The news hit my feed at 6:42 AM Vancouver time — a time when the city is still wrapped in its signature drizzle, and the crypto world is usually quiet. Ormat Technologies, the undisputed heavyweight of geothermal energy, has announced a pivot to AI-driven Enhanced Geothermal Systems. The market's reaction was predictable: a modest bump, a few excited tweets, a round of think pieces on how AI and clean power are finally merging into a single narrative. But as I read through the press release, the same question kept biting at me: who is this story really for?
This isn't a story about energy. It's a story about trust. Or rather, it's a story about how we choose to verify the claims that shape our infrastructure. And the first thing I want to do is throw out the entire framework that this announcement assumes. Because if you look at this through the lens of someone who spent the last decade building decentralized systems, you'll see that Ormat's pivot is not a revolution. It's a consolidation.
Let's start with the technology. Enhanced Geothermal Systems — EGS — is not a new idea. The concept of extracting heat from hot, dry rock by fracturing it, injecting water, and then harvesting steam has been around since the 1970s. The US Department of Energy, Japan, Germany, and France have all run EGS testbeds. The fundamental challenge has never been about finding the hot rock. It's about creating a reservoir in rock that has no natural permeability — and then keeping that reservoir open for decades. That means dealing with induced seismicity, water loss, and the slow degradation of thermal output.
So what does AI actually change? It changes the optimization game. Machine learning can analyze geological data to identify the best targets. It can simulate and adjust hydraulic fracturing plans to reduce seismic risk. It can optimize the flow rate between injection and production wells in real time. It can predict equipment failures before they happen. That's a meaningful value proposition. It's also the same kind of value proposition we see in wind farm predictive maintenance or solar array digital twinning. It's not a fundamental breakthrough.
Ormat's 'pivot' is a branding exercise, not a technological leap.
But the branding matters because it's directed at a very specific audience: the AI data center industry. And this is where the story gets interesting. AI data centers don't have the luxury of intermittent energy. They need 24/7, they need zero-carbon, and they need reliable. That's exactly what geothermal delivers. It's the only renewable source that can provide baseload power without the need for massive storage. Solar and wind are great, but they require grid-scale batteries. Geothermal is the gift that keeps on giving. And with AI model training cycles running for days, a single outage can cost millions. For these companies, the promise of a constant, dispatchable, zero-carbon source is not just an energy narrative. It's a survival mechanism.
Now let's get to the core of my argument, and I have to warn you: this is where I'll be skeptical of the entire framing. Because in the world of blockchain, we have a saying: 'Code is law, but people are the soul. ' The same applies to energy. If the physical infrastructure is centralized, then no amount of on-chain voting or governance tokens will decentralize the actual power. Ormat's pivot, no matter how well they talk about AI and machine learning, is still a single company controlling a physical resource in a specific geographical location. It's the equivalent of a sovereign sidechain with a multisig controlled by a single entity.
This is where the 'dual-audience' tension comes in. For the data center industry, Ormat is offering stability. For the crypto world, they're offering a narrative that the energy is clean, the AI is smart, and the future is green. But what does this mean for the average person? What does it mean for the communities that live near these geothermal sites?
The article I read about Ormat's pivot completely ignored the environmental risks. EGS projects can induce earthquakes. The hydraulic fracturing process can alter groundwater resources. And if something goes wrong, it can cause significant subsidence or contamination. These are not peripheral concerns. They are existential risks to the project's social license to operate. When the original report for the article I was reading mentioned 'AI-driven geothermal,' it was probably trying to sell the idea that AI will solve all these problems. But AI can't prevent an earthquake. It can only predict the probability of one. And even that prediction is probabilistic, not absolute.
Now, here's the contrarian angle that I think is missing from this conversation. The financial incentives for Ormat aren't necessarily aligned with the 'AI revolution.' Ormat is the largest independent geothermal operator in the world, managing about 1.5 GW of capacity. They've got decades of experience in traditional hydrothermal resources. EGS is a new frontier. And they're coming in as a challenger, not a pioneer. There's a company called Fervo Energy, a private firm backed by Bill Gates and Google, that has already demonstrated successful commercial-scale EGS projects. They've already signed a power purchase agreement with Google for a data center in Nevada. So why is Ormat suddenly making this announcement? Could it be a reaction to losing out on the AI market? Could it be a way to attract investors who are piling into the AI narrative?
Based on my years of auditing governance protocols, I've learned to ask: who is the counterparty in the deal? When a company suddenly announces a 'new strategic direction' without providing specific technical data, the market is often seeing a narrative designed to capture attention and capital, not to share transparent information. And that's precisely the case here. The announcement is heavy on vision and light on numbers. There's no LCOE estimate, no stated drilling depth, no fracture design details. That's a red flag.
But let me not just be a skeptic. There's a real and massive opportunity here. The energy needs of AI data centers are structural. This isn't a short-term trend. The demand for 24/7 zero-carbon energy is a multi-decade, multi-trillion dollar opportunity. And geothermal is positioned better than any other renewable to meet that demand. It provides a natural hedge against grid instability and the intermittency of solar and wind. If Ormat can actually deliver on the promise of EGS at commercial scale, they will be a key player in the AI supply chain. But the key question is: can they deliver at a scale that matters? And can they do it fast enough before more nimble startups like Fervo take the best sites and the best contracts?
There's also a hidden policy layer. The article I read didn't mention the Inflation Reduction Act (IRA). This is a glaring omission. The IRA provides a 30% investment tax credit for geothermal projects, and it has specific allocations for EGS demonstration projects. In fact, without the IRA, the economic case for EGS is much weaker. The US Department of Energy has committed billions to EGS demonstration projects. So Ormat's pivot is not just a technological bet; it's a policy bet. They're betting that the IRA will remain intact and that they'll get their share of the tax credit. But there's a political risk. The next election cycle could change this. The IRA is a bipartisan bill, but the details are still subject to political pressure. If the tax credit is reduced, the project's economics could fall apart.
Now, let's bring it back to the blockchain. I am a DAO governance architect, and I have lived through the liquidity traps of the DeFi summer and the winter of value. I know what it means to build something with a philosophy of decentralization and then watch it crash on the reality of technical constraints. The same is true for energy. Decentralization is a verb, not a noun. It's a process, not a fixed state. Ormat is not decentralizing energy. It's centralizing it under a new, more technically sophisticated banner. The AI narrative is a shiny object, but the underlying structure remains the same: a single entity owns the resources, controls the distribution, and dictates the price. This is the opposite of what we're trying to build.
So, what is the 'information gain' here? I want to offer you a new lens, a different way to think about this announcement. It's not about AI and geothermal. It's about the clash of two different types of 'trust' in the global economy. On the one hand, you have the financial trust of the market, which is based on the promises of a company and the speculation of investors. On the other hand, you have the computational trust of the blockchain — a trust that is verifiable, transparent, and, ideally, decentralized. And what I see happening is a conflict: the energy sector is trying to use the 'trust' of AI to buy the trust of the market, but it's not building a system that is verifiable.
Trust isn't a feature; it's a process.
I know this from my experience with LibertyDAO. We had a beautiful governance model, but we lacked the operational discipline to execute it. We had a smart contract, but we had no human security. We learned the hard way that code is law, but people are the soul. The same is true for Ormat's pivot. They have a sophisticated technical plan, but the soul of the plan is still a centralized company maximizing shareholder value. The data center's demand for power is a concentrated source of influence. The company that controls the power controls the data. And the company that controls the data controls the AI. It's a recursion of the same problem we've been trying to solve in the crypto world.
So what's the alternative? Do I suggest that we need to decentralize energy? That's a tough sell. But I do suggest that we need to think more carefully about the governance structure of our energy infrastructure. If we're going to rely on EGS for the AI boom, we need to consider the social, political, and economic impacts on the communities where these projects are located. Will they get a say? Will they benefit from the revenue? Or will they simply bear the risk of induced seismicity and water resource depletion? The answers are not included in Ormat's announcement.
I'm reminded of a story I read about a similar project in Iceland. The country is a natural haven for geothermal, and it's been able to build a decentralized energy grid in which communities have a say. But Iceland is an exception, not the rule. Most of the world's energy infrastructure is built on a top-down model. The announcement of Ormat is not a step toward decentralization. It's a step toward more efficient centralization. It's a step toward more efficient data extraction.
Let's go back to the market signals. The market is excited about this story because it merges two very trendy themes. But if you look at the actual market dynamics, there's a race to the bottom. The competition for data center PPAs is intense. The first movers will get the contracts, and the others will be left out. So, Ormat needs to move fast. But moving fast in EGS is risky. Drilling deeper, fracking harder, and building bigger carries significant technical and operational risk. And the market doesn't reward risk; it rewards the appearance of progress. The press release is a form of marketing, not a technical roadmap.
So, what do I think should happen? I think we need to shift the narrative. Instead of asking, 'How can AI help geothermal?', we should ask, 'How can geothermal help a more decentralized future?' What if we tokenize the energy output of an EGS project? What if we allow local communities to own a stake in the project, or at least have a governance token to vote on how to allocate the energy surplus? What if we create a transparent, verifiable system that tracks the carbon footprint from the drilling to the data center? That would be a real innovation. But that's not what's being proposed.
Decentralization is a verb, not a noun.
It's not something you can claim in a press release; it's something you have to practice in the code, in the contracts, and in the community. And I'm not sure Ormat is ready for that. They're a traditional energy company. They're not built for participatory governance. They're built for shareholder value.
Let me share a piece of my own technical experience. During the DeFi winter, I was brought in to audit a DAO that was using a simple multisig to manage their treasury. The multisig was secure, but the governance process was a disaster. The 'governance token' was held by a few whales, and the community had no real voice. When the market crashed, the treasury was drained by a cleverly crafted attack that exploited the speed of the process, not the code. The lesson was that technology is not a substitute for a social process. It's the same with EGS. AI can't replace the social license to operate. It can only help you get to the site more efficiently. But it won't help you build the trust.
So, I have to look at the crypto angle. This announcement is generating buzz in the crypto community because it's tied to AI. But is it a good investment? I'm not a financial advisor, but I am a risk manager. The risk is not just technical; it's a policy risk. The IRA is a good support, but it could be a sunset. And the competition is intense. Fervo is already there, and they have the benefit of being more agile. Ormat has scale, but scale can be a weakness in a high-tech niche. They're not built for speed; they're built for stability. The AI race requires speed.
But I want to step back and see the bigger picture. The combination of geothermal and AI is a signal of a larger trend: the 'energy-constraint' of the AI economy. Every single large model needs to be trained and served, and every single request consumes energy. The infrastructure that powers the AI is just as important as the algorithm. This is a huge opportunity for renewable energy companies, but it's also a huge responsibility. We're putting the world's digital economy on the back of a few energy sources. If we're not careful, we might end up with a new energy 'monopoly' that is just as powerful as the current tech giants.
Code is law, but people are the soul.
If we don't get the governance right, we're building a new empire, not a new world. And that's what scares me about this announcement. It's a step forward in technology, but it's a step backward in the 'decentralized' ethos that we're trying to create. We can't just be pro-crypto and pro-AI and ignore the energy constraints of the world. We have to be 'green' in the true sense of the word. And that means not just reducing carbon emissions, but also ensuring that the benefits of this energy are shared fairly.
So, what do I see the future? I see a bifurcation. On the one hand, we'll see centralized energy systems that are optimized by AI, serving the biggest tech companies. On the other, we'll see a push for more community-owned energy grids, built on open-source technology, where the data is transparent, and the governance is shared. The question is which one will have the upper hand. If the market is the only driving force, the first one will win. But if we can align the incentives of the community, the environment, and the energy producers, we might have a chance to build a better system.
I'm not saying Ormat's project is bad. I'm saying it's incomplete. It's a piece of the puzzle, but the puzzle isn't complete. The AI data center needs power, and geothermal is a great source. But the power is not the end goal. The end goal is a more sustainable, equitable, and resilient future. That future is not guaranteed by a press release. It's not guaranteed by AI. It's guaranteed by the choices we make as a community. And the community needs to be informed.
This is the call to action. Let's not be passive consumers of this narrative. Let's be active participants in the process. Let's ask the hard questions. Where is the project located? What are the environmental risks? Who owns the infrastructure? What are the community benefits? Let's demand transparency, not just buzzwords. Let's use the tools of blockchain — or even just the tools of good governance — to create a better energy future.
Decentralization is a verb, not a noun.
We can start by treating energy as a public good, not a commodity. We can start by supporting projects that are transparent about their technical and social risks. We can start by demanding that the energy that powers the AI revolution is not just clean, but fair. The future is not a set of press releases. It's a set of choices. Let's choose wisely.
The news from Ormat is a tiny shift in the market, but it's a symptom of a larger shift in the energy landscape. It's a wake-up call for the crypto world to look beyond the 'digital' and see the 'physical' and the 'social'. We are not just building a financial infrastructure; we are building a human one. And that requires us to be more than just technical architects. It requires us to be social architects. The question is: are we ready to take that responsibility?
I'll leave you with a question: Can we have a decentralized AI if we have a centralized energy grid? The answer is likely no. So, let's start thinking about how we can decentralize the entire stack, from the energy source to the data center. That's the real challenge. And I'm looking forward to the day when the news headline is not about a company pivot, but about a community transformation.