Katherine Keddie:
Hello and welcome back to Scaling Green Tech with Katherine Keddie and Matt Jaworski.
Matt Jaworski:
Hello, hello.
Katherine Keddie:
We are here with a very exciting guest, Paul Domjan, who is the founder and chief policy and global affairs officer at Enoda. Thank you so much for joining us, Paul.
Paul Domjan:
Thank you so much for having me. I'm really excited to do this.
Understanding the Grid
Katherine Keddie:
So our first question, as usual, to keep things simple, is how would you describe Enoda, a very technical business, to a five-year-old?
Paul Domjan:
So I'd begin by explaining that when I grew up, In my town, Austin, Texas, but the same thing was true here in London, there was a giant power station literally in the middle of the city on the river that burned horrible, smelly coal. Here, it's what's now Tate Modern. It burned oil. And if you're five years old and you go to Tate Modern, it's probably not that interesting. But if you go in the basement, at the back, it smells terrible. Because 25 years later, that oil has still seeped into it. And that's actually pretty cool if you're five. So you can imagine it was really stinky burning all of this coal and oil in the middle of the city. And at the same time, when I was five, We didn't have iPads. We didn't have YouTube. If you wanted to watch television in America, Sesame Street and Mister Rogers, you turned on the television at 4 o'clock and everyone in the entire country sat down in front of the television at the same time to watch Sesame Street and Mister Rogers. And that was really helpful because the people in the power plant knew they could burn some more coal or some more oil to make sure there was enough power for all of those televisions to come on and watch Sesame Street and Mister Rogers. But today it doesn't work that way anymore. We want to stop using coal and oil and use sun and wind. But we all know that sometimes the weather's good and sometimes it's not. Sometimes it's sunny and sometimes it's not. So if we want sun and wind to provide our energy, The system seems a bit wobbly and unstable. We don't have sun all the time. We don't have wind all the time. And we don't all want to watch the same television program at the same time. You want to pick up your iPad when your parents let you and watch Peppa Pig. And there needs to be energy there to let you watch Peppa Pig exactly when you want to watch Peppa Pig. And that's not how the energy system worked when I was born. So we have a brand new technology for Enoda, the Prime Exchanger, our software that sits on top of that. And that smooths out those wobbles. makes it much easier to ensure that you have the power to do the things you want to do when you want to do them, whether your parents like what you're doing or not.
Katherine Keddie:
Amazing. Wonderful. Very, very good.
Paul Domjan:
I love the Peppa Pig part.
Katherine Keddie:
Yeah, me too.
Paul Domjan:
My children are a bit older. I wasn't sure if Peppa Pig was still popular. I hope it is.
Katherine Keddie:
I think it is. I think it's one of those things that lasts the test of time. We hear a lot of good ones. I think that was definitely up there with the storytelling in particular.
Paul Domjan:
I love the one about the boat and the roof. From for adaptation.
Katherine Keddie:
Oh, yeah Yeah, yeah, he's a prime storyteller absolutely For the technical listeners for the people who are adults and not five-year-olds Listening, please. Give us your elevator pitch of Enoda of kind of give us a bit more flavor about your work your technology Countries you're active in etc.
The Tesla-Westinghouse AC system and its Struggles With Modern Renewables
Paul Domjan:
Certainly so I'll take it from kind of a personal story perspective. When I started my career in my late teens, I was a software developer in .com 1.0. And in those days, we used to talk about how we had the browser at one end and we had the web at the other, and in between was Alexander Graham Bell. And we never got to the modern internet until we fixed the middle piece and replaced Alexander Graham Bell with broadband and mobile and distributed storage and now distributed data centers, et cetera. The energy transitions in a very similar place. We have rooftop solar and electric vehicles and heat pumps and distributed digital loads like urban data centers at one end. At the other end, we have utility scale solar and wind. We have the removal of spinning mass generators. We have huge new requests, especially in the US, but also, for example, in Poland, where there are data center interconnection requests that are more than four times the size of the current system to connect huge assets at the top. And in the middle supporting all of that, we have a 138-year-old system that Nikola Tesla, George Westinghouse, and William Stanley built. And that system isn't designed to work that way. Why is it not designed to work that way? Tesla had a brilliant insight when he built the AC system and designed the AC system, which is that the power generator could do two jobs. It could provide bulk power, but it could also vary its output very precisely in real time to keep the balance of supply and demand, the balance of generation and load stable and provide a stable frequency. And it did that so well that apart from a few very nerdy people in the corners of electrical engineering programs and systems operations departments, no one cared about frequency services. It was just a matter of how do we have sufficient bulk power. So we said, you know, how do we get decarbonized bulk power cheaply? Solar and wind. But solar and wind can't provide any of those services. So we end up in a situation where the system's increasingly unstable because we have to match generation and load in real time, but the resource that did that previously isn't available. It's driving up energy prices, it's imperiling energy security as we saw in Spain, and it's really limiting our ability to utilize distributed resources. At the same time, at the bottom of the system, The system was always designed for us all to do the same things. We all came home. We watched television at the same time, et cetera. They watched, maybe not Peppa Pig, because it was quite predictable, but they watched the football in the control room at the power station, not because they wanted to watch football. Maybe they did, but because they needed to know when the game stopped in order to burn more coal, because people would open the fridge and power would rise, because we lived this very synchronized life. You know, they would call the power station when the referee had added, called out a time in order to be ready to start shoveling because they knew the power demand would peak. That doesn't happen anymore. You know, we have these tremendous distributed resources and the system isn't able to cope with them. You know, my roof might be generating power at the same time as my neighbor's charging her electric vehicle. We're pulling differently on our transformer. We're generating harmonics. We've really broken the Tesla Westinghouse system. And fixing that is our mission at Enoda. Providing frequency services in a distributed fashion to enable renewables to deliver that bulk power cheaply and affordably. Making the distribution system much more flexible so we can accommodate local energy choices and utilize all of that power. Addressing all of the power that's currently being lost, you know, up to 10% of the power we generate that's currently being lost as noise and harmonics. And we do that with an integrated platform of hardware and software.
Katherine Keddie:
Give us a sense of the scale of your work so far. So you mentioned actually on the way in the variety of offices you have around the world. I know you've done a lot of work particularly in Poland but give us a sense of the stage your technology is at.
Distribution Transformers as the Grid's Critical Bottleneck
Paul Domjan:
Sure, so let's look at the whole stack. I think it's important to know what we're talking about first. So at the end of your road, here in London, there'll be a little building that says UK Power Networks Danger of Death. In that little building is your distribution transformer. That distribution transformer is probably 50 years old. It supports, depending on how big they are, between 60 and maybe 120 homes with single-phase power. And it's the workhorse of our network. You'll remember, possibly, I certainly remember, when fiber broadband became available because the equivalent distribution node in the telecom system was replaced with a digital node, and now I could have ADSL, and then it was upgraded, and I could have fiber. We're today very much in the era of modems. That local node is designed to flow in one direction. It's designed on the assumption that all of the demand across the phases is even, et cetera. So we replaced that with our device in Enoda of prime exchanger. It transforms voltage for doing the core job the transformer did, but it also enables us to have tremendous real-time sensing of the network to actively manage the voltage of the network to remove constraints to electric vehicles, rooftop solar, allowing that rooftop solar to flow back upstream, corrects harmonics, etc. Sitting on top of that device is a platform, our nodal asset management platform, which enables those devices to coordinate to light up the utilities. If you think about it today, the electricity system is the biggest machine humanity has ever built, and it's an analog machine with almost no data. You know, we can't see what's happening. If you talk to a typical electric utility about how they understand what's happening in their network, it's basically because of customer complaints. So, we know that there's a problem, not because we had preventative maintenance data, but because someone called up because the power's gone out. And increasingly they can't call when the power comes out, so you have to wait until the power comes back for them to call. So we have that nodal asset management system, which is really an edge to cloud system, because there's so much data that's produced locally on that type of device that we have to process it locally and then send it up to the cloud. And then we have a platform ensemble that enables those devices to coordinate to provide frequency services. We currently have deployed those prime exchangers into a couple of European countries. You might have seen at the end of last year a press release about our first project in Spain, for example. And the intention there is to scale that business to the point where every single transformer in the world at that distribution level is a prime exchanger. Just like today, nobody installs an analog switch in the telecoms network because the analog switch doesn't support the world we have today. We're piloting the nodal asset management layer on top of that, and we're currently working on our first deployments of our aggregator to be able to let those devices and third-party devices to coordinate to provide frequency services.
Matt Jaworski:
And how long did it take you to get to this point? Also, how big is your team? What's been your journey both as a company and as a company founder?
Paul Domjan:
Well, I'll talk about my journey first and then intersect with the company. So I started my career in oil and gas, transportation and politics, building big pipelines out of basically out of and avoiding Russia at Shell and then just down the river in Shell Center. And while I was at Shell, I then got a call from the NATO commander's office to say, would you come become the NATO commander's first energy security advisor and help us develop an energy security strategy for NATO? So this was 2004, as the war in Iraq was beginning, and it was really clear that energy security was a lot more complicated than people realized. So I spent several years at NATO, developed NATO's first energy security strategy. At the time, US European Command, which was the US component of NATO, was also responsible for Africa, so we covered basically former Soviet Union, Europe, and Africa. And what became really clear to me during that process was that there isn't the only real solution for allied energy security is a system based on renewables, because it's the only resource you can produce that you can produce pretty much everywhere domestically, and that you can produce domestically at a price that is an index to oil and gas. So you're able to have both physical energy security and economic energy security. But at the same time, and this is part of what we talked a little bit earlier as we were coming in about terrorism against the electricity system, but at the same time, we learned very vividly from the Irish Republican Army that the electricity system is not inherently stable and will become much less inherently stable as we bring in renewables. So, you know, that journey for me really started in 2005 at NATO. I met my technical co-founder, Andrew Scobie, in 2019, and he and I and our fellow founders, Jackie and Jen, founded this business in 2021. We're now about 150 people, the bulk in our headquarters in Edinburgh, also here in London, Madrid, Krakow, and we're moving into the United States.
Katherine Keddie:
Talk to us about the context in which we're working. So you very interestingly framed renewables as part of this crucial security mandate really that came in. And you also mentioned previously having a modern system that is completely different from knowing when people are going to open the fridge after the football, for example. So what is the context of the demand on the grid, the electricity system now?
Renewable Curtailment and System Balancing
Paul Domjan:
I think it's really important. The first phase of the energy transition was proving that we could build renewables at scale. So you, I think, have a lot of people from the policy community, for example, regulators, et cetera, who had a huge focus for a long period of time on, can we prove that we can build utility-scale renewables? And at that point, we were getting to 10, 15% renewable. And there was still plenty of thermal generation to provide balancing. And the distributed side of it, we call DERs, distributed energy resources, had scaled much less. So all of this local instability was much less of an issue. I worry that some policymakers haven't made the mindset shift to realize, yes, we can build renewables at scale. We know how to do it. You can go to China, you can see it being done. You can go to Texas, you can see it being done. They're totally different systems. They both work really well. But If you take the grid for granted, we will never deliver that power effectively. We see that in this country. We have some of the highest power prices in the world, and we curtail a huge amount of energy because the grid isn't set up to work that way. That's kind of problem number one. Problem number two is this issue of distribution and flexibility. So the grid was originally built on the assumption that demand was totally inflexible, and all of the flexibility had to come from generation. We now have a generation that's not flexible, whether that's solar and wind because they're intermittent, or nuclear because it ramps slowly and in some cases can't ramp, and its economics favor providing continuous base load power. So you need to be able to provide flexibility around that. That's going to have to come from the load side from distribution, not from generation, or we're not going to have a successful transition. If we think about a traditional power system, you would build to the 99.999% maximum load. In the future, we can build to the 80% maximum load because we know we have 20% flex. will be able to deliver a huge amount more power, much more cheaply, power a lot more economic growth at a much lower cost, and more quickly, critically. And we see this in all sorts of areas, from Ireland, which is offering non-firm connection agreements for data centers. One of the first non-firm connection agreements is the electric bus garage in Waterloo. the progress around adopting what I would consider kind of early versions of demand flexibility that are quite expensive and clunky. And then you move to something like us where we can provide a tremendous amount of flexibility across the entire network, not by moving a small number of loads a lot, but by changing the voltage for everyone a little bit in a way that doesn't impact you at all. So it's about moving from a system where Flexibility had to be intentional to a system where flexibility is baked in from day one and doesn't impact the customer. That's kind of transformation number one. And I think that's increasingly understood, but just beginning to be. Transformation number two, which unfortunately is not well understood at all, is that when you open your physics textbook, I don’t know when the last time you did physics was, but if you open your physics textbook and there'll be a chapter on electricity and it will have a beautiful AC sine wave, That's what we consume, the energy that's contiguous with that perfect sine wave. Everything we're doing in the energy system is damaging that. So if you actually plug an oscilloscope into your plug socket, you get something that looks horrible like this. It has voltage transients and harmonics. And if you go upstream to where the three phases from the individual homes intersect, the phases are imbalanced. And that means two things. One is it means an enormous amount of damage to equipment, whether that's inability to connect new data centers because the power quality is too poor. Or it's heating and transformers causing a transformer that should have lasted for 40 years to fail after 20. Or it's the utility saying, because of our operating specifications and our concerns, we're not going to allow this new rooftop solar, or we're not going to be able to reverse power flow this, or we won't enable you to connect a new electric vehicle charger. To put that in context, on a high renewable energy day, you might be losing something like 8% of the power we generate as distortion. So fixing that harmonics problem and distortion problem is making a bigger contribution to primary power than building Hinckley Point C, the UK's new nuclear power station. And we're only beginning, I think, as a kind of policy and wider community outside of the kind of electrical engineering nerds to really appreciate how important it is that we, how big these problems are and how important it is that we fix them. And, you know, one of the things that we're able to do is once we install a prime exchanger in that substation, we can actively correct all those aspects of the electricity signal. So you get back, coming out of that substation at the medium voltage level, a perfect sine wave from your textbook. Even if below that you've got cheap Chinese inverters on rooftop solar, and heat pumps switching on and off, and electrified motors, and all of this turbulence and noise, we're not polluting the system with that. And that lets you scale much faster.
Katherine Keddie:
Sounds like, in some ways, a simple efficiency problem. Why do people not know about this? Why do policymakers not actually?
Flexibility vs Efficiency in Energy Systems
Paul Domjan:
Yeah, let's argue about efficiency. Part of the problem in energy, this is a personal pet peeve, is we talk all the time about efficiency. All right. The price of oil and gas is a function of the cost of getting out of the ground and competition in the market. It's almost always positive, and it's usually quite high. The price of electricity. It doesn't make any sense. Why does it not make any sense? It doesn't make any sense because we don't consume electricity. We participate. And what do I mean by that? So the Tesla Westinghouse system is designed to deliver power at 50 hertz. 60 hertz in the US, parts of Japan, 50 hertz the other half of Japan, everywhere else. That's a function of the level of generation and the level of load. If there's too much generation, and relative to load, the frequency will rise and the system will collapse. If there's too little generation relative to load, the frequency will fall and the system will collapse. When there's too much generation, consuming electricity doesn't have a cost. It makes a positive benefit. So we don't need to optimize for efficiency. If we can reduce power consumption by 20%, we're stuck in the same problem. We have to optimize for flexibility. And I think most people, I didn't even understand the scale of this. So the only country I have really good data for in terms of the impact of that balancing on curtailing renewables happens to be Poland, because they publish it. And in Poland, more than 90% of renewable curtailment is not due to grid constraints, the classic like North-South, although they have a North-South thing just like we do in the UK. It's due to the inability to balance the system. So put differently, you could reduce curtailment in Poland by more than 90% if you could balance the system in real time and consume power when it's available. So we have to think about it as an active participation. So, and that's not how we do it today. And that makes it much more difficult. We have to get out of the idea of efficiency. I mean, efficiency, thermal efficiency, insulating buildings, great. But it's more important that you make your demand flexible than that you make it smaller.
Katherine Keddie:
Flexibility, not efficiency.
Paul Domjan:
Flexibility is flexibility first. Because the more flexible we can be about how we use energy, in particular electricity, the less we need something to complement renewables. If we want to get the maximum value out of renewables, we should be able to use them fully when they're available and use less power when they're not. And that can be batteries, it can be storage heaters, it can be training AI models at certain times of the day and not at others. It can be really simple. You probably never realized it, but the rate at which your kettle boils varies by about 20% depending on the time of day and what's happening in the system. Because the voltage is allowed to vary plus and minus 10%. At Enoda, we take that variation and use it intentionally. So if there's too much renewables, we'll make your kettle boil faster. If there's too little, we'll make it go slower. And that kind of small change across the entire system gives us the flexibility to be able to deploy a lot more renewable energy a lot more cheaply.
Matt Jaworski:
Okay, so we know that the transition is happening, that it has to happen. What is influencing its rate, especially now that big parts of it became impacted by politics and international tensions?
Paul Domjan:
Can we zoom out slightly to start?
Matt Jaworski:
Yeah, absolutely.
Resolving the Energy Trilemma
Paul Domjan:
So there have been, in the whole history of technological revolutions, the typical model, it's what happened to the internet, it's what's happening today with AI. Scientists, innovators develop something new. They bring the new thing to the market. Everyone says, wow, that's a new thing. Is it really going to do new stuff? There's a whole innovation process. And then regulators wake up and are like, goodness, there's a new thing. I don't know what I'm going to do about it, and rush to try to make policy. And that's exactly what we see today with AI. The energy transition is the only policy-driven technology revolution in history. We didn't do it because we invented these technologies. We did it because policymakers said we need to emit less carbon. And so it's always been about politics. And I think that's important to bear in mind. So the question is, why is the politics of it perhaps more challenging today than it has been previously? And by that, I think there's an interesting side note that energy has always been political. In the medieval period, you needed a royal warrant to gather firewood. Politics is about how we define and distribute society's prosperity. And energy is the primary driver of society's prosperity. So they've always been deeply intertwined. It's that prosperity point that I think is part of what's imperiling the transition today. The transition just seems too expensive. And the reality is that humanity is not going to accept that, nor should they. So the question becomes, actually, how do we avoid locking in legacy choices that are going to make the transition very expensive, and quickly move forward to new technologies, including ours, among others, that will enable us to deliver a much more affordable energy transition. If you look, for example, at energy prices today, the price of primary power, I mean, it's risen because of the war in Ukraine, but basically the price of primary power is not the fastest increasing part of energy bills. The fastest increasing part of energy bills is system charges balancing costs, all of these costs around enabling the grid to work with this increasingly decarbonized primary power source. So if we can fix those pieces, we can bring down the cost of the transition and make the transition more sustainable. The way that I would think about it is in terms of the energy trilemma. Traditionally, the trilemma is framed where we can have energy that's secure, energy that's affordable, or energy that's clean, but only two. And I find that already problematic, because I spent the first part of my career just trying to deliver secure, affordable energy without being at all concerned about cleaning. We couldn't do that. We've been trying to do that since the Arab oil crisis. But we certainly couldn't do all three. If you're able to resolve the trilemma, so you're able to make the energy transition something that drives down costs and improves security, then which point in that trilemma is most important to you as a societal choice? You know, do I as a society want more, want to relatively weight more towards affordability or more towards security or more towards a clean transition? But you've resolved the fundamental tension. With the current technology toolbox we have, that fundamental tension isn't resolved. So we see in this country, for example, tremendous concern about if we continue to transition and prices continue to increase, what does that mean for our economy and security? The response is, well, that trilemma exists because of the current level of technology that's available. But the current level of technology that's available is not what will always be available. We would never have built the modern internet by stacking up modems. And that's effectively the equivalent of what we're trying to do, building the energy transition by stacking up batteries and inverters. Why? Why are we doing it? Why are we doing it? I think partly because of that policy mindset we started with. The fact that there was such a concern for so long about how do we build more renewables. Partly because we now have a set of incumbents who are committed to that part of the transition. We talk about the political role of oil and gas, but if you look at the political role of large offshore wind and the latest auction in the UK, there obviously is a political weight there. But I think the biggest one is that There's often, I think, a mistaken mindset of assuming that we have all the tools for the transition. One of the pieces of research that I like the most from when we started in Enoda is this piece of research by McKinsey that showed that less than half of the technologies to execute the transition are currently commercially available. So we have to look at this as an innovation problem. That's kind of one issue. The second issue is when we look at it as an innovation problem, we use poor analogies. So we talk about a moon shot, like the space program was an extremely expensive, very unreliable, quite dangerous way to go to the moon a couple of times. We have to build something that's reliable every single day and affordable. Or they talk about the blitz spirit in the UK, but the whole reason the blitz spirit was sustainable is because eventually you would win the war. We have to build an energy transition that people want to be part of for the long term. rather than frame it as a kind of heroic one-off thing that we're going to get through. No, we need to be building and deploying technologies that people want to be part of their lives forever.
Matt Jaworski:
Something you spoke about quite a few times in different interviews, et cetera, is that those requirements for technologies that are specified at the policy level, they have to be tech agnostic and focused on the solutions, specifying what solution is needed, even if this solution doesn't currently exist.
Technology-Agnostic Policy
Paul Domjan:
That's completely right. My favorite example of this is batteries. Nobody needs a battery. All right, what do we need? We need short-term frequency balancing. We need a way of shifting energy from daytime to nighttime. And we need some solution for two weeks out of the year when it's really cold and there's very little wind. Those are functionally defined problems. Batteries are one of the technologies that solve each of those problems. But far too often policymakers say, for example, we need batteries. Batteries will be part of this. I'm not trying to pick too much on batteries. Batteries will be part of the solution. But you don't need a battery. You need a provider of frequency services, or a provider of day and night diurnal shifting, or a solution for Dunkelflaute. And you should let innovators come up with the best possible way of addressing that functional specification. So I'll give you a simple example. If you use a technology like ours to provide all of the frequency services, then you open up the battery storage space to a whole bunch of innovation in chemistries that aren't fast enough to balance the grid, but are great for day-night shifting. you open up a really interesting issue in how power is used. In order to incentivize that innovation in new battery chemistries, you have to decouple the actual problems to find them functionally and let technologies compete. Otherwise, you may take the risk of making a policy-driven technology choice on a specific solution that is not going to be the one that wins out. And anyone who had a mobile phone in the US in about 2005 knows how badly that can turn out.
Matt Jaworski:
So would you say it's kind of like, what was it called? Electricity wars, whatever is that? Yeah, exactly. When they were happening right at the end of 19th century in the States, if the government or some local authority specified that we need to have DC grid infrastructure, then only Edison's company would be able to build it. and they would be sort of stuck with that and they would be missing out on the benefits they could have gotten from AC Current and Stevenson and the other gentleman whose name I couldn't remember.
Paul Domjan:
My hero is Tesla, but not the cars. But I think that's right. I think the difference today is that You know, we've invested more money in building the AC system than pretty much any project humanity has ever done. So you sometimes hear ideas about, you know, we need to move to DC and, you know, there will be DC microgrids in Africa, but, you know, we need solutions. This is part of our design process. something called design by rationalized constraint to understand what are the constraints in the physics, in the economics, in the human dimension in the policy within which a solution to some of these grid problems has to lie. One of those constraints is we have to be backward compatible with the AC system we currently have. Because we're not going to be able to move away from it, although it might build differently in other places. And there are lots of good reasons to choose AC still. But that's part of why energy innovation is difficult. Because in the days of the current wars, they were starting with nothing. They were building power stations, laying cable, trying to get across the country as quickly as they could. Today, we're starting with this enormous legacy infrastructure. And we have to figure out, how do we transition that legacy infrastructure?
Katherine Keddie:
Who do you think is doing this well? I know you said you do a lot of work in Poland that's quite forward-looking policy in that area. Who do you think is a good example of tech agnostic, forward-looking, considering the weight of this legacy system that we've inherited?
Paul Domjan:
I mean, I think there are different countries that are leading in different ways. So I think what the U.S. is doing well is recognizing that the energy system has to deliver prosperity. I mean, it's the underpinning of our economic growth, and that means that an energy system in which we don't consume much more energy in the future than we do today is not a good one. So that's a good, important insight. I think the European Union is doing a good job on a lot of these technology agnostic areas. So saying, if you want to provide demand flexibility, for example, you can provide it from anywhere in the system as any type of actor, et cetera. I think in Poland — since you mentioned it — Poland has done very well. Something like energy security has to have a you have to have a long-term view that goes across multiple governments You know, I mean I started working in Poland just after I left NATO in 2006 and since then we've had you know an incredibly contentious political process But through all of that a really consistent commitment to building energy security, you know first diversifying natural gas pipelines then building LNG then building renewables, then building nuclear, now building flexibility in batteries. Energy isn't something you can easily deliver in a political cycle. And then I think the other piece that's important, and I'm not sure which country has gotten this right yet, is recognizing that different countries will do the transition differently. So China has built a big grid first approach to the transition. And we don't hear about it as much in Europe. In the US, we talk all the time about the electron gap, China's ability to deliver electricity to data centers quicker and more cheaply than the US. The US is going to close the electron gap in a totally different way to the way that China has. Europe is going to build a very different system that involves a lot more community choice. All of those approaches are valid. The key thing is to start from a perspective of realism about what actually do we have, what are the problems we really face, et cetera, rather than from a perspective of excessive optimism or taking things for granted or creating obstacles where there shouldn't be any.
Katherine Keddie:
So you mentioned in those examples a couple of times the energy demand from data centers, and also the idea that a future with the same energy demand as we have now is not a prosperous future. How is the huge shift in energy demand that's coming from AI changing policy expectations in this area, maybe creative thinking approaches?
AI Data Centre Energy Demand
Paul Domjan:
Well, I think the first thing is understanding the scale. Right. The way that I would think about it, let's look at a very long-term history. If you look at human energy consumption per capita for the last half million years, it's basically flat. The Roman Empire is a little teeny bump. And then in about 1750, it starts to take off. And GDP tracks energy consumption from most of humanity being dirt poor to our modern world. And I think back to that, you think Queen Victoria was one of the most powerful people in the world, and she couldn't get a mango. And that shows you the way in which energy has transformed the world, that all of us eat mangoes without even considering the effort that was gone to try to bring Queen Victoria a mango in a world without aviation or refrigeration, et cetera. So we've delivered an enormous amount of economic growth as a function of rising energy consumption, and we will continue to. The difference with AI is that when you think about, we manufacture cars out of steel, we manufacture intelligence out of energy. Energy is the primary input to AI compute. So the scale of interconnection is enormous. In many jurisdictions, I gave the example earlier of Poland, you see the same thing in Texas. You know, you have requests for connection of new AI data centers that are 4, 5, 6, 10 times the total energy generated in that place. And I think the UK should be extremely worried as a country about the fact that our interconnection queue is not that big. Because it shows that we're doing something wrong. So the demand is enormous. And I think when I talk to data center developers, their response is, look, yes, I've put in lots of applications, but I will build everything I get. Because we need that scale of capacity to fuel AI. I think that we sometimes think about AI a kind of monolithic way. There's going to be hyperscale at the edge. And that hyperscale hopefully will become more flexible and dispatchable. You know, so that we are matching model training, for example, to the requirements of the system. You know, model training is a great dispatchable load if we have a little bit of programmability. And if you think about chat GPT versus, you know, traditional Google search, chat GPT uses about 10 times more energy. All of that additional energy, almost all of it's asynchronous. So the ability to integrate that effectively with the energy system is huge. We're also going to have a lot of edge data centers. You know, bringing compute closer to people. So we think about data centers, typically from that hyperscale perspective, but the energy system is also going to have to cope with, you know, how do we put a lot more compute capacity into London. Because we have an industry here that needs to use AI. And even if those models are being trained in France with nuclear power, they need to be deployed here. with very low latency in real time. And then I think the last area of that is increasingly there'll be competition around not just how cheaply can we train the model, but how quickly can it learn. So the capability of those edge data centers is going to have to be greater than it is today. And the energy system needs to power all of that.
Katherine Keddie:
So fundamentally, having flexible use of this energy is a foundation for future prosperity. We're all aware of the demand coming from AI. How do you define, in this context, energy security in an interconnected world? What does that mean to you?
Energy Security in a Post-Fossil-Fuel Grid
Paul Domjan:
So I think we need to think about two aspects of energy security. We talked about it briefly earlier, physical energy security and economic energy security. So physical energy security is when I require power, it's available. Power, oil, gas, whatever. Economic energy security is that power is available or that oil or that gas at a sufficiently stable, predictable, mark a determined price that I can grow my economy. If we think about the energy transition, fundamentally what the energy transition is doing is trading reliance on imported fuel for reliance on the domestic energy system. So we're moving away from as much exposure to geopolitics and we're a lot more exposed to grid instability. We saw that in in Spain, we saw it in Berlin, we saw it in Texas. Texas, there was obviously a linkage to natural gas. Spain and Berlin were both endogenous to the energy system. Berlin, I would be very, very surprised if it was not a hostile act. We'll leave it at that. It's extremely well targeted. You're not unlucky in quite such a specific way.
Matt Jaworski:
In several places at once. Hold on. When you mean Berlin, do you mean you're referring to something else?
Paul Domjan:
The blackout.
Matt Jaworski:
Okay, the blackout. Sorry, I was thinking about Nord Stream explosions.
Paul Domjan:
No, no, no, the blackout. So why did the blackout happen? I mean, I don't know if you've got it, but the blackout happened as a result of a fire on a bridge. that happens to be the bridge, which carries a whole range of HV and MB cables, which happened to go to a section of the grid, which is an islanded loop. So if you were going to start a fire on a bridge to try to take out power, that is the bridge you'd want to start a fire on. And a bridge is a really good target, because it doesn't, by definition, have a perimeter. So that was extremely well targeted. If it was an accident, I'm surprised. But yeah, we're trading reliance on imported fuel for reliance on our own domestic power system. And so we need to think much more about cybersecurity. We need to think much more about grid resilience and flexibility. We need to think much more about speed of recovery. We need to understand our electricity grid much better than we do to know actually where are those most critical loads that have to be kept on because that's where the energy security problem is going to shift to. as we move away from what I spent the first part of my career doing, which was really worrying about how do we maintain our geopolitical leverage versus Russia? How do we protect the oil market from piracy and banditry in West Africa, et cetera?
Matt Jaworski:
At the beginning, when you explained your technology in the five-year-old version and then fully, The emphasis was quite a bit on the sustainability. So for example on the pollution from the coal power plant, from the oil power plant like at Tate. And then what we've been hearing so far today mostly was the resilience security angle. So how has this changed? How has this shaped? How have you decided, right, which narrative to focus on?
Paul Domjan:
From our perspective at Enoda, our goal is to resolve the energy trilemma. That's the technological solution we built. Addressing climate change is one aspect of that. If you resolve the energy trilemma, you enable people who are particularly concerned about security to address climate change as well. From our perspective, it's really about choice. What does each society want from its energy system? and removing the constraints to those choices. You know, the energy transition is struggling. We've seen that. You know, I think it's, deal activity in climate tech is down 30% this year. It's struggling because of the perception that those are trade-offs. What we have to do is resolve the trilemma and then open up that opportunity for choice. So from our perspective, you know, what we really, our mission is sustainable prosperity for everyone. Part of prosperity is about being able to pursue what you think is important in your society and not having your energy system dictate that.
Matt Jaworski:
And I think it also ties in with sentiment of you. We've been hearing from investors and other founders in the space that, right, your solution has to have real economic impact on resilience, on lower costs, in this case, better grid. And then sustainability kind of comes as a part of the package. But that's not the only or the primary benefit that you bring in. It's kind of a byproduct of everything being better by a big margin.
Paul Domjan:
When I started my career in oil and gas, and at that stage knew nothing about climate, we still had a very, it's attributed to the former Saudi oil minister, but the saying, the stone age didn't end for lack of stones, and the oil age isn't gonna end for lack of oil. The transition's gonna succeed because it's offering something better. If you heat your home with a heat pump, it's a much better experience. It happens in this country to be more expensive, but it's much better. You should do it because you will actually be much happier and warmer and more comfortable. We're building the biggest utility scale deployments of renewables are in Texas, not because Texas is a particularly, it's my home state, not because it's a particularly pro-climate place, but because It's actually a cheap way to generate large amounts of power, especially if you have a good deregulated system and you don't have nimbyism. So, you know, we're going to move through this transition successfully, not because we have to save the planet, but because we are building the future, which is significantly better in many ways, including sustainability.
Katherine Keddie:
You're about to move into the U.S. more as a company. Obviously the context is a big change in climate policy, but I think that iterates your point that fundamentally it's an exciting market for you. Can you tell us a bit more about what that market looks like for you, what the opportunity is?
Paul Domjan:
without sharing more than I would want to about our plans, I think the key thing about the US is, number one, you have this tremendously deep capital market, which makes it a very appealing market to be in from many different perspectives, which we don't have in the same way, certainly, in the UK. Number two, you have a real understanding and commitment to load growth. Like, if you go to a US utility conference, by about the second day someone will say, can we please stop talking about load growth? We all know that we need to serve AI load. If you say that to, there are a couple of ministers in the UK government who understand that, but very few people. It's just fundamentally a different mindset about what the energy system needs to deliver. And then the last piece is the reliability challenge is in many ways more present in the U.S. Six and a half percent of U.S. homes own a generator. If you add in home batteries, about 10% of U.S. homes have some kind of backup power. So the ability to use flexibility and use better data and use advanced analytics and digital twins and all of the tools we provide with NAM to actively manage the system and improve reliability has a very direct on the ground impact in the U.S. It does in Europe, but not in a way that necessarily has quite the same salience. So from our perspective, there's an enormous amount of attraction in the U.S. market. I mean, the European market is equally appealing. And I think in both cases, what is compelling for us is that ability of enabling individuals and communities to make the energy choices that they see most appropriate for themselves.
Katherine Keddie:
So for the US, it's capital infrastructure, and then also customer awareness and understanding.
Paul Domjan:
Infrastructure requirement, I think, is the way to think about that. There's an infrastructure requirement in both places, but there's more of an awareness in the US of how large that is. And I think Europe, and especially the UK, needs to move faster on some of these things in a way that's a little bit more humble. in the sense of saying, maybe we got to where we are with one set of solutions, but those aren't the solutions that will get us to the next step. And frankly, I know this will make me very unpopular in London with some of the people here, but I respect what President Trump is saying in terms of energy dominance. Energy is what drives economic growth and prosperity. And if you're going to build a successful, prosperous society, you need substantial amounts of energy. China understands that. That's why they're building renewables so quickly. That's why they're building grids so quickly. It's entirely possible, coming back to your AI question, that actually, in at least the short term, China will win the AI race, not because it has better compute, but because it has chips that are four or five times worse, but four or five times more of them. And they can power all of them at once. It's the ability to power them, yeah. So you end up with a... And for the US, I think that tension is quite existential. In Europe, we take it for granted a bit. In the US, under Biden, still under Trump, there's a tremendous focus on how does the not just Greenland, but how do you actually secure the minerals you need and execute the transition? Lobito Corridor Project, domestic mining, mining in Canada, deep sea mining in the Clarion-Clipper Zone. These are all real significant attempts at ensuring that you have an independent, sovereign Western supply chain for critical minerals. We don't do that in the same way in Europe, you know, there's there's some have some things happening in in Zambia, you know They're there certainly are European countries with areas of the clarion-clipper zone But you don't have the same kind of push to begin actively mining nodules like we need to see it in Europe as more as an existential competition that's about our prosperity in the future in the way that the US doesn't in the way that China does
Katherine Keddie:
Fundamentally approaching it in a more competitive light.
Paul Domjan:
Well, recognizing that it's not optional. You know, that we are not going to deliver successful, prosperous, unified, stable societies unless we can meet people's aspirations. And doing that requires energy.
Matt Jaworski:
Our time is about to run out, so before we finish, tell us about your plans for the next 12 months, what's coming up at ENODA in 2026?
ENODA's Plans for the US Market and European Commercial Deployment in 2026
Paul Domjan:
I think I'll highlight three things. First, more trials and commercial deployments in European grids. Really being able to scale and demonstrate what we're able to do in terms of offering a better European energy solution here in Europe. Moving from our current generation of devices to incorporate all of the learnings of the programs we've been through so far to our version one commercial device, which will be manufactured here in Europe. And then entering the US from a commercial perspective, from a corporate perspective, and also from a manufacturing perspective.
Katherine Keddie:
Fantastic. And do you have any shout outs, notices to share with the audience, places they can find you?
Paul Domjan:
Yeah. Well, I mean, in terms of shout outs, I guess I'll pick up a few. So you will have seen last year that we, end of last year, we announced our first pilot in Spain. There's a lot more of those coming. We obviously met through Climate Connection, which is amazing. And if you're in the UK and you're not part of that process, you should be. We have a collaboration with the University of Edinburgh around understanding those flexibility programs. It's one of, and how we can deliver more flexibility in the energy system. It's one of a number of university industry partnerships that we're engaged in around these issues. And then I'd also like to give a shout out to my colleague, Anna Baisley, who is one of the founders of Out and Climate in the UK and bringing that organization from the UK, from Europe, from the US to the UK.
Katherine Keddie:
If you want to find out about Climate Connection, we actually have an episode with Juliet, who is the founder, talking about building co-founder relationships, communities, et cetera. So please check that out if you're a listener. And that's all we have time for. So thank you so much, Paul. This was such an interesting conversation.
Paul Domjan:
Thank you so much for having me.
Matt Jaworski:
Thank you very much for joining.
Katherine Keddie:
And you can catch us on the next Scaling Green Tech. Goodbye.
Matt Jaworski:
Bye bye.
.jpg)
