Google Superhot Geothermal Deal Transforms Data Center Power

8 min read
0 views
Nov 19, 2026

Google just locked in 50 MW of superhot geothermal power from Oregon volcano rocks for its data centers. The early 2030s delivery could rewrite the rules of clean energy for AI, but one major hurdle remains...

Financial market analysis from 19/11/2026. Market conditions may have changed since publication.

What if the same volcanic heat that once shaped entire landscapes could quietly keep the lights on for the next wave of artificial intelligence? That question stopped me cold when I first heard about Google’s latest energy move. The company has locked in a long-term supply of electricity generated from superhot rocks deep beneath Oregon, aiming to feed its existing data centers starting in the early 2030s. It is not just another clean-power contract. It feels like a quiet turning point.

Google’s Landmark Superhot Geothermal Agreement

Google has signed a power purchase agreement for 50 megawatts of superhot geothermal electricity with Quaise Energy. The power will come from Project Obsidian, located near the Newberry Volcano in Central Oregon, and is earmarked for Google’s data center operations in The Dalles. This marks the first commercial-scale commitment of its kind for electricity drawn from rock heated to nearly 800 degrees Fahrenheit.

Traditional geothermal systems rely on naturally occurring hot-water reservoirs that rarely climb past 600 degrees. Superhot rock takes a different path. Engineers drill far deeper, inject cold water to create high-pressure steam, and spin turbines with far greater intensity. Quaise Energy claims the approach can deliver up to ten times more power per well than conventional geothermal setups. That kind of leap changes the economics almost overnight.

I have followed energy deals for years, and this one stands out because it pairs a hyperscaler’s real, immediate need for always-on power with a technology that still sits on the edge of commercial reality. The early 2030s timeline gives both sides room to prove the concept works at scale. If it does, the implications stretch well beyond one data center campus.

Why Superhot Rock Matters Right Now

Data centers already devour electricity at a staggering rate. Recent projections suggest global demand from these facilities could jump 170 percent between 2025 and 2030, with more than 60 percent of that growth concentrated in the United States. Power has become the permanent bottleneck. Solar and wind help, yet they remain intermittent. Batteries store energy, but they do not generate it. Superhot geothermal offers something closer to firm, flexible, carbon-free baseload.

The Clean Air Task Force has estimated that tapping just one percent of the superhot rock potential inside the United States could yield 4.3 terawatts of firm power. That figure is nearly four times the country’s current total installed generating capacity. Numbers like that are hard to ignore when AI models keep growing hungrier by the month.

In my view, the real story is less about the absolute potential and more about timing. Hyperscalers need solutions that can be contracted today and delivered within a decade. Quaise’s millimeter-wave drilling technology is designed to attach to existing rigs, which should lower the barrier to reaching those extreme depths and temperatures. Without that breakthrough, the resource stays locked away.

How Millimeter Wave Drilling Changes the Game

Conventional drill bits struggle once temperatures climb high enough to destroy mechanical components. Millimeter-wave systems use directed energy to vaporize rock instead of grinding through it. Quaise positions this method as the only practical way to reach superhot conditions economically and at the scale needed to compete with fossil fuels on cost.

Project Obsidian aims to become the first commercial demonstration. If the technology performs as expected, the cost of accessing deep heat drops dramatically while the energy returned per well rises sharply. That combination is what makes the Google agreement more than a symbolic gesture. It creates a commercial foundation that other developers can build upon.

Together with Google, we are building the commercial foundation for superhot geothermal and cementing the United States as the global leader in geothermal power.

– Quaise Energy leadership

Those words carry weight because they come with a signed power purchase agreement attached. Talk is cheap in the energy sector. Contracts with real megawatts and real delivery dates force progress.

Parallel Progress at the Same Volcanic Site

Interestingly, Newberry is also home to work by Mazama Energy. That company recently announced a chilling technology designed to protect critical equipment in rock measuring 645 degrees Fahrenheit while increasing drilling speed by roughly 20 percent. The dual approach—protecting gear while accelerating progress—attacks the cost problem from both ends.

Mazama raised $135 million earlier this year, drawing strategic capital from major energy players alongside private investors. The presence of established oil and gas venture arms signals that traditional energy companies see genuine opportunity in the next generation of geothermal. They understand drilling. They also understand the value of firm power.

Having two serious efforts underway at the same geologic feature creates a quiet race. Competition often sharpens engineering. It also attracts more capital and talent. From where I sit, that is healthy for a technology that still needs to prove itself at commercial scale.

The Broader Context of AI Power Demand

Artificial intelligence is no longer a future energy problem. It is a present one. Training large models and running inference at scale requires continuous electricity. Grid operators already feel the pressure in regions with heavy data-center concentrations. Adding intermittent renewables helps the carbon profile but does not solve the reliability equation on its own.

Superhot geothermal sits in an interesting middle ground. It is renewable in the sense that the Earth’s heat is effectively inexhaustible on human timescales. It is also dispatchable in ways that pure solar or wind cannot match. Once a well field is established and the surface plant is running, output can be managed to follow demand more closely than most renewable sources.

Perhaps the most interesting aspect is how this technology could reduce the need for long-distance transmission of power. If superhot resources can be accessed near existing load centers or industrial sites, the economics improve further. Oregon’s Newberry site happens to sit relatively close to Google’s long-standing data center presence in The Dalles. That geographic coincidence is not trivial.

Economic Realities and Remaining Hurdles

No energy technology arrives without challenges. Drilling to the depths and temperatures required still carries technical risk. Equipment must survive extreme conditions. Regulatory pathways for deep geothermal projects remain less mature than those for conventional oil, gas, or even traditional geothermal. Community acceptance and water management also matter.

Yet the cost curve for many advanced geothermal approaches has begun to bend downward. Parallel improvements in drilling speed, materials, and reservoir management all contribute. The Google contract provides a clear offtake signal that can de-risk further investment. Developers can point to a creditworthy buyer and a defined volume of power. That changes conversations with lenders and equity investors.

I have watched similar dynamics play out in offshore wind and utility-scale solar. Early offtake agreements from large corporates helped those industries cross critical thresholds. Superhot geothermal may now be entering a comparable phase.


What Success Would Look Like

If Project Obsidian delivers its 50 megawatts on schedule and at competitive cost, the demonstration effect will be powerful. Other hyperscalers already exploring advanced geothermal would gain confidence. Utilities facing rising load growth might open more aggressive procurement programs. Policymakers could refine permitting processes to accelerate responsible development.

Success does not require every project to hit the absolute upper end of the theoretical resource. Even a fraction of the estimated potential would represent a meaningful addition to firm clean capacity. The key is proving that the resource can be accessed reliably and economically enough to attract repeated investment.

  • Consistent well performance over multiple years
  • Drilling costs that continue to decline with experience
  • Surface plant reliability matching or exceeding conventional geothermal
  • Clear regulatory frameworks that reduce uncertainty
  • Growing interest from additional large energy buyers

Those milestones would turn an interesting pilot into an industry.

Implications for the Wider Energy Transition

The conversation around clean firm power has intensified in recent years. Nuclear, long-duration storage, hydrogen, and advanced geothermal all compete for attention and capital. Each has strengths and limitations. Superhot rock geothermal’s particular advantage lies in its potential geographic breadth and its ability to leverage existing drilling expertise from the oil and gas sector.

Many of the skills, supply chains, and even some of the capital that once flowed primarily toward fossil fuels can be redirected. That transition is already visible in the investor lists of companies working on next-generation geothermal. It is pragmatic rather than ideological, which often proves more durable.

For data center operators the appeal is straightforward. They need large volumes of reliable, low-carbon electricity under long-term contracts. Superhot geothermal offers a path that does not depend on the weather and does not require the multi-decade construction timelines associated with new nuclear plants. That combination is rare.

A Personal Take on the Timing

I keep coming back to the timing. Five years ago this deal would have seemed speculative. Ten years ago it would have sounded like science fiction. The combination of AI-driven load growth, improved drilling concepts, and corporate appetite for novel clean resources has aligned in a way that feels genuine rather than forced.

Will every technical challenge be solved on the first try? Unlikely. Energy projects of this complexity rarely follow a perfectly straight line. Yet the presence of a major technology buyer willing to commit to 50 megawatts creates accountability and momentum that pure research funding cannot match.

In my experience, markets respond more strongly to demonstrated commercial interest than to optimistic resource estimates alone. Google’s agreement supplies that interest in concrete form.

Looking Ahead to the Early 2030s

By the time the first electrons from Project Obsidian are scheduled to flow, the data center landscape will look different again. AI workloads will have evolved. Efficiency gains will have continued. Yet the fundamental need for reliable power will remain. Technologies that can deliver firm clean electricity at competitive cost will occupy a privileged position.

Superhot geothermal is still early in its commercial journey. The Google deal does not guarantee success, but it does remove one of the largest barriers: the absence of a clear first customer willing to sign a meaningful contract. That step matters more than most people realize.

Other developers are watching closely. Capital is beginning to move. Engineering teams are iterating on the remaining technical problems. The next several years will determine whether superhot rock becomes a meaningful part of the energy mix or remains a promising but limited niche.

For now, the signal is clear. One of the world’s largest buyers of electricity has decided that heat from deep volcanic rock is worth contracting. That decision alone shifts the conversation. The real test will come when the first wells reach temperature and the first turbines begin to turn. Until then, the industry has a concrete target and a powerful incentive to hit it.

The Earth’s heat has always been there. What is new is the combination of technology, capital, and urgent demand that might finally unlock it at scale. Google’s agreement is one of the clearest signs yet that the unlocking process has begun in earnest.

❝
The blockchain is an incorruptible digital ledger of economic transactions that can be programmed to record not just financial transactions but virtually everything of value.
— Don Tapscott
Author

Steven Soarez passionately shares his financial expertise to help everyone better understand and master investing. Contact us for collaboration opportunities or sponsored article inquiries.

Related Articles

?>