top of page

Google Quaise Geothermal Deal Secures 50 MW

34 minutes ago
11 min read
Illustrative photograph. Photo: Zifeng Xiong / Pexels.

Google has signed an agreement to purchase 50 megawatts of geothermal electricity from Quaise Energy’s planned Project Obsidian in Central Oregon. Announced on November 17, 2026, the Google Quaise geothermal deal establishes a customer for future generation, with deliveries targeted as early as 2030. It does not establish that the project has already delivered commercial electricity.

The agreement also contains a distinction that matters for assessing Quaise’s technology. Its initial development phase will use conventional drilling. Millimeter-wave drilling, which uses high-frequency electromagnetic energy to break through rock, belongs to a planned second phase. Google is backing a development pathway whose commercial and technical milestones remain ahead.

Google already has geothermal experience through Fervo Energy, providing a useful reference for this agreement. The central tension concerns the distance between securing a buyer and operating a dependable power plant. Quaise has advanced the first task. Its next challenge is turning that commercial commitment into infrastructure that serves Google’s existing operations in The Dalles and surrounding communities.

What the Google Quaise geothermal deal changes

The agreement gives Project Obsidian a named customer before commercial generation begins. That changes the project’s business position, even though it leaves construction and operating performance unresolved.

A power purchase agreement, or PPA, establishes terms for buying electricity. For a developer, having a buyer can make a proposed project easier to explain to potential lenders, investors, contractors, and suppliers. Each can evaluate a project with an identified destination for its output.

That commercial role deserves attention because a generating project needs more than an attractive resource. Equipment orders, construction planning, and financing discussions depend on expectations about future revenue. A customer commitment helps address that question without settling every condition required to start building.

The geothermal purchase agreement identifies the capacity, project, intended beneficiaries, and earliest anticipated delivery date. Those are concrete commitments and expectations. They should remain separate from conclusions about whether the plant is financed, completed, or operating.

The announcement does not provide the full contract needed to assess payment obligations, delivery remedies, or the allocation of construction risk. Without those details, readers cannot determine precisely how much development uncertainty Google has agreed to absorb.

Likewise, purchasing future electricity does not automatically mean purchasing an ownership stake in the developer. Commercial support, equity investment, project lending, and equipment financing perform different jobs. Treating them as interchangeable would overstate what this announcement establishes.

The distinction between power and energy also matters. Megawatts describe a rate of production, while megawatt-hours describe electricity generated over time. A contracted capacity figure alone cannot establish annual deliveries without operating assumptions.

Readers should therefore resist converting the headline capacity directly into a guaranteed quantity of clean electricity. Actual generation will depend on availability, maintenance, and sustained project performance. Those operating results will become meaningful evidence once the plant runs.

The announcement also leaves Google’s current electricity supply unchanged by this particular project. A future purchase supports development; it does not mean geothermal electricity from Obsidian is already reaching Oregon data centers.

This makes the agreement consequential in a specific way. It gives Quaise a stronger commercial foundation for pursuing development. The remaining question concerns execution: whether that foundation supports a plant that meets its promised operating role.

That question leads directly to Google’s motivation. The buyer needs electricity at useful locations and times, while the developer needs confidence that successful generation will find a market.

Google needs generation that fits existing operations

Google’s interest connects future geothermal supply with an existing electricity demand. The intended application is concrete: supporting data center operations in The Dalles and the region’s communities.

For a cloud operator, a clean electricity purchase has several dimensions. The amount matters, but so do delivery timing, location, availability, and the relationship with the surrounding grid. A project can look attractive on paper while still requiring substantial integration work.

Geothermal energy uses underground heat to produce useful energy, including electricity. Its appeal here is the prospect of generation that continues through periods without sunlight or strong winds. That characteristic explains why technology companies are considering it alongside other resources.

The International Energy Agency’s geothermal energy outlook identifies both the opportunity and the obstacles. It describes transferable oil and gas expertise while highlighting development risks, permitting, environmental concerns, and social acceptance.

For Google, the relevant purchasing question extends beyond whether geothermal can generate electricity. It concerns whether this particular development can supply useful power within the company’s infrastructure plans. That creates pressure to evaluate engineering schedules alongside commercial commitments.

The same pressure reaches Quaise. A named customer raises the importance of translating development plans into measurable milestones. Future updates will face a more specific standard than general enthusiasm about underground heat.

Local institutions also have practical questions. A proposed generating plant must fit its electricity system and its community. Claims about affordability or reliability require arrangements that show who pays for infrastructure, how service works, and which obligations protect other customers.

Those questions do not establish that the project will harm local consumers. They identify the information necessary to evaluate promised benefits. A favorable statement from a buyer or developer cannot substitute for the relevant commercial and regulatory details.

For enterprise technology buyers, the connection is indirect but important. A cloud service operates through physical facilities whose electricity requirements persist beyond individual software releases. Infrastructure development therefore belongs in assessments of long-term capacity and sustainability claims.

The announcement provides no basis for predicting changes to Google Cloud pricing, service availability, or individual AI products. Connecting this PPA to an immediate software benefit would skip several unproven steps.

It does, however, show the length of the planning horizon. Electricity procurement for existing computing operations can involve projects expected to arrive years after an agreement. That timing complicates any narrative suggesting new generation instantly follows rising demand.

Google’s position is therefore both supportive and demanding. It can help establish a market for an emerging energy approach, while still needing dependable infrastructure. Quaise must satisfy the latter requirement for the former to become a lasting commercial success.

Conventional drilling comes before the millimeter-wave phase

The first development phase and the later drilling ambition require separate assessments. Combining them would make the Google Quaise geothermal deal sound like a completed demonstration of technology that remains under development.

Quaise says Obsidian’s initial phase will use conventional drilling to reach the conditions required for its project. Its planned second phase would introduce millimeter-wave drilling. This sequence is central to understanding the announcement.

Conventional drilling refers here to mechanical drilling methods, rather than the electromagnetic approach Quaise is developing. Using those methods initially means the first commercial development step should not be described as proof of millimeter-wave performance.

The choice also distinguishes the resource from the method used to reach it. A project can target unusually hot underground conditions while using mechanical equipment. The temperature ambition does not, by itself, identify which drilling technology created the well.

Quaise describes its approach as an enhanced geothermal system, meaning an engineered system for extracting underground heat rather than relying solely on a natural hot-water reservoir. That adds further work beyond reaching a target depth.

A useful well must function within a system that brings heat to the surface and converts it into dependable electricity. Drilling progress therefore addresses an important component, but it cannot represent the whole generating plant.

The staged plan has a plausible development rationale. It allows the company to pursue a commercial project while continuing work on a different drilling approach. That is an interpretation of the sequence, rather than evidence that either phase will succeed.

It also creates distinct standards for judging progress. Initial development should face questions about construction readiness and useful generation. Later millimeter-wave development should face questions about field performance, repeatability, and integration into actual project work.

A successful initial plant would provide evidence about Obsidian’s commercial execution. It would not automatically establish that the later drilling technology can reproduce that success elsewhere. Conversely, progress on the drilling technology would not automatically settle the plant’s operating economics.

This distinction matters because a striking drilling demonstration can dominate coverage. Readers may remember a device cutting rock while overlooking the steps between that demonstration and electricity sales. Those steps remain essential to the business.

The company’s broader ambition concerns access to underground heat beyond favorable locations. Testing that proposition requires evidence across relevant conditions. One successful development would be meaningful without proving universal applicability.

The most useful assessment therefore follows the actual sequence. First, examine whether conventional drilling supports the proposed commercial development. Then examine what the later electromagnetic approach contributes, using results specific to that technology.

That approach also creates a fairer comparison with competitors. Companies should be compared through their demonstrated milestones and project obligations, rather than the most ambitious version of their future plans.

Fervo provides a reference for Google’s geothermal strategy

Google’s earlier geothermal work supplies a reference point, rather than a verdict on Quaise. It shows an existing procurement interest while leaving each developer responsible for its own results.

In November 2023, Google described its operational geothermal project with Fervo in Nevada. The company said the project supplied electricity to the local grid serving its data centers.

Google’s account described horizontal wells and fiber-optic monitoring of flow, temperature, and performance. Those details matter because they connect geothermal development with observable operating conditions. They provide a more useful reference than an unsupported claim about which technology will dominate.

The relationship subsequently expanded. In September 2026, Fervo announced a 396 MW agreement with Google for development at Cape Station in Utah, expected online in 2028. Fervo also described its earlier 115 MW agreement with Google and NV Energy in June 2024.

These agreements establish that Google’s geothermal interest predates Obsidian. The Quaise announcement fits a continuing effort to procure this category of electricity. It does not establish that Google has chosen one developer to replace another.

The different project capacities also require care. A larger contracted figure does not automatically establish better economics, higher reliability, or lower development risk. Projects differ in geography, engineering, schedules, and commercial arrangements.

Fervo’s operating experience offers a concrete comparison point. Quaise’s new agreement establishes a commercial relationship attached to a proposed development. Those are different milestones, and describing them accurately is more informative than forcing a winner.

For Quaise, the comparison raises an evidence standard. Buyers and observers can ask how its project progresses from contracted capacity to verified generation. The existence of another developer’s operating project makes that question more tangible.

For Google, working with multiple geothermal developers creates exposure to different development approaches. Whether that becomes an effective procurement portfolio depends on delivery and integration. The announcement alone cannot establish the eventual performance of the combined effort.

There is also a limit to comparing company narratives. A developer may emphasize drilling speed, another resource quality, and another commercial structure. Those dimensions matter only when connected to dependable project output and repeatable execution.

The main contest in this story remains between commercial commitment and completed infrastructure. Fervo supplies context for that contest, rather than becoming a separate rivalry that overwhelms Obsidian’s own milestones.

Readers should watch whether developers publish comparable evidence. Useful disclosures would distinguish operating output, construction progress, contracted capacity, and future options. Mixing those categories makes an industry appear further along than the underlying projects justify.

A market can accumulate credible buyers before it accumulates dependable plants. The next stage of geothermal development will be judged by how consistently companies close that distance.

A customer commitment leaves development risks unresolved

Securing demand reduces one uncertainty while leaving several engineering and execution questions open. That is the central tradeoff in backing an emerging energy project before its commercial operating record exists.

The IEA’s analysis of geothermal development risks describes challenges involving exploration, upfront financing, and execution. It also argues that interest from data centers alone does not resolve the sector’s commercialization difficulties.

That assessment provides an independent counterweight to company optimism. A motivated customer helps establish demand. It cannot, by itself, establish resource performance, construction readiness, or successful long-term operation.

For Obsidian, the practical questions start with the development pathway. What evidence supports the generating design? Which milestones remain before construction and commissioning? How will the developer communicate changes to its expected schedule?

Commissioning means testing and bringing equipment into service before normal operations. Reaching that stage would represent substantial progress, but sustained operation would answer a further set of questions about reliability and maintenance.

The announcement’s earliest delivery date should therefore remain a target. It should not become an unconditional forecast through repetition. Readers need enough schedule detail to distinguish a development expectation from a completed obligation.

Financing requires similar precision. A purchase agreement can support a financing case, but the announcement does not reveal every funding condition. It would be premature to conclude that all required capital is available solely because Google has signed.

Operating economics also remain a separate question. A credible buyer does not establish the cost of producing electricity over a plant’s life. That assessment depends on construction, maintenance, performance, and financing details beyond the headline announcement.

Environmental and community questions deserve project-specific treatment. Broad concerns about geothermal development should not become allegations about Obsidian without evidence. Equally, broad assurances of local benefits should not bypass examination of the actual project arrangements.

The same principle applies to technology claims. Statements about future output advantages or wider geographic reach should remain attributed expectations until relevant results support them. An agreement does not convert a forecast into independent validation.

This does not make the contract meaningless. It makes its contribution identifiable: commercial support for a development effort that still needs to meet technical and operating requirements.

The Google Quaise geothermal deal therefore supports a qualified conclusion. Google has strengthened the project’s customer position. Whether that commitment produces dependable electricity remains a question for subsequent evidence.

A balanced assessment can recognize both points without treating caution as rejection. Early commercial agreements are part of developing new infrastructure. Their value becomes clearer as projects pass milestones that customers cannot supply through endorsement alone.

Three signals will show whether Obsidian is advancing

The next useful evidence should narrow a specific development uncertainty. Over the next several months, readers should prioritize project milestones over additional expressions of confidence.

The first signal is a clearer development schedule, including any disclosed permitting, construction, or grid-connection milestones. Grid connection means the arrangements and equipment that allow a generator to deliver electricity into the network.

More detailed milestones would strengthen confidence that the project is progressing toward its intended role. A changed schedule would require reassessment, but it would also provide useful information about the constraints developers face.

The important distinction is between a new statement of ambition and an update that changes the project’s readiness. A named contractor, completed approval, or documented construction step would answer a more specific question than another broad technology forecast.

The second signal is field evidence relevant to commercial generation. Readers should look for clearly described testing conditions, measured results, and explanations of how those results support the proposed plant.

Useful reporting would distinguish underground temperature from sustained fluid performance and usable electrical output. Each addresses a different part of the system. A favorable result in one category should not silently stand in for all the others.

Strong results tied to the actual development plan would strengthen the case for execution. Limited or ambiguous results would leave more uncertainty. Either outcome should be evaluated against what the test was designed to establish.

Any millimeter-wave update should also identify its relationship to the planned second phase. That would help readers separate progress on the later technology from progress on the initial commercial development.

The third signal is greater clarity about the commercial delivery structure and local responsibilities. That includes disclosed arrangements for electricity delivery, infrastructure obligations, and the treatment of surrounding customers.

Such details would make claims about community benefits easier to assess. They would also clarify the link between a generating project in Central Oregon and Google’s existing operations in The Dalles.

An absence of immediate disclosure would not prove that development has stalled. Some commercial information can remain confidential. It would, however, limit the conclusions outsiders can draw about financing, risk allocation, or likely consumer outcomes.

These signals should appear in that order because they follow the project’s central challenge. Development readiness establishes a pathway, field evidence supports performance expectations, and commercial detail explains how successful generation reaches its intended users.

For developers and enterprise buyers tracking infrastructure news, keep a record of dated commitments alongside subsequent evidence. Separate announced capacity, construction milestones, operating results, and future technology plans. That makes it easier to recognize progress without treating every announcement as an equivalent achievement.

Follow the Google Quaise geothermal deal through those distinctions. Ask whether each update explains what changed, which uncertainty narrowed, and what remains necessary before delivery. Compare the answers with the original schedule rather than the latest promotional framing. An identified customer is a meaningful starting point; documented execution will determine how much confidence the project earns. When the next announcement arrives, does it bring Obsidian closer to dependable generation, or simply restate why that generation would be valuable?

Give every agent the context to do better work

Connect your agents to the knowledge, decisions, and history already organized in remio.

remio currently supports Windows 10+ (x64) and Macs with Apple silicon.

Your AI Partner at Work
Get more done with remio

Plan. Create. Deliver.
All in one place.

bottom of page