ESMERALDA COUNTY, Nevada—Geothermal company Zanskar has privately raised $115 million in its latest funding round to find and develop overlooked geothermal fields such as one in Nevada, concealed beneath an arid expanse of high desert.

The land doesn’t vent steam, emit fumaroles, or boast hot springs. In other words, it has none of the telltale signs that companies have typically relied on as they have scoured the West for heat to turn into electrons.

Zanskar, based in Salt Lake City, Utah, said it can bypass finding these signatures altogether and unlock hidden troves of concentrated energy, as well as squeeze out more from known fields, all thanks to artificial intelligence. For years, the company has been feeding field data collected by its geologists to a custom-built model that now reliably spits out prospects, including Big Blind, the Nevada field.

Co-founder Carl Hoiland said Big Blind is proof that the Western U.S. is rife with hidden heat that can produce gobs of power.

“There’s more systems to be discovered than anybody believed,” he said in an interview.

The company has raised $180 million in equity funding to date and didn’t disclose a valuation. Spring Lane Capital led the newest round, which included returning investors Obvious Ventures, Union Square Ventures and Lowercarbon Capital. Zanskar said the funds would support its exploration and development campaign and its plans to build several geothermal power plants before 2030.

The startup says its discoveries can help sate the U.S.’s thirst for power. Tech companies are locked in an AI race that requires ever more electricity, factories are sprouting up around the country, and more Americans are driving electric vehicles. The booming demand is making geothermal hot again because it can deliver firm, clean electricity around the clock.

But so far, it is the companies that develop what is called enhanced geothermal that have caught the limelight. Leading the pack is Fervo Energy, which is backed by Bill Gates’ Breakthrough Energy Ventures. It has raised about $1.5 billion to deploy technology pioneered by oil-and-gas drillers to frack rocks, circulate water, and create geothermal reservoirs.

In theory, companies such as Fervo can create these artificial systems almost anywhere. This eliminates the costs and risks associated with pinpointing naturally occurring pockets of hot water, which are like needles in a haystack. But carving out the reservoirs remains expensive today, although the cost is expected to keep coming down as companies drill more efficiently.

Zanskar says its technology not only helps derisk exploration, it opens access to fields that can be tapped much more cheaply than engineered reservoirs. It says it expects its model to improve drastically over time.

The company still has to prove that it can pull heat from a brand-new site and transform it into electricity. But geothermal experts note that conventional geothermal deserves a second look.

“There’s just huge potential there,” said James Faulds, a professor and geologist at the University of Nevada, Reno. “It could be many tens or hundreds of gigawatts of energy.”

Hoiland and co-founder Joel Edwards befriended each other on a raft floating down the Zanskar River in Northern India when both were young geologists. They later reconnected and launched their company in 2019, with one goal: Help kick-start languid geothermal exploration.

Geothermal today accounts for less than 1% of U.S. power generation, and nearly half of that capacity came online in the 1980s, according to the U.S. Energy Information Administration. In the following decades, the industry was hindered by declining wholesale electricity prices, the plummeting costs of competing wind and solar power, and the risks inherent to exploration.

Most of the publicly available geothermal data was collected when oil-and-gas companies were wildcatting across the West between 1960 and 1980. Zanskar estimates that only about 0.4% of the region has been covered by exploratory drilling for geothermal. Hoiland and Edwards said they believe that if they could brush up this data, they could predict where geothermal fields are and make exploration less risky.

“We were going to have to come in and collect data at scale and do it really cheaply,” Edwards said.

He and Hoiland set out to compile, clean and analyze that existing data with AI. Then, they generated predictions and sent geologists into the field to test them. The initial results were disappointing. Most of the time, the researchers found no trace of heat. But the company kept improving its model thanks to data the geologists sent back. Meanwhile, Zanskar built a team of data scientists to make adjustments and improve the model. In time, the software started delivering—and the breakthroughs started coming.

The company aimed its model at an old geothermal plant it bought in New Mexico and found it had unexploited potential. There, it drilled a well that it says is now the most productive pumped geothermal well in the country. Last summer, it announced that it had discovered a field in an area of Nevada known to the industry but unsuccessfully developed.

Then, there is Big Blind. In the fall of 2023, a crew of geologists reported that the public land it was looking at in Esmeralda County didn’t seem all that promising. They were about to end their survey when the company decided to make one last push. The more data they collected, the more interesting it became: They had stumbled upon a “blind” geothermal field, one that no one had known about.

“When this data flowed in, it was extremely obvious that, like, wow, this is very hot,” said Edwards.

Zanskar acquired leases from the Bureau of Land Management, and, last summer, it drilled two exploration wells. At about 2,700 feet, it found a heat reservoir with temperatures of some 250 degrees Fahrenheit. Zanskar’s model shows that it will find higher temperatures when it drills deeper, the company said.

One day in December, a handful of geologists trekked about the site carrying a gravimeter, a blocky instrument that weighs about 25 pounds and is used for surveys. The crew was looking for clues that would help it better understand what the subsurface looks like, and where to eventually drill production wells.

“Fractures, fluid, pathways, permeability, that’s kind of the name of the game for us,” said Nils Caliandro, the lead geologist for Big Blind.

Zanskar said that it has identified dozens of sites, each with the potential to power tens of thousands of homes, and that it expects to build new power plants in the next three to five years. Like other energy companies, it is facing delays to connect to the grid. But it said that, in some cases, it should be able to shorten lengthy development time for geothermal fields because it expects to find resources on private land and because the projects have a relatively small footprint.

The company said it could eventually create artificial heat reservoirs next to its fields, pairing conventional and enhanced geothermal to produce even more power.

“Zanskar is discovering sites that the rest of this industry is going to be very interested in,” said Hoiland.