household electric bills have risen another 10%, and the policies he’s enacted are set to exacerbate the problems at hand.
Due to the GOP megalaw signed by Trump last month, the U.S. could install as much as 62% less clean energy over the next decade, per Rhodium Group estimates. That’s a huge deal: It’s expected that 93% of the new electricity capacity built this year will be solar, wind, or batteries.
If renewable energy construction slows at the same time data centers and consumers require more power, it will create a clear dynamic of too much demand and not enough supply. The result will be even higher energy bills for Americans, Rhodium and others forecast — the exact opposite of Trump’s grand vow to rein in costs.
If your power bills are getting higher and higher, you’re not alone. That’s probably little comfort, but here’s some proof anyway: Utilities requested or were granted a total of $29 billion in rate increases in the first half of 2025, according to a study from advocacy group PowerLines. That’s more than double the total in the same period last year.
The biggest reason for these rising prices stems from the piece of the grid you can see from your window, as Heatmap reports. Utility poles and wires, also known as the distribution grid, shuttle power from high-voltage transmission infrastructure into homes and businesses. Over the last few years, building and maintaining these lines has become the biggest source of costs that utilities recoup via power bills, according to a December report from the Lawrence Berkeley National Lab.
Natural disasters are also driving up expenses as they force utilities to repair and harden their grid for future weather events. California utilities, for instance, have to rebuild after wildfires and in some cases are spending even more money to underground lines. In the Southeast, utilities routinely look to raise rates to cover post-hurricane restoration costs.
Then there’s the fact that natural gas remains the U.S.’s dominant energy source and that prices for that fuel remain higher than they were over much of the last two years.
Now for the second big question: Will things get better anytime soon? Probably not, for a few reasons.
For starters, power demand is on the rise, stemming in large part from the construction of energy-hungry data centers. Tech giants plan to keep building facilities to run their AI operations, and how they’re powered — and how that demand is managed — could end up making everyone else’s electricity more expensive.
That demand could be largely satiated by new solar and wind farms, which are typically quicker and cheaper to stand up than fossil-fueled and especially nuclear power plants. But the One Big Beautiful Bill Act that Republicans passed in July will soon wipe out federal tax credits that incentivized clean energy construction.
Instead, the Trump administration is pushing to keep aging fossil-fuel power plants online past their retirement dates — a mission that could end up costing utility customers as much as $6 billion each year by the end of President Donald Trump’s term. A federal order that kept a Michigan coal plant open past its planned closure cost its operator $29 million in its first five weeks, and just this week the Energy Department reupped the facility’s extension until November.
Treasury rules tighten access to clean energy tax credits
The U.S. Treasury Department has released guidance that will make it harder to access wind and solar tax credits before their ultimate expiration, Canary Media’s Jeff St. John reports. The One Big Beautiful Bill Act gives wind and solar developers two options to tap the credits: They must either put their project in service by the end of 2027 or begin construction by July 2026. The Treasury’s new guidance narrows the federal government’s longstanding definition of what marks the start of construction.
Still, things could’ve been a lot worse, experts told Jeff — deadlines to finish work could’ve been accelerated, for example. And with these rules, developers have the clarity they’ve been waiting for to make decisions and get building.
USDA pulls support from solar, wind on farmland
Federal assistance for solar and wind power on farmland is fading. On Tuesday, the U.S. Agriculture Department announced that it will “no longer fund taxpayer dollars for solar panels on productive farmland or allow solar panels manufactured by foreign adversaries to be used in USDA projects.” It will also render wind and solar projects ineligible for the agency’s Business and Industry loan program, and bar Rural Energy for America Program loans from being used for ground-mounted solar projects larger than 50 kW.
The Trump administration has already taken multiple shots at REAP, Canary Media’s Kari Lydersen reported in July, freezing nearly $1 billion in funding for farmers and closing a window for new applications before it even opened.
“Come to America and lose $1B”: Foreign offshore wind developers have faced steep financial losses over the past few years, and they’ve only intensified under the Trump administration’s anti-wind policies. (Canary Media)
Polluting the post: Republican U.S. senators move to strip federal funding for the U.S. Postal Service’s transition to an EV fleet to save taxpayer money, though industry observers say the move would have the opposite effect. (Associated Press)
Steel’s dangerous warning: Last week’s fatal explosion at Pennsylvania’s Clairton Coke Works underscores the urgent need to decarbonize the coal-reliant steelmaking industry. (Canary Media)
Solar still rises: The Energy Information Administration estimates the U.S. will add 33 gigawatts of solar power to the grid this year, amounting to half of all new generation brought online in 2025. (EIA)
A red flag for gas stoves: A new Colorado law will require gas stoves to come with labels that warn buyers about the carcinogens and pollution the appliances emit, though a lawsuit has delayed its implementation for now. (Canary Media)
Cruising to electrification: New York City debuts its first hybrid-electric ferry, which is making trips from Manhattan to an emerging climate-change research hub on Governors Island. (Canary Media)
Counting on cleanup: California advocates worry Phillips 66 may shirk its responsibilities to clean up a “lake of hydrocarbons” that has accumulated under a Los Angeles-area refinery slated for closure later this year. (Capital & Main)
This story was originally published by THE CITY. Sign up to get the latest New York City news delivered to you each morning.
When a heat wave hits New York City, many customers can soon expect a message from Con Edison, asking customers to conserve energy.
The reason is to protect the heat-strained electric grid, which, when taxed to the point of failure, can lead to blackouts and brownouts.
Addy Spiller, an Upper West Sider and founder of a product management business, said those messages from Con Ed drive her bananas.
“Listen, I don’t know how to use less electricity,” she said. “I already have the AC at a reasonable temperature. I don’t think I can do enough to help Con Ed on my own.”
But this summer, Spiller and her dog, Ranger, are among 65 households across the city actually doing more to help — and they don’t have to stop blasting their ACs on sweltering days. That’s because they’re participating in an experiment that connects their air conditioning units to small batteries in their homes. The batteries, about the size of a small microwave oven, plug into wall sockets.
The pilot program, called Responsible Grid, is run by the company Standard Potential in partnership with Con Ed. When demand for energy is high but the utility company needs customers to lay off, the company powers the participants’ AC units with the battery instead of the electric grid.
“There’s a class of large portable phone chargers almost, and instead of powering a whole building, they power a single device and take it off the grid,” said Andrew Wang, Standard Potential’s CEO. “Because we have the battery, it allows folks to participate in the program without having to adjust their comfort levels.”
If more New Yorkers were to connect electric appliances to batteries in their homes, this approach could make the city more resilient, add to the stability of the electric grid, and keep people cool. Responsible Grid is one of about a dozen programs residential Con Ed customers can enroll in to reduce energy during key windows and get financial rewards.
Participants who have the freely provided batteries in their homes through September will also receive about $100 per air conditioning unit plugged into them from Responsible Grid, as Con Ed pays the company to reduce demand.
In southeast Queens, participant Farudh Emiel noticed several times over the hottest days of the summer that his three air conditioning units plugged into the batteries he got through the pilot program kept pumping even as he saw lights dimming. It was likely Con Ed reduced the voltage in his neighborhood to protect the electric system, but his AC units, relying on the batteries, were unaffected.
“I run my ACs 24/7, three of them at the same time,” Emiel said. “One thing I will spend money on is electricity because I don’t want to sweat.”
Outside the individual homes of the participants, batteries have the potential to reshape the electric-supply system and protect ratepayers’ wallets.
When demand for power is high, especially in the summer, fossil-fuel-fired peaker plants kick in to meet that need. Those plants, often located in and around low-income neighborhoods, can be highly polluting and costly to rely on.
“By switching your AC to a battery rather than the outlet, you’re providing a measure of relief to the grid, avoiding more expensive, dirtier power plants turning on,” said Jamie Dickerson, senior director of climate and clean energy programs at Acadia Center, a research and advocacy nonprofit.
The small batteries in participants’ homes have served as a source of backup power in other instances.
In the midst of a heat wave in July, Emiel had just finished cooking a meal when the power went out in his neighborhood. He scurried around his home — a detached, multistory house — to connect his refrigerator, WiFi router, and TV to the batteries.
“We were the only house with electricity because of the stand-alone batteries,” said Emiel, who works as a policy manager for a clean-energy advocacy organization. “We had internet still, we were charging our phones, we had a lamp connected. The air conditioning was still working.”
The blackout lasted for about four hours, he said.
Spiller, too, relied on her batteries in early June, when her prewar apartment building had a planned electrical outage to do some upgrades. The day was hot, and she began feeling stressed as she wondered where she should bring her dog and how she’d get her work done. But then she remembered the battery.
“With the battery, I was able to continue working. My AC worked, my WiFi worked,” Spiller said. “It was such a relief to realize I had a little bit of a buffer and didn’t have to leave my house — I was able to continue just living.”
New York state is looking to deploy large batteries to help make the grid more reliable, especially as officials look to add more forms of renewable energy to replace fossil-fuel sources, and as electric heating, stoves, and vehicles become more common. Wind and solar projects produce power intermittently, but batteries can store extra energy and discharge it back into the grid when the wind doesn’t blow or the sun doesn’t shine.
But connecting big batteries to the grid requires navigating lots of red tape and finding major real estate, two tough tasks in New York City that can slow down adoption.
Jesse Jenkins, a professor in energy and engineering at Princeton University, called the pilot a “compelling model and a good way to avoid the very high costs and bureaucratic headaches of trying to install a grid-connected home battery or solar system.”
But he added that eventually, getting more customers to put the batteries “comes down to the cost of these devices, and whether the value delivered exceeds that cost.”
Looking ahead, Wang said he’s looking forward to scaling up the program to include more participants next summer, and to potentially try pairing the batteries with electric heat pumps in the winter.
.svg)
