Unlocking Hidden Geothermal Power: How AI is Revolutionizing Energy Discovery

AI is revolutionizing the search for geothermal energy, enabling companies to uncover vast, previously hidden resources. While some geothermal hotspots are evident through surface features like geysers and hot springs, many significant reserves lie thousands of feet underground, obscured from traditional detection methods. Now, artificial intelligence offers a powerful solution to locate these "blind" geothermal systems.

Zanskar, a pioneering startup, recently announced a significant breakthrough: the identification and confirmation of a commercially viable blind geothermal system in the western Nevada desert using AI and advanced computational methods. This marks the first such discovery in over 30 years. Historically, locating new geothermal sites was an expensive, trial-and-error process involving extensive deep drilling. Zanskar's approach introduces a new level of precision.

"We aim to solve this problem that had been unsolvable for decades, and go and finally find those resources and prove that they’re way bigger than previously thought,” states Carl Hoiland, Zanskar’s cofounder and CEO. The company's technology helps pinpoint locations with the critical elements for a successful geothermal plant: high temperatures at accessible depths and permeable rock structures for fluid movement. The newly discovered "Big Blind" site boasts a reservoir reaching 250 °F at approximately 2,700 feet below the surface.

As global electricity demand escalates, geothermal systems provide a consistent power source without emitting greenhouse gases, making them a crucial tool in the fight against climate change. Zanskar's cofounder and CTO, Joel Edwards, highlights the broader potential, noting, "We have dozens of sites that look just like this." For Big Blind, extensive fieldwork has already validated the model's predictions.

The process begins with regional AI models trained on known geothermal hotspots and simulations. These models analyze vast datasets, including geological information, satellite imagery, and fault line data, to predict potential new sites. A key advantage of AI is its ability to process complex phenomena that are challenging for human analysis. Hoiland explains, "If there’s something learnable in the earth, even if it’s a very complex phenomenon that’s hard for us humans to understand, neural nets are capable of learning that, if given enough data."

Once the AI identifies a promising area, a field team conducts further investigation, including drilling shallow holes to detect elevated underground temperatures across a roughly 100-square-mile zone. This prospecting phase provided sufficient confidence for Zanskar to secure a federal lease for Big Blind. Subsequent deep drilling in July and August confirmed the presence of the anticipated hot, permeable rock.

The next steps involve securing permits for plant construction and grid connection, along with lining up necessary investments. Long-term testing of heat and water flow at the site will also continue. John McLennan, technical lead for resource management at Utah FORGE (a US Department of Energy-funded geothermal field site), describes the discovery as "promising" and emphasizes the "tremendous need for methodology that can look for large-scale features."

While Big Blind represents Zanskar's first confirmed discovery of a previously unexplored system, the company has successfully applied its tools to other projects, including identifying and reviving a geothermal power plant in New Mexico and confirming a site previously explored but undeveloped by the industry.

"This is the start of a wave of new, naturally occurring geothermal systems that will have enough heat in place to support power plants,” concludes Edwards, signaling a promising future for AI-driven geothermal energy development.