← Energy Gang
Energy Gang
Electric vehicles create problems for the grid. Could they also help solve them? The plan to turn EVs into reliable grid infrastructure
00:54:29
Loading summary
Transcript64 lines
- [00:00]
A
Today we do not have the capital, we do not have the resources, we don't have the human capital, frankly to go build out the grid as much as all the load growth needs are pushing us to get to. Not to mention that it's causing huge affordability challenges.
- [00:13]
B
No matter how good the technology is. Unless you can do that job of persuasion, you're not going to get adoption, you're not going to get to where you want to be and where you could be in theory, with the technology that you've got.
- [00:23]
A
This is an infinitely flexible resource. You can do whatever you want as long as you send me the right signal and I can do, you know, I can manipulate my behavior and I think that's sort of where we, we have to learn to operate and I think grid operators are also getting used to that.
- [00:40]
B
LNG 2026 is the premier global conference and exhibition for the LNG industry. Taking place in Doha, Qatar on the 2nd to 5th February 2026, the event will highlight how LNG can meet growing demand for energy, drive economic development and and enable a lower carbon future. It will also explore issues including the impact of AI, changing ESG regulations, emerging low carbon fuels including hydrogen and ammonia, and ways to cut methane emissions. To understand where the industry is heading, LNG 2026 is the place to be. Explore the exhibition floor, see the latest technologies in action and hear from the experts. Register today@lng2026.com Foreign. Hello and welcome to the Energy Gang, a discussion show from Wood Mackenzie about the fast changing world of energy. I'm Ed Crooks and on this show we're going to be talking about one of the most active debates in energy at the moment and that's the role that new technologies can play in solving the new problems that electricity grids are facing. To talk about that issue, it's a pleasure as always to welcome back Amy Myers Jaffe. Amy is the director of the Energy Climate justice and Sustainability Lab at New York University. Hi Amy, how are you?
- [01:56]
C
I am good. Busy grading theses and examinations. So nice to take a break from that and talk about the exciting things happening in the industry.
- [02:07]
B
Absolutely. It's that end of semester feeling, isn't it? I'm sure, yeah.
- [02:10]
C
On top of the snowstorm and on.
- [02:12]
B
Top of the snowstorm. Very nice. I was saying we've had a snowstorm in New York City. I always say it makes New York City beautiful for about six hours and after that it becomes really horrible. But just those first few hours, very magical here in New York and That's, yeah, very nice, Very nice scene outside the window right now. It's also a pleasure to welcome for the first time Apoor Bhargava, who is the co founder and CEO of a company called WeaveGrid. Hello, Paul. Welcome to the show.
- [02:40]
C
Hi.
- [02:40]
A
Great to be here. Thanks for having me on. Amy and Ed.
- [02:43]
B
Yeah, thank you very much indeed for joining us. Now we're going to come on to Weave Grid in a moment because that's really the heart of what we're going to be talking about in this show, what the company does, where you fit into this debate over the future of the grid. But before we do that, the first time we have anyone new on the show, something we always like to do is ask them a bit about their careers in energy, how they got started, how they came to the roles that they now hold. So what's your story, Apoorv? What first got you interested in energy and what made you decide to pursue a career in it?
- [03:12]
A
Yeah, well, it's funny. I'm an energy lifer and there's a person on the show that's near and dear to my energy career and that happens to be Amy because she was one of my professors many years ago back at Rice University when I was getting my degrees there. So I had the privilege of having Amy as a professor. I was a chemical engineer and an energy economist by training at Rice. And I just sort of, I saw energy as a foundational opportunity. But problem growing up, I was very fortunate to kind of grow up all over the world and notably grew up in the tiny Mediterranean island of Cyprus. And when, you know, when, when you kind of think about places like that out in, out in that part of the world, you can imagine that the geopolitics of the area is very much underpinned by oil energy. All of those questions of geopolitics meets security meets, meets energy independence or all of those kind of themes of the early 2000s. And so that got me really interested, even in high school and middle school thinking about energy. What is the role of energy? And then energy took me very quickly to climate change. I got interested in the question of climate change and how societies would have or not have resilience. And I just, I sometimes joke. I ended up becoming my own version of Greta Thunberg at the age of 13. I was like, you know what? I'm interested in energy. I can't really imagine doing anything else. It's been an energy journey through and through. I absolutely love the space. I think there's so much to learn and there's so much to do.
- [04:45]
B
Let's talk about WeaveGroup then. So you founded that company in I think 2018. What is your thinking then about what you were trying to do and what does the company do now?
- [04:56]
A
Yeah, so you're exactly right. We started around 2018. My co founder John and I were at Stanford together. John had worked at Tesla for many years and had sort of seen, I think, that that tremendous innovation that was coming, of course, on the automotive side. And if we take a step back, you know, of course we're going to talk a lot about data centers and load growth and all that. But in 2017, 18, early 19, the thing that was very clear was that electrification was going to be something that was going to come. It was coming faster than most people were expecting. Because if you had looked outside of the US and you looked towards China, you saw that electrification was starting to become a serious priority of the Chinese government. And you could see that it wasn't just Tesla, but every other automotive company that was starting to invest deeply in electric products, whether plug in hybrids or full battery electric vehicles. And so John and I started weavegrid with this thesis that we needed a platform. We needed to build a platform that would allow utilities, of course, as well as their customers, to take advantage of the massive amounts of flexibility that sat in electric vehicles. These are huge batteries on wheels that are increasingly being attached to the grid. But the reality is utilities had very little to no visibility on those electric vehicles. They had no understanding of what their driving behavior, and hence their charging behavior was, what impacts that was creating on the grid. And so rather than presuming that EVs were a distributed asset, we actually started with the premise that we first need to see what they're doing and transform them from being a liability, a grid liability, which is something that they really were and continue to be, if not managed well, and trying to transform them into being a grid asset. And so you could see that weavegrid has built what I would say is one of the most purposeful DER management platforms out there. And there's a lot of acronym soup that we'll of course untangle and talk through.
- [06:54]
B
Was just about to mention that. So DER for a start. Why don't just explain what a DER is?
- [06:59]
A
Absolutely, yeah. It's one of my favorite terms that I hate to hate. I love to hate, rather. A doctor is a distributed energy resource and that can mean many different things to many different people. Beauty is in the eye of the beholder. And so for the most Part though, when I talk about distributed energy resources, we're talking about batteries, electric vehicles, sometimes things like thermostats and so on and so forth. The problem with the term, in my view, is that just because something is a distributed asset or a distributed or customer owned resource, does not necessarily mean it is providing value. And so this concept that everything is a distributed energy resource, quote unquote, is actually not true. A lot of these things are devices that we've bought for consumer benefits like heating and cooling, mobility in the case of automotive, vehicle in the case of the automotive sector, and our ability to harness the value versus letting it just become a liability that grid planners and grid operators have to think about, really comes down to how we orchestrate and manage those resources. And so back to weave grid, we built a platform that sat between three stakeholders, utilities being our core customer, automotive companies, where we collect data, we send back signals to their telemetry to manage people's charging, and of course, the shared customer between the two of them, to enable the customer to be in full control of their behavior, to be in full control of their charging, but at the same time to be able to offer up what is a very flexible resource. Right. You think about your charging behavior of an electric vehicle. Most people are charging for two hours every 48 hours. And so there's so much you can do with that giant battery. And of course, the real sort of magic in the platform is that we're actually providing at a very granular level, down to what's known as the grid edge, true visibility, monitoring and prediction, and then management and control of these aggregated resources on behalf of a utility. And I think that's very distinct from yesteryear because traditionally utilities have not tried to leverage DERs in a grid operations way. They have tried to leverage DERs for what's known as demand response. And demand response is a very specific product that, you know, there's been a series of DRMS platforms, Dr. Management System platforms built around where five times a year, when your peak or power plant is really overloading, you just say curtail, curtail, curtail. You send out an emergency signal, you tell every thermostat to shut down as much as possible, heat it up, cool it down, do whatever possible. But it's really not meant to be proactive. It's not meant to be a part of, of your actual infrastructure. It's meant to be a reactive tool that is only meant to work in these emergency moments. And that's totally different from how we've approached the whole problem we came in at saying, what if we could make EVs, batteries, all these other devices a part of the fabric of the grid infrastructure? What if we could make managing them day in, day out as reliable as more copper in the ground and more steel on the ground? And that's a very distinct and different approach. And it's something I'm happy to, of course, dive into more.
- [10:27]
B
Yeah, I do think that's really interesting. And definitely you're going to have to walk us through that a little bit more because that I think is a distinction I still don't fully grasp. I mean, going back to the question of sort of different kinds of business model in this area, one term I think you haven't used yet is vpp Virtual power plant. Controversial, much discussed topic. Would you describe what you do then as operating a vpp?
- [10:53]
A
I would say that Weavegrid runs a platform. Some people call us a derms platform. For example, an edge derms, A true edge derms, A D E R M.
- [11:02]
B
S. So Distributed Energy Resource Management System. Derms, right?
- [11:07]
A
Correct. Yeah. I also hate this acronym. Sue, I promise one of the things that we offer as running a platform on behalf of the utilities to actually understand and then manage those, those distributed resources, those customer owned resources, is the ability to aggregate those resources as a portfolio and use them for grid management. Whether it be deferring actual, you know, upgrades that you need to make on your distribution system, notably around transformers and substations and feeders, or whether it be aggregating them and shifting the energy utilization so that you don't have to turn on other power plants. And when we run an utility owned and operated program, some people would call that a virtual power plant. Now my trouble with the term is not that these aggregations of customer resources don't drive value. The problem is in the term virtual power plant. You know, Ed and I talked about this over a coffee once. There's a favorite Voltaire line of mine about the Holy Roman Empire. It was neither holy, neither Roman, neither an empire. The problem with a virtual power plant is it diminishes the value that this aggregation can deliver. Because actually aggregations of resources can deliver a lot more value on the local system, on the distribution and transmission system, which a power plant cannot deliver. And at the same time, real power plants out in the field deliver a lot of value that customer resources don't have. The telemetry, don't have the dispatchability, don't have the reactivity to be able to deliver. And so it's caught in this awkward Place where it's neither a true power plan, but it's actually so much more powerful than a true demand response product. And I think one of the things that utilities and grid operators, and I think notably distribution planners and distribution operators are really grappling with is how do we value and how do we trust it. And that's something that weavegrid is really uniquely good at because we started with a distribution first mindset where we said instead of thinking about those five events a year on the generation side of the house, we're going to think about reliability down at the distribution edge where every time four cars charge at exactly the same moment, which may only happen once every blue moon, those four cars overloading a transformer causes a transformer to lose half of its life. And that's a really extreme scenario, but it is something that happens and we have to think a lot more about that.
- [13:41]
B
Yeah, it totally happens. I have a colleague, in fact a woman, MacKenzie, who exactly that has happened to where I think with them they can take two cars in their street charging simultaneously. If someone tries to plug in a third EV to charge it up, bang, all the lights go off. Same goes the transmitter.
- [13:56]
C
So it's really an interesting puzzle. And in conversations with utilities and others, it really comes into this domain that everybody pretends doesn't exist on certain level. Because when you get into this discussion about what can I count on or what's a firm asset and what's not a firm asset, people try to make it black and white, but it's not black and white. So even in the traditional VPP terminology, I wrote a substack about this because people say that term and they mean different things. Exactly right. Because demand response is totally different than a battery platform system. And you know, Apoorv, you and I have discussed the fact that some of the really top notch aggregators, it is a firm resource because they put batteries into whether that's an industrial client or whether that's a residential system, they're sizing the batteries larger than the host of the battery needs. And so it's really what we call an oil spare capacity. In other words, you're charging it maybe with solar panels or some other kind of distributive resource, but then when you discharge it, there's a percentage of the hard asset that is only committed to the utility or only committed to a capacity market. And so when people talk about it, they kind of ignore that part. And then they talk about whether or not they can count on a customer to turn down the, thermostat or turn up the thermostat, whichever season it is. But I think your system goes even beyond that in the sense that you're able to solve problems at the block by block level, which, again, the utilities don't have great visibility on. Right, Correct.
- [15:53]
A
Yeah. One of the things that's most misunderstood about the grid, if we really kind of distill down to the simple concepts, is that we have unbelievable visibility on the generation and pretty fundamentally good visibility on the transmission system. But when you get down to the distribution system, it's not great, actually. The lower the voltage on the distribution system, the worse the visibility and the monitoring that a utility has. And that's okay. You know, we've generally run these systems to failure because an average transformer used to cost 25, $30,000. It was fine to run it to failure in a neighborhood, and it lasted anywhere in the order of about 40 years. Well, the average transformer in the United States today is aged at 25 years, the average transformer. And so that means that you've got a whole set, as you can imagine, right? The Median transformer being 25, you've got a whole set that are even older. And now you age it out faster by overloading it with something like an ev, but or sending a bunch of back feed from solar and batteries that are not monitored properly now, you're going to start seeing accelerated degradation, you're going to start seeing higher cost of spending on distribution systems. And boom, here we come to that energy affordability crisis that we're all talking about, which is met with that perfect moment of also load growth coming from AI data centers. And so this challenge on the edge is actually huge. It's really, really, really a massive problem. But to your exact point, Amy, if we can take advantage off these resources and actually build a trust layer between utilities, these DERs, and of course, their owners, the customers, we don't have to compensate people for the exact nameplate capacity in the battery. I actually think that's absurd. If somebody's got an 11 kilowatt hour battery, I don't believe that 11 kilowatt hours is going to be dispensed when a utility needs it. What we want is a firm counting of how much is going to be delivered when I ask for it. And I think one of the things that weavegrid has actually done, I mean, honestly, it stems from my own skepticism of a lot of DERs. And so I built a company to make sure that we can get over my own skepticism and hence help our Utility operator friends also get over the skepticism, which is that these are real resources, they do have real value. And the thing is, we just have to be able to really account for that value carefully. We have to be able to monitor these systems. We have to understand their flexibility. And then we have to be able to dispatch them reliably day in, day out. Not so that they shed a megawatt one day, but then shed 0.1 megawatts another day. That's not reliable. I'd be much happier If I got 0.3 megawatts every day, and I knew that I could rely on that day in, day out. And I think we've built, I think, a lot of that repeatability and that dispatchability in a way where now I've got engineers at Utilities distribution engineers saying, I trust this as good as copper. I'm good. And that's very unique. I do not think that we've seen that from others.
- [18:56]
C
Apoorv where does the company sit? Like, you sell this system to the utility, you've got your own people sitting in the utility. Explain to us a little bit the business model. I mean, I think the function sounds very positive, but what's the business model?
- [19:12]
A
Yeah, so we're an enterprise SaaS platform. It's boring as that sounds. We sit within the utility and we're a cloud software system. So we're not on prem like some of the old sort of, sort of older software systems that utilities have, but we do often hook up to pretty, pretty important internal OT and IT systems within the utility. So it's not just hooking up to, for example, your billing system and so forth, like other players do. I think it's going even deeper into SCADA and understanding their capacity, their hosting capacity, and all of these other things that utilities have, to be completely honest with you. Imperfect information. Right. Utilities are trying. They're in the process of sort of digitizing as much as possible off a lot of those physical infrastructure components out there. They're trying to put more SCADA and more sensors out there. But the reality is that there's a lot of information gaps. And so we take what imperfect information exists and kind of bring that together with our own data and telemetry at the edge and can build a much better understanding of what's happening as people come home, plug in their cars or people have batteries installed in their homes or, or they're doing all of these other things. And then by building and dispatching those resources, building this optimization and dispatching those resources, again, all very much in the cloud. It allows us to build a far more granular optimization, something that really looks at the customer by customer, the device by device. And that's really unique because again, in yesteryear, what everyone used to do and where there was such a huge gap in the market was, was they just said, I don't actually care what happens with Amy's thermostat. I don't really care what happens with Ed's thermostat. I'm just sending a signal to all thermostats and hoping that it sheds a megawatt. That's not reliable. What I need to know is what is happening with Amy, what is happening with that? When I add you up together, what is that reliability factor that I can trust on from each and every asset together? And then how do I turn that into a portfolio that has 99.9% reliability for utility?
- [21:18]
C
And is it system wide or it's literally at the distribution level? Because that's another thing that people debate about, like, am I getting system flexibility, which some utilities say, well, it's great, but it doesn't help me at that particular transformer. Right. So does it help is this localized in nodes or it's. Or it's just, it aggregates system wide?
- [21:41]
A
Yeah, that's a great question. So behind my head you can see a disco ball. And part of the reason is our platform is called Distribution Integrated Smart Charging Orchestration or DISCO for short. And it's a patented approach we built which actually pays homage to a lot of distributed systems that you see in modern cloud computing and so forth. The idea is it is system wide flexibility that starts with a bottoms up approach. So we're going from the transformer all the way to the generator. So we can create absolutely bespoke groups that can solve those challenges down at that localized transformer. Or we can go up to feeders or to the substation or yes, build a portfolio that serves your entire service territory. And so we've got programs with utilities on the east coast and the west coast and the Midwest where we're doing everything from co optimization of wind energy. Over in Colorado, where there's too much wind in the middle of the night, and so we turn thousands of cars on in the middle of the night to ensure that their charging is matching that excess production of wind. But sometimes when you're charging too many cars in, let's say, a place like Boulder, now you're creating feeder challenges because now you've got too much charging happening in one pocket. And so then we can co Optimize that charging so that we're not creating those distribution challenges and we're also sucking up all of that excess wind. So that's really how the system works. It works in an incredibly flexible way for the whole system.
- [23:07]
B
Right. And to go back then to the question of visibility and how you're getting this data from these assets, you're Getting data from EVs in particular, are you? And is that something, do customers opt into that? I mean, if I've got an ev, do I need to choose to be part of the Weave grid system? Weave grid network. How does that work?
- [23:27]
A
Yeah, that's a great question. So we started with EVs because it was the most misunderstood asset, is what I would say the most poorly, the least visible asset on the system. I think there were people who were, you know, as you can imagine, EVs are being bought pretty quickly. There's 7.5 million plug in vehicles today that has nameplate capacity of about 40 gigawatts in the United States. For context, that's more load than what the total data center requirements are over the next few, over the next few years. And so it's a tremendous amount of load that's already being built up and very little of it is being taken advantage of by, in terms of its flexibility and its ability to be useful to the grid. So we started with EVs very different. They also have a mobility component that is far more important to the consumer, frankly, than anything you do. On the energy management side, if I'm paying $70,000 for a car, I don't really care for your $10 a month, unless you're going to make sure that my experience of mobility is not tarnished in any way. That unique asset also has particularly unique attributes on the distribution grid. You know, EVs don't really cause a bunch of problems to. We're already pretty familiar with people's charging behavior. We know that most charging behavior will happen in the middle of the night. But again, back to that experience that you talked about Ed earlier. Two to three cars charging at the wrong moment could cause a localized problem. So, you know, by starting with EVs, we actually built the platform architecturally the right way to start Bottoms up so we could get that visibility. And yes, to get the visibility, the customer has to opt into a utility program where utility is offering you an incentive that allows us to then basically talk to, let's say Tesla Server or Rivian Server or GM Server and be able to pull that data so that we can now have Better understanding and visibility of how your car is charging on that local part of the grid. As we're adding other devices though, for us it's very similar. It's exactly the same concept. And the reason we're doing it, and it's much easier in some ways is because those are already pre established connections and relationships. I mean, your nest thermostat's been in your home for many years. You know, we've got other devices that we've already documented pretty well. Energy management on cars was pretty poorly understood. There wasn't really a standard layer. We'vegrid's been building very much that sort of standard operating layer for a lot of these automotive companies. And at the same time as we're adding more and more resource types, it's allowing us to give visibility to utilities in a much fuller way of what's really been going on with these devices, with these resources.
- [26:03]
B
So it's not just cars then? It is what if you've got home energy storage or for your appliances, anything like that? You can read data from all of those. Yeah.
- [26:13]
A
So as we're now expanding beyond EVs, I think the sort of paradigm shift we're trying to bring to the market is that historically other devices have only been used, I would say, in a 50% of their value sort of way, which is that we've only really used other devices as classic demand response resources. And that's perfectly fine. There is value in dispatching a fleet of thermostats to do sort of doctor Responsiveness in those moments where maybe you've got a peak or power plant that's turning on. But actually building holistic, systematic, wide flexible capacity for the whole system is something that requires you to again, go all the way down to the distribution edge and from the distribution edge up to the generator. And so I think one of the things we're really trying to do, as we think More holistically beyond EVs is starting to say, okay, what is the extra capacity and the extra flexibility and the extra value we can bring to the system by adding these devices that are much more cookie cutter, much more off the shelf, they don't move unlike your ev. I've never heard of a thermostat rolling off a wall and so bringing all of those other resources to build a more holistic flexibility such that you can actually start to say, okay, how do we ensure that? When I'm thinking about whether or not I upgrade my substation, maybe by using weavegrid's platform, I can actually now have more reliable megawatts that every day are being managed in a more intelligent way. And to be completely frank with you, I'm not sure that some of these customer resources can be used every day. There is some amount of customer friction that is felt when you tell someone to dial their thermostat up or down. And so we have to be very careful that we're not overselling the capability of some of these distributed resources. I do think that folks, folks are willing to let go of, let's say, charging behavior because it is a daily behavior, because it is something where you know that as long as your car is fully charged at the end of the night, you're fine. But you know, when you're trying to change someone's comfort levels, you do have to be very careful on how you manage it. And so we do really trust our partners on this. On this. You know, whether it be some of the heating and cooling companies, whether it be the, you know, the thermostat manufacturers themselves, we trust that they know more about the customer's comfort than we do. But one of the things we're just trying to do is make sure we build the fullest amount of flexibility possible for the utility and again build that trust, which I think today has been shattered by years of frankly, poor performance from these resources. And I think there's a way that we can kind of bring up the bar of performance given what we've been able to accomplish with EVs.
- [28:58]
B
LNG 2026 is the premier global conference and exhibition for the LNG industry taking place in Doha, Qatar on the 2nd to 5th February 2026. Under the theme Leading Powering Today and Tomorrow, the Executive Program will feature plenary spotlight sessions with global industry leaders discussing key LNG sector opportunities and strategic topics from across the industry's value chain. The event will host highlight how LNG can meet growing demand for energy, drive economic development and enable a lower carbon future. It will also explore issues including the impact of AI, changing ESG regulations, emerging low carbon fuels including hydrogen and ammonia, and ways to cut methane emissions. To understand where the industry is heading, LNG 2026 is the place to be. By attending, you can join the conversation network with global leaders, explore the exhibition floor and experience Doha firsthand. Register for a delegate pass@lng2026.com.
- [30:00]
A
When I first started doing this and I would talk about climate change, it was like another subject like geology, hydrology, meteorology and it was well received and then at some point it got politicized. What Made climate change political, was the most comprehensive, longest running propaganda campaign in U.S. history.
- [30:18]
C
I'm Amy Westervalt, the host of Drilled, a true crime podcast about climate change. Listen and subscribe wherever you get your podcasts.
- [30:29]
B
Right. So it seems to me, though, you've raised a really important point, which is right up a heart. A lot of these issues around VPPs, whatever you want to call them, the kinds of advanced grid technologies that you're providing, which is this question of control. If you are a utility, a grid operator, you need control. If you've got a problem locally on the grid, across the system as a whole, you need to be able to make changes and you need to be able to make them rapidly in order to keep the grid balanced and avoid having problems across the grid. But if you're a customer, you also want control. You want to be able to charge your car so that it's available when you need it to get to work in the morning. You want the thermostat set to a level that's going to keep you comfortable in the summer and in the winter, you want to be able to keep your food frozen in the freezer, whatever it might be. You need to be able to make the changes that you have to make in order to receive the energy services that you want to receive. And if you can't do that at times because you've surrendered control to the grid operator, that's potentially a big problem for you. And as you say, though, I think there is that issue of trust and the human factor then becomes very important and the psychology of the way people relate to the grid operator becomes very important, and people, I think, need to be convinced that they should trust the grid operator. And as you say, I think there's plenty of examples of people having bad experiences with that and pilots have been tried and haven't worked and so on. And so is there a sense in which the technology can do whatever it likes, but actually the job of winning over hearts and minds is the difficult part. And no matter how good the technology is, unless you can do that job of persuasion, you're not going to get an option. You're not going to get to where you want to be and where you could be, in theory, with the technology that you've got.
- [32:31]
A
Yeah, you know, to be honest, I think, I think too much of the conversation has been around technology. I think we, we have again and again done this thing where we talk about, oh, but the tech is better now, and the tech is this and this and this. The technology is not the issue, to be completely honest. And back to your point of control, I think of control as a spectrum. Ultimately it is also okay if we say that a certain asset cannot be controlled to do anything. Like it is actually an uncontrollable asset. If it is knowable that it is uncontrollable, that is also okay. And then there's also the other end, which is that this is an infinitely flexible resource. You can do whatever you want as long as you send me the right signal. And I can do, you know, I can manipulate my behavior. And I think that's sort of where we, we have to learn to operate. And I think grid operators are also getting used to that. I mean, taking a big step back. The thing we haven't talked about on this show yet are about data centers, right? And I think for the most part almost all data centers, and I'm sure some people will, of course, you know, there will be isolated cases here and there which will be different. But for the most part today, most data centers want to run flat out. They want to just run flat out and they do not want to actually have conversations about flexibility. And you know, hopefully people will prove me wrong in that. But that to me is actually a knowable fact that as an operator has. They know that I don't have control over this. So it's just good to know that. And then that allows me to plan my system and what resources I'm bringing and what capacity I have to build out and what infrastructure I have to deliver in a known way. It's similar with distributed resources too. If I know that my portfolio of, let's say EVs or portfolio of batteries and EVs or whatever else is going to deliver me this shape of flexibility, this kind of megawattage. You know, we've got hundreds of megawatts on our platform, right, which we can easily scale to like a gigawatt plus over the next year. The way I look at it is I don't care about the number of megawatts and gigawatts entirely. I don't just care about that. I care about the time and the location of those megawatts and gigawatts and being able to show those shapes not just to, oh, this happens one off, but actually showing it to 95 and 99 and 99.9 percentile moments and building that trust with operators is very different. One of the things that irks me no end is when technology providers turn around and say, but look one off, I managed to make this load drop happen by 50 megawatts. Well, that's great. I can also one off, convince all my friends on my neighborhood to wear a blanket and put on a sweater and to not charge their cars. And in those extreme moments, I can also basically go to zero. But that's not realistic. What I want to have is a realistic distribution of understanding as an operator of what is going to happen when you manipulate the behavior through automation, through all of your intelligence that you're doing on the edge. And then that allows me to plan how much I should or should not build out with how much extra headroom in my system. And that's a really important place to be right now because today we do not have the capital, we do not have the resources, we don't have the human capital, frankly, to go build out the grid as much as all the load growth needs are pushing us to get to. Not to mention that it's causing huge affordability challenges. And so I think it's really exciting. I think for that reason, tools like ours are actually stepping up to the moment and are saying it's not just about the tech, it's also about the trust we've been able to build with customers and with utilities, of course, and kind of step into the moment and show off what we can do to put that downward pressure on rates.
- [36:09]
C
Well, and I mean, I think the point also to the point here is that the utilities can't actually control what consumers do. And so especially at the residential level. But even, you know, when you were talking about the tech companies and other kinds of big industrial users, and so, you know, when people feel like it's not reliable, they go on their own, you know, and they're going to put in a solar system, they're going to put in storage. And then, you know, I was talking to some people from the trading side, right. And you know, the problem we have on the trading side too is you've got people gaming certain markets. And so someone was telling me an interesting story where they were studying trading markets and they found that this couple of big traders that had been moving in and out of markets don't go to Vermont because Vermont has used all these assets, has visibility within Vermont, not necessarily the wing of ISO, but within Vermont. So, you know, super small market and they've completely eliminated the duck curve. And so there's nothing to arbitrage anymore. So no one's trading around in that market.
- [37:24]
A
Yeah, I think, you know, this is one of the other challenges in our sector right now. And I think particularly as somebody who grew up in Europe. I still have sort of a. I do a little bit of discounting of a lot of the ideas that come across from Europe because a lot of European startups and a lot of European companies come across and say there's so much value in energy arbitrage, there's so much value in energy trading this, that Europe is a totally different market. Right after Enron, they were the only ones that actually deregulated outside of Texas, whereas the rest of the US actually stayed very regulated as a generally bundled entity. And I think one of the things that's interesting is ARB arbitrage, right? It's, it's a short game. You know, in 2023 in Texas, the arbitrage value was so high in ERCOT that every battery manufacturer and every battery trader went out and said, I'm going to make a killing. And then you look at the prices today. You can, you can barely eke out a living right now by trying to play the ARB game, right?
- [38:22]
C
Hey, listen, nyu, you know, I'm doing. I have a bunch of students working on this whole. These VPP statistics state by state, and then also working with the SADI data. So SADI data tracks total minutes of electricity outage by average customer per location. You know, we're working with the annual data, but you know, you can get a little more granular. And the outage data for Texas, you know, 2020, 2021, 22 was out of sight and now it's tiny. And you can really attribute that. Cause you know, those same traders are still there, but you know, you can really attribute that to their reduced opportunity because the outage data for 2024 and 2025 is so small because a lot of batteries went in. And then also, you know, you get NRG doing a VPP 1 gigawatt, you've got a lot of other VPP players. It's probably the biggest state or one of the biggest states for vpp. And so there's a lesson there.
- [39:28]
A
Arbitrage is generally a game that you should leave to traders and maybe utilities to go figure out. I don't want consumers holding the hedge risk off things like arb. It's just too complicated. And I say this as somebody who spent my whole career at energy. What I think is different and unique about weavegrid's business model and how we try to work with distributed resources, right. Is we're not trying to be an aggregator playing in power markets and pretending that we're going to be somehow better at that. We're actually trying to aggregate resources on behalf of the utility and actually arbitrage against infrastructure buildout. And what I mean by that is infrastructure is going to take decadal time horizons to build out to the level of load growth that we're anticipating in this country. Localized at the level of distribution, transmission, and of course generation too. But generation, infrastructure gets built way faster than you think. Right, and that's why ARB goes away. However, poles and wires, transformers and substations, that takes time. We don't even have a supply chain for enough transformers today. You know, the prices have gone up so much today. And so our ability to unlock capital flexibility using distributed resources requires us to bring up the bar of visibility, monitoring and control of those devices to a whole different level than what we've done before. Because historically all we needed to do was say, I'm going to hit a button, hopefully a couple megs show up and I don't have to worry about it beyond that. But now it is different. We really are trying to say how do we defer spending more money on capital? And guess what? Utilities are the ones saying that because they have to make a choice between do I provide the infrastructure to that data center and do I, or do I go and upgrade parts of the cul de sac where people live with EVs and batteries and all those things going into homes. And, and the trade off is real because it's economic trade off. And I'd much rather be upgrading the power lines to ensure the data center comes online while keeping costs low for everybody else. And so I think it's a fundamental shift in the business model of flexibility.
- [41:29]
C
But, but they're making people on the cul de sac really, really angry. So they, they're having to balance that. You got to, you know what I mean? So, because, right, because rates are, and they want rates to go up like they're making big money. Their stock is way up, you know, I mean, almost, you know, universally around the country. Well, not universally, but in a lot of key markets. But on the other hand, it's a lot of political pressure now.
- [41:52]
B
Right, okay, so let's talk about scale and speed of deployment then. Because it seems to be, then if you're talking about offering material solutions to some of these problems, you're going to need to get to very large scale pretty quickly. I think you were talking earlier about being in the hundreds of megawatts scale at the moment. U.S. demand is projected to grow by, what is it, 50, 100 gigawatts maybe over the next five years, as you say, because of data centers largely driving that very rapid growth. Is it going to be possible for platforms like Weave Grid and similar companies really to make a material difference to the extent to which we have to have a physical infrastructure builder to meet that increased demand? Or are you really just kind of of a scale chipping away at the edges, maybe making a few little savings here and there, but not really fundamentally changing the structural equation?
- [42:52]
A
No, look, I think, I think that's, that is, that is a question that we're going to keep figuring out through the economics of these, of this build out, and we'll see how much of that demand is real and we'll see how much of that plays out over the next few years. Like projections are projections. At the end of the day, what I think is that better utilizing the distribution grid, from the about 25% utilization that it has today to close to 50 to 60% is something that Weave Grid can facilitate. And I think that Weave Grid can pull other aggregators, notably aggregators that sit outside of the utility fence line, into the fold to actually ensure that instead of again, them just dispatching their resources for those few RB moments, arbitrage moments, you know, they see out in the power markets if we can actually better utilize their flexibility to better, you know, increase that utilization of the whole system. You're now talking about taking the same hundreds of billions of dollars that have been deployed, but spreading it out over a heck of a lot more kilowatt hours. And so there is an opportunity, no matter how many gigawatts of extra demand shows up, to actually keep driving down the cost of the fixed infrastructure. And that's, I think, a given. Now, do I think that we're not going to need to build out more infrastructure? Of course not. I mean, again, I'm an engineer by training. I believe in steel and concrete and copper doing its job. I think those things have been tested a lot more, frankly to, to the limits. But at the same time, I think it's, it's just not possible anymore to assume that we cannot use other resources because we're just going to cause cost increases of a kind that consumers will not be willing to bear. And so we're going to get to a point right where the AI data center folks are going to have to make a choice. Do I continue to invest in the US and do I continue to see the political backlash, and do I continue to see election cycles going against politicians who aren't doing anything here? Or can we find some sort of compromise where Utilities are investing in these solutions and then on the utility side I think they're going to have to really kind of make sense of what I think is a loud and sometimes chaotic and confusing landscape. Some people claim that they can do XYZ magical things with their VPPs. Other people claim that they've got a full suite of solutions that can do everything under the sun. I think the reality is we've just got to get to reliability. I want utilities to use these resources because they have tremendous potential, but I don't think that we should ever over promise on these things. And I think the more trust that utilities get with it, the more they'll want to scale. That's at least been our success story so far. We went out, we started with pilots where utilities said wait, I want to trust the technology. We checked that box. Then they said I want to trust the value. Can you deliver as much value as you talk about? And then we checked that box and then we took it back to regulators and regulators said, okay, let's go from pilots to programs where we can really see these things scale. And that's where we're starting to see the next set of challenges, which is that you need to integrate with a whole series of other systems and so on and so forth. And that stuff takes time too. But I think the ability to scale from hundreds of megawatts to even 10 plus gigawatts is far more doable than frankly waiting on some magical fusion to show up in the next couple of years. And I think that is these are all known problems in some sense. And I think that's why I have a lot of confidence in our scaling.
- [46:08]
C
So I have a futuristic issue to bring up because apoorv part of the reason that I called you is I was at a dinner and we were in a heated debate about VPP and I said, well I was at some conference keynoting and somebody came up to me at the sidelines and said you really need to know as we go AV from the ride sharing world, they are going to have a business model where they're going to have a whole fleet of AVs and then they're going to know that they can arbitrage between mobility services and plugging all those batteries from those AVs into the grid all at one time and providing a flex battery system for a distressed grid or a capacity market or whatever. Like an hour, you know, had to be an hour ahead market, couldn't be a two year ahead market but you know, some kind of shorter timeframe market. And I was telling that story, and everybody said, well, you know, you're the one who knows the poor. You should talk to him, because he's the one that's closest to being able to organize that. And so, you know, let me ask you, like, do you see that in the future?
- [47:26]
A
Yeah, I think. Okay, let me take a bit of a step back. So. So, you know, there's this. There's this term that's become very popular recently when we're talking about data centers. It's this term known as the bit what spread. What is the value of a bit relative to the value of a what? And I think as somebody who believes he's probably been closest to looking at energy systems and mobility systems in the same framework, I'm going to coin a new term, which is what I call the mobility watt spread, which is how much do you value your mobility versus how much do you value the watt that powers your mobility? In this moment, I can guarantee that each and every one of us on this podcast has probably paid a lot more for an Uber or a Lyft or a taxi, and definitely for a flight than we have our average monthly electricity bill. And so we do value our mobility quite a bit, is where I want to start this conversation. What I would say is, as we move to a world of autonomous vehicles, what they're building essentially is a portfolio of mobility. So they will have a better understanding of the supply and demand of mobility needs in any city, in any region. And at the same time, they will also have a better understanding of the. Of the sort of downtime of that fleet and also the auction value of that downtime. And I think one of the things that they will have to weigh is, is the mobility value of even the option of having your car ready to go, rather than draining it off its batteries. You know, if its battery state of charge, is that option value more than the value of the watt that it can provide in a market or in some sort of flexibility program. And what I would say today, based on the modeling, based on the data we have, based on the tens of thousands of vehicles and other devices increasingly we have in our programs, is that today it's very hard to see a pure arbitrage in power Market's case for an AV fleet to make more than incrementally few dollars in their total cost of running the or total value in their business model. However, if there is an opportunity where somebody says, look, Waymo or whoever, right, Zoox, instead of upgrading this substation, if we can actually curtail Your charging by, let's say, 10% on these five to 10 really hot moments or really tough moments in the year. And that means that your fleet might not be fully charged, but might be 75% charged. We can save you two to three years of upgrade time and we can actually offer you that in lower energy cost. I think they'll take that. And so that's where the opportunity, I think comes in again, which is the AVS actually saving infrastructure time and money, which is where I think they will become curtailed versus turning them into a giant aggregated power plant which are trying to bid into ERCOT or ISO New England or something like that. Because I'm just not sure that the value is quite there unless we frankly start seeing energy prices and capacity or energy markets going up another 10x. And so that's where my theory comes about right now, Amy, but I'm happy to debate that over another coffee.
- [50:40]
B
Yeah, definitely a lot more to talk about on this one. Certainly the future of avs is fascinating. I think we're just about out of time though, unfortunately. We're going to have to leave it there. Just before we go though, just time quickly for our free electrons, personal items that we've brought in. I'm going to start myself. I just wanted to flag up the film Eddington, which I just watched. It was out earlier this year, really interesting about data centers and public resistance to new data centers getting built. Very much a film that speaks to our present moment. Pretty grim and depressing, actually, as I take on that, I think, but also, as I say, pretty prescient in identifying data centers and the way the public feel about them as a big issue in America today. And I would say definitely, if you're interested in that issue, it's a film that's well worth checking out. Amy, what's yours?
- [51:33]
C
Mine, Ed, kind of comes from sort of the same thing, but a little bit differently. So a couple of years ago we had a mockumentary like a comedy series that was a film about aliens coming down from outer space. But now apparently you can go to Amazon prime and there's a real documentary that is quoting like real government officials, you know, ex CIA. Apparently Marco Rubio is featured in this documentary, which I haven't had a chance to watch yet, about how the government is hiding from us the fact that there are alien spacecraft that are coming down to Earth. And so this time of year in my cleantech class, each student is supposed to do a report, like a Wall street style report, about a potential sector and one of my students, I guess, kind of was watching this documentary. And so he did his presentation on space solar and whether or not we've come to the moment. Caltech has apparently had some breakthroughs on how you would manage to put some equipment up in space and then beam down solar energy down to Earth and have it be constant and not ever go off like so the sun is shining up there in space 24, 7. So are we at the moment in time when we're willing to consider space Solar and have 24.7Solar. And so I, when he was making his presentation, I showed him, you know, that Florida company that has a data center on the moon.
- [53:20]
B
To be fair, they don't, when you say they have a data center, they don't notice. They don't actually have one. They would like to have one.
- [53:27]
C
Well, it's this little, they put this little data thing up on the moon. Yeah, they did from St. Petersburg. They put some little.
- [53:36]
B
Yeah, data center is a big term, but. Yeah, go on.
- [53:39]
C
Right, yes, yeah, not a data center like, you know, we have here down on the ground. But the interesting thing is thinking about all of this is you have these big tech companies like SpaceX talking about going public and you have other big tech billionaires investing in space travel. And originally I think they were thinking about bringing minerals back that would solve the minerals problem. But now, now there's this new interest in could we solve our data problem in space, whether that's space solar or whether that's space data center, et cetera. And so that is my free electron because it's front of mind because I had a student say that it might be a business finding its time.
- [54:30]
A
Yeah, that's pretty fantastic. Especially with all the conversations increasingly now about Data centers in SpaceX like we do.
- [54:37]
C
And then what, what would happen? What would happen if there are like other civilizations out there, how are they going to feel about all these gadgets, you know, that we're going to put.
- [54:46]
B
Out, just throwing it out, their videos from Amazon prime to find out about life on Earth. Indeed. Yeah. Anyway, Paul, what's yours?
- [54:55]
A
Yeah, well, mine, mine's probably a little less, a little less data center heavy. I'm somebody who loves salt. And so I'm going to give, I'm going to give all your listeners a book recommendation which is Mark Kolinsky's Salt A History of the World. And I love it because I think one of the things we all miss is how much the history of salt. I'm giving away the, I'm spoiling the ending here. But the reason we have salt in abundance is because of energy and the fossil fuel revolution that sort of enabled us to have abundance of salt, which went from being a completely, you know, difficult to access commodity, something we fought wars over and so on and so forth, to now, I think this really interesting moment where the salts are actually coming back to enable abundant energy because lithium and magnesium and all these things matter a ton and so.
- [55:50]
B
And sodium, of course. Sodium batteries.
- [55:52]
A
Sodium. Sodium batteries, Exactly. So I'm going to go recommend you spend your Christmases and holidays reading about salt. It's a pretty fantastic book.
- [56:03]
B
Yeah, that is a fascinating point. Something I've never thought of at all. But the energy salt nexus, as you say.
- [56:09]
A
Energy salt.
- [56:10]
B
Yeah. So we do unfortunately have to leave it there. But many thanks, Amy.
- [56:15]
C
Thank you, Ed, for a great show.
- [56:16]
B
Yeah, many thanks, Paul.
- [56:18]
A
Absolutely. Thank you all and excited to have been here. Thank you for the invite.
- [56:23]
B
Yeah, Thanks a lot for joining us. It's been great talking to you both. Look forward to talking again soon. Thanks to our producers, Stuart Duffy, Toby Biggins, Gilchrist and Dan Cottrell. And above all, many thanks to all of you for listening. We really value your feedback, so please do keep that coming. And we'll be back soon with all the latest news and views on the future of energy. Until then, good.