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:
Artificial intelligence is also involved.

Categorizing Crusoe Energy is not easy. The startup is a Bitcoin miner and data center operator. It’s a “high-performance” and “carbon-negative” cloud platform provider. It’s a darling of the clean tech world that’s raised nearly $750 million in funding. The company has historically powered its operations with natural gas, but its overall business model actually reduces emissions. Confused yet?
Here are the basics. The company was founded in 2018 to address the problem of natural gas flaring. Natural gas is a byproduct of oil extraction, and if oil field operators have no economical use case for the gas or are unable to transfer it elsewhere, it’s often simply burned. If you, like me, have spent time sourcing stock images of air pollution, you’ve probably seen the pictures of giant flames coming out of tall smokestacks near oil pump jacks and other drilling infrastructure. That’s what flaring natural gas looks like, and it is indeed terrible for the environment. That’s largely because the process fails to fully combust methane, which is the primary component of natural gas and 84 times more potent than carbon dioxide over a 20-year period.
That’s where Crusoe comes in. The company’s co-founder, president, and chief operating officer, Cully Cavness was working in the oil and gas industry when he realized that stranded natural gas could be harnessed to power Bitcoin-mining data centers if they were sited directly next to the oil field infrastructure. Burning natural gas for electricity production fully combusts methane, producing CO2 as a byproduct. Still bad, you might say! But it’s definitely not as bad as methane leaking into the atmosphere via flaring, the status quo where Crusoe operates.
So regardless of what one might think of the utility of Bitcoin mining overall, “if you were to delete what we’re doing you would just have a big ball of fire and that would be worse,” Cavness told me.
Plus, it’s dirt cheap. “It is the lowest cost way to generate power that we’ve ever seen,” Cavness said, though he wouldn’t disclose exactly how much Crusoe pays the oil companies for their natural gas. “This is truly a waste product. I mean, there is no value being ascribed to it.”
According to Crusoe’s most recent ESG report, for every ton of CO2 equivalent that the company produced in 2022, it reduced over 1.6 tons through avoided methane emissions. And the opportunity for growth is enormous. “There is a huge amount of flared gas around the world,” Cavness said. “If you captured it all, it would power like two thirds of all of Europe’s electricity and it would power the entire data center industry many times over.”
Of course, in an ideal world, flared gas wouldn’t even be an option. There have been some state level-efforts to ban “routine flaring” in Colorado, New Mexico, and Alaska, but enforcement has often fallen short. “Nothing about flaring should be routine,” Deborah Gordon, a methane expert at the think tank RMI, told me. “It should be an emergency piece of equipment. It’s there to handle a burst of gas that would otherwise present a safety problem to the people on the ground.”
But in the places where Crusoe operates, Cavness said flare gas is available 98% to 99% of the time. Today, the company has about 30 sites located throughout all the major oil fields in the U.S., plus one facility in Argentina.
Gordon views circumstances like this, where gas is being perpetually flared, as “opportunities to decommission” oil wells. But given sheer demand, that may not be an economically or politically feasible solution in the short term. Last year was a record-setting one for oil production, as the U.S. pumped more than any country had in history.
So given that oil isn’t going to disappear overnight, this particular fossil-fuel powered Bitcoin miner has been wildly successful with climate-focused investors. Two years ago, Crusoe closed its $350 million Series C round, led by clean tech investor G2 Venture Partners with participation by existing climate tech venture firms Lowercarbon Capital and MCJ Collective, among others.
“It’s not just the lowest hash rate for Bitcoin mining, or the cheapest cost of compute. It’s also the greenest and when those two things are true, you’ve got an amazing business on your hands,” Clay Dumas, a partner at Lowercarbon Capital, told me. He views shutting down oil fields that flare natural gas as simply “not tractable” given today’s energy environment.
But now Crusoe is shifting its focus on multiple fronts. Cavness told me the company never planned to build its long-term business solely around Bitcoin mining, though historically nearly all of its revenue has come from the famously volatile world of cryptocurrencies. His co-founder, Chase Lochmiller, has a masters in computer science with a focus on artificial intelligence and has long understood AI’s energy demands.
“And so since way before ChatGPT, we’d had a view that GPU computing was going to be actually the bigger opportunity and the bigger driver of data center power demand. And if we could align that with wasted energy sources and other curtailed energy sources, it could be a really effective approach to reduce costs and also reduce emissions,” Cavness told me.
Last year the company expanded its Crusoe Cloud service, which is essentially its version of Amazon Web Services or Microsoft Azure. It works like this: Crusoe builds the data centers (or co-locates with existing facilities), buys the GPU servers, and operates a software layer on top of it all. Then, companies looking to train AI models or synthesize large datasets pay to access Crusoe Cloud over the internet, remotely spinning up Crusoe’s GPU clusters to do the hard lifting.
Last month, Cavness said that the majority of the company’s revenue came from its AI data centers, outpacing Bitcoin revenue for the first time. If all goes according to plan, AI will comprise more than 75% of the company’s revenue by year’s end. “You couldn’t really have timed the launch of a cloud business focused on generative AI much better than they did,” Dumas told me.
Then, as the world (potentially and eventually) moves away from oil, Crusoe is also shifting its focus towards stranded renewable assets. That means sourcing power from areas where there’s excess wind, solar, hydropower, or geothermal on the grid, which leads to curtailment or negative pricing for these resources. “So that’s how we think about operating on the other side of the energy transition,” Cavness told me. This business model, he said, creates an incentive for renewable operators to build even more capacity, since they know they’ll have customers for their excess energy.
Of course, Crusoe isn’t the only company and data centers aren’t the only industry looking to access the cheap power that stranded renewables can supply. Excess clean energy could be used to make green hydrogen, provide heating and cooling for buildings, operate direct air capture facilities, or power microgrids. If renewables are used to mine speculative cryptocurrencies, many would likely argue there are worthier opportunities.
But high compute data centers — whether they’re mining Bitcoin or training AI models — do have one major advantage. “You can talk about highest use from a CO2 avoidance standpoint. But generally, the market is going to treat highest use as the greatest willingness to pay,” Dumas told me. “At this particular moment, it’s hard for me to imagine any application that has a higher willingness to pay, and that is more deployable than data centers.”
Crusoe wouldn’t reveal what portion of its operations run on renewables vs. natural gas. The company’s current focus is expanding its Crusoe Cloud service in Iceland, partnering with an existing data center that’s powered by the country’s abundant hydropower and geothermal energy. Crusoe also says it’s working to develop domestic behind-the-meter wind and solar projects, which would be separate from the main grid and directly supply their data centers with power, though none have been formally announced yet.
Ultimately though, whether Crusoe uses renewables or flare gas, whether it mines Bitcoin or trains AI models, investors have decided that it’s undeniably better than business as usual. “You can complain all you want about the carbon emissions of Bitcoin and compute, but they’re not going anywhere except for up,” Dumas told me, saying it’s incumbent upon us to bring this new computational power to market as cleanly as possible. “And that’s really what Crusoe’s in a position to do.”
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
There has been no new nuclear construction in the U.S. since Vogtle, but the workers are still plenty busy.
The Trump administration wants to have 10 new large nuclear reactors under construction by 2030 — an ambitious goal under any circumstances. It looks downright zany, though, when you consider that the workforce that should be driving steel into the ground, pouring concrete, and laying down wires for nuclear plants is instead building and linking up data centers.
This isn’t how it was supposed to be. Thousands of people, from construction laborers to pipefitters to electricians, worked on the two new reactors at the Plant Vogtle in Georgia, which were intended to be the start of a sequence of projects, erecting new Westinghouse AP1000 reactors across Georgia and South Carolina. Instead, years of delays and cost overruns resulted in two long-delayed reactors 35 miles southeast of Augusta, Georgia — and nothing else.
“We had challenges as we were building a new supply chain for a new technology and then workforce,” John Williams, an executive at Southern Nuclear Operating Company, which owns over 45% of Plant Vogtle, said in a webinar hosted by the environmental group Resources for the Future in October.
“It had been 30 years since we had built a new nuclear plant from scratch in the United States. Our workforce didn’t have that muscle memory that they have in other parts of the world, where they have been building on a more regular frequency.”
That workforce “hasn’t been building nuclear plants” since heavy construction stopped at Vogtle in 2023, he noted — but they have been busy “building data centers and car manufacturing in Georgia.”
Williams said that it would take another “six to 10” AP1000 projects for costs to come down far enough to make nuclear construction routine. “If we were currently building the next AP1000s, we would be farther down that road,” he said. “But we’ve stopped again.”
J.R. Richardson, business manager and financial secretary of the International Brotherhood of Electric Workers Local 1579, based in Augusta, Georgia, told me his union “had 2,000 electricians on that job,” referring to Vogtle. “So now we have a skill set with electricians that did that project. If you wait 20 or 30 years, that skill set is not going to be there anymore.”
Richardson pointed to the potential revitalization of the failed V.C. Summer nuclear project in South Carolina, saying that his union had already been reached out to about it starting up again. Until then, he said, he had 350 electricians working on a Meta data center project between Augusta and Atlanta.
“They’re all basically the same,” he told me of the data center projects. “They’re like cookie cutter homes, but it’s on a bigger scale.”
To be clear, though the segue from nuclear construction to data center construction may hold back the nuclear industry, it has been great for workers, especially unionized electrical and construction workers.
“If an IBEW electrician says they're going hungry, something’s wrong with them,” Richardson said.
Meta’s Northwest Louisiana data center project will require 700 or 800 electricians sitewide, Richardson told me. He estimated that of the IBEW’s 875,000 members, about a tenth were working on data centers, and about 30% of his local were on a single data center job.
When I asked him whether that workforce could be reassembled for future nuclear plants, he said that the “majority” of the workforce likes working on nuclear projects, even if they’re currently doing data center work. “A lot of IBEW electricians look at the longevity of the job,” Richardson told me — and nuclear plants famously take a long, long time to build.
America isn’t building any new nuclear power plants right now (though it will soon if Rick Perry gets his way), but the question of how to balance a workforce between energy construction and data center projects is a pressing one across the country.
It’s not just nuclear developers that have to think about data centers when it comes to recruiting workers — it’s renewables developers, as well.
“We don’t see people leaving the workforce,” said Adam Sokolski, director of regulatory and economic affairs at EDF Renewables North America. “We do see some competition.”
He pointed specifically to Ohio, where he said, “You have a strong concentration of solar happening at the same time as a strong concentration of data center work and manufacturing expansion. There’s something in the water there.”
Sokolski told me that for EDF’s renewable projects, in order to secure workers, he and the company have to “communicate real early where we know we’re going to do a project and start talking to labor in those areas. We’re trying to give them a market signal as a way to say, We’re going to be here in two years.”
Solar and data center projects have lots of overlapping personnel needs, Sokolski said. There are operating engineers “working excavators and bulldozers and graders” or pounding posts into place. And then, of course, there are electricians, who Sokolski said were “a big, big piece of the puzzle — everything from picking up the solar panel off from the pallet to installing it on the racking system, wiring it together to the substations, the inverters to the communication systems, ultimately up to the high voltage step-up transformers and onto the grid.”
On the other hand, explained Kevin Pranis, marketing manager of the Great Lakes regional organizing committee of the Laborers’ International Union of North America, a data center is like a “fancy, very nice warehouse.” This means that when a data center project starts up, “you basically have pretty much all building trades” working on it. “You’ve got site and civil work, and you’re doing a big concrete foundation, and then you’re erecting iron and putting a building around it.”
Data centers also have more mechanical systems than the average building, “so you have more electricians and more plumbers and pipefitters” on site, as well.
Individual projects may face competition for workers, but Pranis framed the larger issue differently: Renewable energy projects are often built to support data centers. “If we get a data center, that means we probably also get a wind or solar project, and batteries,” he said.
While the data center boom is putting upward pressure on labor demand, Pranis told me that in some parts of the country, like the Upper Midwest, it’s helping to compensate for a slump in commercial real estate, which is one of the bread and butter industries for his construction union.
Data centers, Pranis said, aren’t the best projects for his members to work on. They really like doing manufacturing work. But, he added, it’s “a nice large load and it’s a nice big building, and there’s some number of good jobs.”
A conversation with Dustin Mulvaney of San Jose State University
This week’s conversation is a follow up with Dustin Mulvaney, a professor of environmental studies at San Jose State University. As you may recall we spoke with Mulvaney in the immediate aftermath of the Moss Landing battery fire disaster, which occurred near his university’s campus. Mulvaney told us the blaze created a true-blue PR crisis for the energy storage industry in California and predicted it would cause a wave of local moratoria on development. Eight months after our conversation, it’s clear as day how right he was. So I wanted to check back in with him to see how the state’s development landscape looks now and what the future may hold with the Moss Landing dust settled.
Help my readers get a state of play – where are we now in terms of the post-Moss Landing resistance landscape?
A couple things are going on. Monterey Bay is surrounded by Monterey County and Santa Cruz County and both are considering ordinances around battery storage. That’s different than a ban – important. You can have an ordinance that helps facilitate storage. Some people here are very focused on climate change issues and the grid, because here in Santa Cruz County we’re at a terminal point where there really is no renewable energy, so we have to have battery storage. And like, in Santa Cruz County the ordinance would be for unincorporated areas – I’m not sure how materially that would impact things. There’s one storage project in Watsonville near Moss Landing, and the ordinance wouldn’t even impact that. Even in Monterey County, the idea is to issue a moratorium and again, that’s in unincorporated areas, too.
It’s important to say how important battery storage is going to be for the coastal areas. That’s where you see the opposition, but all of our renewables are trapped in southern California and we have a bottleneck that moves power up and down the state. If California doesn’t get offshore wind or wind from Wyoming into the northern part of the state, we’re relying on batteries to get that part of the grid decarbonized.
In the areas of California where batteries are being opposed, who is supporting them and fighting against the protests? I mean, aside from the developers and an occasional climate activist.
The state has been strongly supporting the industry. Lawmakers in the state have been really behind energy storage and keeping things headed in that direction of more deployment. Other than that, I think you’re right to point out there’s not local advocates saying, “We need more battery storage.” It tends to come from Sacramento. I’m not sure you’d see local folks in energy siting usually, but I think it’s also because we are still actually deploying battery storage in some areas of the state. If we were having even more trouble, maybe we’d have more advocacy for development in response.
Has the Moss Landing incident impacted renewable energy development in California? I’ve seen some references to fears about that incident crop up in fights over solar in Imperial County, for example, which I know has been coveted for development.
Everywhere there’s batteries, people are pointing at Moss Landing and asking how people will deal with fires. I don’t know how powerful the arguments are in California, but I see it in almost every single renewable project that has a battery.
Okay, then what do you think the next phase of this is? Are we just going to be trapped in a battery fire fear cycle, or do you think this backlash will evolve?
We’re starting to see it play out here with the state opt-in process where developers can seek state approval to build without local approval. As this situation after Moss Landing has played out, more battery developers have wound up in the opt-in process. So what we’ll see is more battery developers try to get permission from the state as opposed to local officials.
There are some trade-offs with that. But there are benefits in having more resources to help make the decisions. The state will have more expertise in emergency response, for example, whereas every local jurisdiction has to educate themselves. But no matter what I think they’ll be pursuing the opt-in process – there’s nothing local governments can really do to stop them with that.
Part of what we’re seeing though is, you have to have a community benefit agreement in place for the project to advance under the California Environmental Quality Act. The state has been pretty strict about that, and that’s the one thing local folks could still do – influence whether a developer can get a community benefits agreement with representatives on the ground. That’s the one strategy local folks who want to push back on a battery could use, block those agreements. Other than that, I think some counties here in California may not have much resistance. They need the revenue and see these as economic opportunities.
I can’t help but hear optimism in your tone of voice here. It seems like in spite of the disaster, development is still moving forward. Do you think California is doing a better or worse job than other states at deploying battery storage and handling the trade offs?
Oh, better. I think the opt-in process looks like a nice balance between taking local authority away over things and the better decision-making that can be brought in. The state creating that program is one way to help encourage renewables and avoid a backlash, honestly, while staying on track with its decarbonization goals.
The week’s most important fights around renewable energy.
1. Nantucket, Massachusetts – A federal court for the first time has granted the Trump administration legal permission to rescind permits given to renewable energy projects.
2. Harvey County, Kansas – The sleeper election result of 2025 happened in the town of Halstead, Kansas, where voters backed a moratorium on battery storage.
3. Cheboygan County, Michigan – A group of landowners is waging a new legal challenge against Michigan’s permitting primacy law, which gives renewables developers a shot at circumventing local restrictions.
4. Klamath County, Oregon – It’s not all bad news today, as this rural Oregon county blessed a very large solar project with permits.
5. Muscatine County, Iowa – To quote DJ Khaled, another one: This county is also advancing a solar farm, eliding a handful of upset neighbors.