The Exchanges

Q Last week, your guest did not mince words of not thinking direct air capture was going to be a solution, uh, for climate change. What's your view on the potential of direct air capture?

A You know, I think, uh, I've read the IPCC reports like everybody else and recognize that there is no realistic pathway for us to get to one and a half degrees Celsius of, of temperature rise or even really two degrees without substantial carbon, carbon removal. There's just no way we're going to do it. Um, And substantial being gigatons. So, and, you know, within carbon removal, there are a bunch of different approaches, as we've talked about, uh, but most of them are difficult to scale as well as you could scale a direct air capture system, just because you can literally put it anywhere. You're pulling ambient air, um, And you are sequestering CO two. And then as long as you can store it somewhere permanent, then it is permanent removals. So there is, you know, I think the fun, the first principles premise, which is why so many people are interested in direct air capture is basically, we need a lot of this. There's just no way, no way we're going to solve this problem without a lot of it. With that said, I think that it is not going to be a smooth path from today to 10 gigatons in twenty-fifty or whatever the number is going to be. We're going to have, this is, you know, a market that is extremely nascent with technologies that are mostly pretty unproven. They are expensive, like very expensive by any real normal standard today, and they have a pathway to get cheaper, but even…

AI assessment note: “So I'm sort of somewhere in between. I, I, I'm, uh, excited”

Q having better conversations about this, and that the industry doesn't necessarily see that as threatening. I mean, thinking about your episode last week about Um, supply chains for minerals, for rare minerals, for batteries. It seems like we're having these conversations a little bit more. So are we getting better about talking about these bottlenecks, or are we just not fully recognized the ones that are coming down the pike?

A I feel like we, we wake up to the next big problem, like, collectively a little bit later than we should. Why are we talking about that? Why is the world talking about battery minerals? Right now, because it is a current shortage. We are in a, you know, actual supply crunch right now. There were some people who were, and including Kurt from Kobold, who was on last week, and a bunch of other people who were saying this is going to become a bottleneck, you know, five or 10 years ago. But the sort of public consciousness wakes up to it when it becomes An acute crisis. You can say the same thing right now for fertilizer, for example. Uh, and so I think we will, we collectively wake up to it a little bit too late. Um, but there's a small, always a small group of people who are sort of looking one step into the future and saying, well, if it is true that we are going to do X, then there are a bunch of knock on effects of that. And we need to watch out for all those challenges that are going to arise.

AI assessment note: “we wake up to the next big problem, like, collectively a little bit later”

Q Okay, next question from at Winston Reed. From an investment perspective, what are your thoughts on policy-resilient climate tech solutions, i.e. market-driven, versus policy-dependent tech solutions in the current environment?

A Well, um, I don't know, I'm, I'm always thinking of it from an investor perspective. Obviously, all things equal, you want policy-resilient solutions, like, you don't want to be reliant on policy if you can avoid it. I think the interesting thing, there's sort of a couple of categories that have emerged fairly recently that I think ultimately are policy dependent, but may have a near term world that is policy resilient. So the obvious example of this is carbon removal, where I think anybody who looks at this would, would admit that in the longterm, this market does not scale without substantial policy intervention. Like this has to become mandated in some fashion or another. Uh, otherwise there's no way we're going to get to gigatons of removal. But in the meantime, there's enough demand from the private sector that is policy resilient or not linked to policy that you can like jumpstart the market. So there's a lot of investment going into that. I didn't get the risk that, that those companies are taking is that the policy does eventually show up at large enough scale. Some people might tell you, you don't need policy there. I don't, I don't, subscribe, subscribe to that view. Um, but generally speaking, I think things that are, you, you want things that have the, um, ultimate pathway to become policy resilient, right? Like, Someday this will just be better, or someday this wil…

AI assessment note: “all things equal, you want policy-resilient solutions, like, you don't want to be reliant”

Q And that, that sort of, you know, magnifies, or at least, um, emphasizes how acute this challenge is, right? Because they're saying, like, we're desperate, we're willing to pay if you can deliver it to me next year, or in two years, and they're not even able to get it, in many cases, get power on those kind of timelines.

A Right. Ok, so, High level, you know, the world has woken up to the gauntlet in the past six months. Some folks were already awake to it, but certainly now it's much more prominent. Let's talk about, like, what is starting to happen as a result of it. I mean, you mentioned one thing already, which is that utilities in these pockets and these locations where it's most prominent, which tend to be the data center regions, the existing regions, are reckoning with, like, Totally unprecedented and kind of unpredictable load growth and having to figure out how to deal with that from a planning perspective. So, like, tip of the spear on this is, is Northern Virginia and Dominion Energy, which has been the biggest data center region historically. Like, they're already completely overrun, um, and it's, it's actually hard to predict how much more capacity is, is possible in that region. But then there's these secondary regions behind that, um, Georgia Power, um, Issued an amended IRP a year after the first one. You see this in places like Phoenix with APS and some others. So utilities are, are sort of getting hit with this one by one, some more than others. That's one impact. Let's talk about the impact that we're starting to see as utilities are being forced to reckon with this new load growth. What is it that they're planning to build? Because this is one of the This is one of the side e…

AI assessment note: “utilities in these pockets... are reckoning with, like, Totally unprecedented and kind of unpredictable load growth”

Q will be on a node, but we've simplified it. NP-fifteen. Come up with the optimal schedule for next day or the day ahead market. We're going to put aside all the complexities around market bidding and trading, and the idea is to take this energy storage asset and come up with a good use case for it, For tomorrow's market. How do I use it for energy and ancillary services?

A And to be clear, Basically what needs to happen, you laid out a list of acceptance criteria for this, for this code, and basically what needs to happen is that the system, the code needs to be able to pull day ahead energy and ancillary service prices from CAISO. It needs to be able to incorporate sort of various factors, load patterns, historical data, maybe weather data. It needs to process that information alongside the parameters of the battery, which you've set out. You said 25 megawatts, four hours. You also gave us a cycle limit, roundtrip efficiency, minimum state of charge, all the sort of key parameters of a battery, uh, and then it needs to have an algorithm to optimize the schedule for charging and discharging that battery for the next day, print that schedule, and forecast the revenue. Is that basically right?

AI assessment note: “needs to have an algorithm to optimize the schedule for charging and discharging”

Q the credit. So in that world, um, uh, you had a enhanced set of governance, like the, the, the stuff that Vera put in place and gold standard, you'd say it's better than the CDM stuff, uh, in terms of like the guardrails, the audit, et cetera. Uh, but the bar was very low to begin with, right? So, um, uh, it did not solve the problem in its entirety.

A No, and I will say that that world, though, I think you're about to say that we were in a new phase now, That world definitely still exists. Like, that is ongoing. As an example, I was on a flight yesterday, and if you go to the Oakland airport right now, there are signs that say, offset the CO₂ associated with your flight, two dollars for every thousand miles that you travel, and there's literally no additional information. Tells you nothing else about what you're purchasing. It has your little QR code you can go to to, to make the purchase, right? That's very much, in my mind, in the, in the world that you're describing, of sort of like, It's, it's storytelling.

AI assessment note: “No, and I will say that that world, though... That world definitely still exists.”

Q Okay, so you brought up fertilizer. We got a few questions around the nexus of energy and food, so I'm gonna jump to a question from at Kawasaki Steven, which is, given the war in Ukraine's impact on fertilizer food supply chains, Can you share more about innovation in the climate tech sustainable fertilizer space?

A Yeah, I mean, fertilizer is, we're in a literal global crisis. Um, it was even prior to Ukraine, we were in a pretty tough spot, right? The fertilizer prices, um, are driven in large part by natural gas prices, and so natural gas prices, particularly in Europe, had already been up prior to Russia's invasion of Ukraine. You had already seen some of the biggest fertilizer manufacturers in the world, companies like CF Industries shutting down Haber-Bosch plants, which is a big deal because there are only three hundred-some Haber-Bosch plants in the world. So all of the ammonia that we produce basically today comes from like three hundred-some facilities. So you shut a couple of them down. That immediately has fairly significant ramifications. So we were already seeing record high fertilizer prices. Then of course, Russia invades Ukraine and both Russia and Ukraine, um, have a big role to play in the global fertilizer and sort of downstream of fertilizer market, as well as obviously the impacts on natural gas from Russia directly. So it's become, this is another one where like we're in a really, really dangerous crisis right now. And the global South is feeling it much more than We are certainly, though we're seeing it in rising food prices already, and we'll continue to. Uh, there has been a bunch of really cool innovation in, um, reducing the need for nitrogen fertilizer, uh, or …

AI assessment note: “there has been a bunch of really cool innovation in, um, reducing the need for nitrogen fertilizer”

Q All right. Nuclear industry, you're on notice. Coming at us from Benjamin Tank, having been through a first clean tech boom and bust cycle a few years ago, where would you say we are currently in this climate tech cycle, and what do you think happens next?

A We're definitely, well, okay, let's geographically constrain this first. Um, Because I do think it is pretty markedly different in the U.S. versus in Europe, for example, other parts of the world. But presuming that this is predominantly a U.S.-focused question, though I know Ben and he lives not in the U.S., so apologies, Ben, that I'm going to make this kind of a U.S.-centric answer. Um, we're, we're clearly in the bust part of the cycle. I think it's been, one difference between this time and last time in my mind is that the fall from, from glory has been a lot steeper and faster, because it was really driven by a political change, the, the change in administration, and then the, the swiftness with which that new administration has really gone on the attack against a lot of climate tech stuff, um, Has, is different. Last time around, it was more of a slow roll, right? Like nothing precipitated the bust quite as much. It was just sort of a series of failures and lack of capital and so on. I guess you could say Solyndra last time around, but I don't actually think that's true. It was just like, that was just like a marker of, and it was an example. It wasn't the cause. This time it has been pretty steep. Um, So that is different, and that's probably more challenging, because I think a lot of companies sort of feel like they had their rug pulled out from under them. Um, you kno…

AI assessment note: “we're clearly in the bust part of the cycle”

Q All right. Tesla, you've got your marching orders. Let's talk for a second about plastics. So a question from Justin Brody Comet. Plastics are one of the cheapest materials on the planet and one of the largest uses of petroleum outside energy. How do we decarbonize our plastic sector, Shale?

A Okay, so I'm, I'm not the expert here, and I'm gonna wait for a bunch of listeners to tell me the ways in which what I'm about to say is, are wrong, but I think there is a common misconception where people confute, conflate a couple of things. One, they conflate the use of petroleum with an emissions problem. Right? Which it's not. Burning petroleum is an emissions problem. Using petroleum is not necessarily. So plastics do contain petroleum and hydrocarbon and stuff like that, but, but except to the extent that those things are burned, um, they do not create emissions that cause climate change. The second thing that I think people conflate is a broader environmental concern with plastics with climate. There are many reasons from an environmental perspective that you might want to care about plastic, uh, You know, pollution of the oceans, microplastics, blah, blah, blah. But those are not inherently climate problems. To my understanding, the majority of the, the reason that plastic is, has a lot of embedded emissions is largely in the, is largely the energy consumption of the production of the precursors. So like the production of ethylene uses, is, is steam cracking of natural gas predominantly, and that does produce a fair amount of emissions. But that's a very specific thing, and so you can You could solve that in a bunch of ways. You can introduce new mechanisms of producin…

AI assessment note: “You can introduce new mechanisms of producing, producing ethylene. You can electrify that process”

Q people are willing to pay for the service being delivered is, I think, an open question because they're, you know, they're charging, you know, open AI right now is just a major, it's hemorrhaging cash like you can't believe, right? There's no obvious way that they become profitable Without a massive increase in the demand for their services and the ability to raise the cost, and that is really unclear.

A Yeah, I mean, setting aside the model hallucination stuff, because I think there's definitely an argument there, but there's also a counter-argument of like, well, these models are improving sufficiently, we're doing better and better at removing the false answers and things like that, but the, the, will the revenue show up To match the capex is a big question. My, uh, my friend David Kahn at Sequoia wrote this piece a while back called the six hundred billion dollar question, which is probably now like a two trillion dollar question. I don't know what the number is today, but it's basically like that's the amount of capex that has been announced to spend on, on AI data centers. And so that is the bare minimum of the revenue that has to show up to support that. And it, we're, we're nowhere near that level on that side. So there is a bit of a leap of faith that the market is taking there. And so I think You know, again, it comes down to this on the demand side alone, and then I definitely want to hear your view on the energy efficiency question. Of course, it is very difficult to predict what is the actual demand. If you remove other constraints from the equation, what is the actual demand for AI compute? It does seem clear to me that at a Lots of new applications, model improvement, all sorts of things that are gonna lead to lots of additional demand. How high that goes, though…

AI assessment note: “will the revenue show up To match the capex is a big question.”

Q going to accrue value in a market perspective? Uh, and what does that mean for data center construction? You know, does it mean that if everybody can go do a very lightweight, but very effective model, uh, Are they going to go back to hosting it in third-party data centers, or is it going to be, you know, is it going to be meta that runs everything on their own?

A Well, that and the scale of the individual data center is an interesting, from an energy perspective, is like a very salient, very important question. So the, the trend line has been, okay, we need to train bigger and bigger models. These models need to be in one site. So we need the data centers themselves to get bigger and bigger and bigger is why we're talking about gigawatt scale and Stargate, who knows, you know, like, Multiple gigawatt scale individual data centers. And that itself is becoming harder and harder to cite from an electricity perspective. Data centers as circa, you know, five, seven, 10 years ago, you'd see lots of these like 20 megawatt scale data centers. Not tiny, but, but way, way smaller. If indeed you get the sort of optimal performance off of an AI model, if you can build it at 20 megawatt scale, it unlocks a very different set of opportunities for where you can cite this stuff. And how you can power it. So that, that's like an important question here. Does the results of this efficiency mean we can build 30 X bigger or more powerful models on gigawatt scale? Or does it mean we're building this level of performance, but at 20 megawatt scale?

AI assessment note: “the scale of the individual data center is an interesting, from an energy perspective”

Q be that I could connect a customer in two years, and now it's seven years. And that's, that's okay in sort of the utility world, right? Now, it's not ultimately in their best interest. They, they, they should want to go faster and get more customers and, but their business model is not designed in such a way to operate in this sort of theory of constraints sort of world.

A I think it, I think that's, Mostly true, though. I think people, I, I know you know this, but a lot of people don't appreciate all of the constraints, not to reuse that term in different contexts, that utilities have on them, and so they're, they're thinking about, like, yes, build, you know, they're incentivized, generally speaking, to invest in CapEx, to build, spend CapEx, and earn their regulated rate of return on that CapEx, and then minimize costs so that the, the profits they actually gain are as high as possible. Um, But they are subject to regulators who, you know, are unique and specific and have different things that they require. They also, you know, have to place the highest premium on reliability and resiliency and all these things. It doesn't necessarily cut entirely against what you are saying, but I think what my experience of watching this all play out over the last couple of years has been, okay, a bunch of Huge customers started showing up on utilities doorsteps faster than they ever had before. Utilities Started by saying, okay, I'll put you through my normal process, which led to, okay, lead times are getting longer and longer and longer, but they're adjusting to that now. And so I'm curious what you think about, like, we, we've seen some of this publicly there, there was a settlement recently. I can't remember which utility in MISO territory, um, between …

AI assessment note: “I think that's Mostly true, though. I think people... don't appreciate all of the constraints”

Q theory, you could have a photocatalyst, but it would have to, this is where some of the research questions that we're going to get to in the report, you know, it's just what that, the, some, some of the, the, this is like some of the material science, right? What is it About the catalyst that is going to make it be effective in ambient UV conditions at ambient temperature.

A I've actually, this one I've seen, I've seen a couple of startups, um, proposing this exact thing. Surface treatment with some kind of spray-on catalyst that you put either on wind turbine blades or something else that, in principle, would oxidize methane, um, in the atmosphere. You know, I think there are, it still seems clearly very early in terms of proving it out. And this one is one where, like, you would have this, you know, you'd get into the realm of some of the similar processes MRV challenges, I think, that you end up with certain forms of carbon dioxide removal too, which is like, you put your, you put your surface treatment catalyst on your wind turbine blade, and then how do you know how much methane you have actually removed? Can you, can you measure that? Do you have to model it? Like, it starts to get a little complicated there.

AI assessment note: “I've seen a couple of startups, um, proposing this exact thing.”

Q Right. As far as the solar production side goes, we've seen this some with wildfires as well as blocking, blocking the sun and reducing our, our anticipated solar production. Do you have an idea of how meaningful that is, or how we are starting to factor that into models?

A I don't know that it's really getting factored into models. I mean, one thing I will say in general is that we've, we've not yet woken up to how much we need to be modeling these more extreme weather events into, um, electricity reserve allocation modeling, basically. Uh, there, there's a bunch of data around, so like Texas, another example of this, right? You know, where you see these kind of multi-day weather events now, basically, uh, Every three years or so, pretty predictably, but we're not, but they're, each one of them is individually considered to be a one in a hundred year event or something like that, and so we don't, we don't yet fully account for them, and that, that causes issues. Wildfires maybe being one, one among that broader category. Um, you know, we haven't yet, I don't think, faced, there, there were some concerns when there were wildfires here in California a couple years ago about, um, About reserves on the grid. We haven't yet had blackouts or brownouts that you would, I think, ascribe to wildfires. But again, it sort of falls into this category of, like, the closer we are to the razor's edge in terms of having enough predictable capacity on the grid, the more things like that or other weather events start to really matter.

AI assessment note: “I don't know that it's really getting factored into models.”

Q Great. The first is from Mike Hogan, PhD. So Mike sees a clear way chemists and physicists can contribute to climate tech, but Mike is a basic biologist. So he wrote in, what are the challenges in green tech that you think are best solved with a biological solution? And in particular, he's talking about molecular biology and genetic engineering.

A Oh, that's very specific. I mean, I guess I'll, I'll answer a broader question about biology. I've, for a while, wanted to do, I think it will do this at some point, uh, a deeper conversation, particularly pitting, not that they're actually in opposition to each other, but pitting biology against electrochemistry, because I think there are a bunch, a bunch of places in, in decarbonization where there are electrochemical pathways to do something, and there are biological pathways to do something, and they're sort of competing with each other, at least in some cases. On biology specifically, I mean, there are some clear areas within quote unquote climate tech that like biology is well suited, right? Acceleration or improvement of existing biological processes being the obvious one. So take nature-based solutions for carbon removal and stuff like that. Um, everything to do with agriculture. Like, those are all categories that are inherently biological, and biology obviously has a big role to play to help solve those. Beyond that, you can look at more, slightly more esoteric, but also really important from a climate perspective thing. So using biology to Change animal feed, right? Um, I guess that falls into the agriculture bucket. Um, and then you get into some categories that I think are interesting and where there starts to be this kind of, uh, emerging battle between biology an…

AI assessment note: “using synthetic biology approaches to produce commodity chemicals”

Q We've tried. We've tried. We've tried everything, and we've even tried, like, giving them smaller portions, and, and they ask for more, and they throw that food on the floor, and it just, it's unavoidable. Like, we're going to create food waste. Then what are we going to do with it?

A I will say my, I have a son who's, uh, about to turn 11 months old, and he has just discovered throwing food on the ground, like, as a, as a fun thing to do, you know, forget, like, that he doesn't want the food. He's just like, oh, this is neat. Uh, and one, I'm throwing a lot more food out, but two, I will say there is an alternate, I would say the biggest competition to Mill in that specific context of the baby throwing food on the ground is, uh, Our dog, who is a great solution to that problem, generally speaking. Not that I don't still have a ton of food waste that I need to do something with.

AI assessment note: “the biggest competition to Mill in that specific context of the baby... is, uh, Our dog”

Q Yeah, what are your thoughts? You in, you out?

A Yeah, so, compressed air energy storage is cool. It has the promise of much better siteability than pumped hydro. There's actually lots and lots and lots of places we could do this, uh, and to Leah's point, it doesn't use as much above ground land, uh, and I like the sort of usage of the oil and gas infrastructure and talents and all that kind of stuff. I think that's all important and real. Historically, the challenge with compressed air energy storage has to do with this Trade off between efficiency and cost. Um, so the basic challenge is when you compress air, you create heat, right? And if you let that heat dissipate into the atmosphere, you're just going to lose a lot of energy in the process of storing it. So most of the innovations around compressed air energy storage have to do with what do you do with this heat?

AI assessment note: “compressed air energy storage is cool. It has the promise of much better siteability”

Q Will you, another example, will you actually, when you get the EV, uh, And I think the message going forward is like, this is cheaper than an ICE vehicle. How much cheaper, and can I rely on that, uh, month over month and prove it? And right now, and so you, you really need this data, the tariff optimization to be able to do that right.

A That's another thing, by the way, that I haven't seen anybody do that I think would be a cool thing to do, which is, uh, really surface to the customer. They're like tracking over time the savings that they are gaining. You know, when, when I have my EV, I, it should be really, really Obvious to me how much savings I'm getting, and I should be seeing that stack up over time, and it's like a little dopamine hit that I should be getting over and over again every time I charge my Vehicle and it's going to create a, you know, higher NPS and it's going to make me recommend it to my friends and stuff like that. You know, I, I want to know because otherwise it gets, it gets buried in the same way that with, with rooftop solar, right? Not a dissimilar thing. Like I should be seeing the counterfactual of what my electricity bill would have been. And then here's what my electricity bill is. And here's my savings on a monthly basis and cumulatively and so on. Like if these things are going to be sold on the promise of savings, um, Make that savings really clear to me, if you indeed believe you're going to deliver on it.

AI assessment note: “I haven't seen anybody do that I think would be a cool thing to do”

Q I think, just some high-level climate tech questions that I think are very much within your safe zone. And so, we're going to start with Matt Waller, handle at wattmallerone.com. Any updates on whether solar costs will keep seeing big declines, and where the biggest costs are, whether they're soft costs, and also whether it matters, i.e. is storage the only piece left for a complete and swift renewable takeover?

A Oh man, that's a big question. I actually think we're in an interesting moment now where there has been, for good reason, a presumption for the past number of years that solar costs are on this, like, steady downward trajectory, and though I think it Has generally been true historically. It's definitely not true today. So our costs actually have been increasing for a number of reasons, right? So we've seen the supply chain bottleneck that has plagued many industries has done the same thing in solar. And so things like steel prices and aluminum prices, which go into solar projects have gone up. Um, all the input costs, like everything that's causing inflation everywhere has affected solar. So prices were already going up, uh, and had basically stopped their decline. And then, in addition to that, now, particularly in the US, we have the kind of specter of this new trade dispute, which may, is already wreaking havoc in the solar industry, delaying projects, potentially raising their prices, and creating a ton of uncertainty. So I think we're actually like, though the long-term trajectory for solar is probably still going to be cost declines, uh, it's possible for the next couple of years, at least, we will see steady to increasing Prices for solar projects or for electricity from solar, which is actually a little bit hard to reckon with and makes it challenging for all these othe…

AI assessment note: “It's definitely not true today. So our costs actually have been increasing”

Q than we could conceivably power, like, Is just good evidence that it's not, not going to happen. Like, I'm just, I'm just very skeptical. Like, what, I guess I, I would just ask you the central question. What do you think? Like, do you, do you think demand is going to get even close to the sort of higher end projections, or are you kind of a deflationist on this?

A I don't think I am smart enough to know the answer to that question. I don't think anybody really knows the answer to that question. I do think that It's, there is pretty universal agreement amongst people who are building these models that we're gonna need, to your point on sort of the, like, will it shift to the edge? Yes, inference might, and that's like an open question. There is going to be, for some period of time in the future, demand for more and more and more powerful models. Those require the centralized, big, big-ass data centers. Um, we are already starting to have a harder time to find sites on the grid to power those in, in the time period that people want. And so I think there's going to be some period of time where we are bursting at the seams from an electricity perspective. I don't know how long it lasts, right? And I don't know whether it gets to the point where, like, Elon wants to put a hundred gigawatts a year of orbital data centers in space.

AI assessment note: “I think there's going to be some period of time where we are bursting”

Q that forward. And eventually, you know, I could, you, of course, can imagine financing products, something like, you know, there's a company, um, called Sealed that was, was, was, or is doing a lot of this. But I think, ultimately, If there is a real OPEX reduction, that's essentially a finance product, right? That, that's a problem that you can build some solution to, um, to bridge those two things.

A That's right. But it is about, it is about that spark spread, um, which is tricky. And, and Northern California, where you guys are starting, where, where you and I both live, is a place where the price of electricity has gone up, I think, maybe more than anywhere in the country. It's not clear to me, but it's a lot, is the point. Is it, A barrier to price of electricity at this point? Or are we still in the kind of early adopter-ish segment of customers where, look, like, maybe it's, it's hard to tell over the long term, am I going to save money on this or not? But I care about this because of X, Y, Z. It's a better customer experience, no fossil fuels, et cetera.

AI assessment note: “That's right. But it is about, it is about that spark spread”

Q Yeah, I mean, one other example would be CATL, which is the largest battery company. You know, what would it be if LFP, lithium-ion phosphate, had not been invented in America, and then first commercialized by A. One to three, which is a venture-backed company?

A That's right. Now, that's a good example of, there is a separate question that I think is a good question, which is, do the, do the investors, the venture investors, and the companies that they invest in reap the full reward of that invention and commercialization? And A-one-two-three is a perfect example of, like, that company went public and then went bankrupt pretty quickly. Um, so maybe the answer to that is no, although some early investors did make some money on it. So, I, I don't know. That, that is a different question, and that's one that I grapple with all the time. But, If you're just taking the big step back, 30,000 foot ecosystem view, um, I still think it is true that again, relative to the dollars deployed, because where are the dollars actually getting deployed? They're getting, you know, in, in total volume, they're either getting deployed into the BYDs and CATLs of the world, or they're getting deployed in infrastructure, right? The real money goes to building big solar projects and things like that. Um, but on a, you know, sort of impact Two dollars invested ratio. I would still claim venture capital plays an outsized role.

AI assessment note: “A-one-two-three is a perfect example of, like, that company went public and then went bankrupt”

Q challenge in this case for the equipment manufacturers, which is how much do you step it up? How much new transformer manufacturing capacity do you build and when do you build it? Um, and again, this gets back to the problem of the uncertainty of all this new electrification and electricity demand. How fast is it coming and when is it coming and exactly what scale is it arriving at?

A Yeah, and on the, on the transformer supply chain pinch, I mean, there's, there's even more nuance to it as well, which we probably don't need to get into, but like, it could actually get worse before it gets better, because there are new standards that are being contemplated right now for, for transformers, which would be, make them more energy efficient, but also make the supply chain even a little bit more challenging. Transformers require in the U.S. this, like, particular kind of steel, electrical steel, and There's only one domestic supplier of that kind of seal. Like there's all these, all these things that we are discovering as, as, as you said, this sort of like, um, the paradigm shifts. And I think for me, that's, this is the, this is the high level point to this whole thing about the gauntlet, which is that it's, it's kind of just like we're, we're in this period where there's a wake up call for the power industry, which is that despite the fact that if you've been in the industry for a long time, as you and I Both have in one form or another. There has been a lot of change, right? Like the, and again, the generation mix is, is probably the biggest one. We retired, we have retired a lot of coal. We have added a lot of new natural gas. We started to add a lot of new renewables. Like things have changed, but fundamentally the supply demand balance has been pretty stead…

AI assessment note: “on the transformer supply chain pinch, I mean, there's even more nuance to it”

Q Okay, well, here's the actual question from Benjamin. Why aren't there more investors launching innovative project finance shop? There's nothing in Europe. Are there new ones in the U.S. outside of the usual suspects?

A So, That is a very good question, I think. Um, project finance, I think, has long been the biggest challenge for a lot of these deep tech or hard tech companies bringing new products to market within climate and energy and so on. And I think that there's a part of it that is fairly well solved and there's a part of it that is definitively not solved. The part of it that's very well solved is if you have a pretty mature technology with a good operating history and customers who are bankable and, uh, you know, financial structures that are well known, like project finance, uh, for renewables, for example, and even for grid scale battery storage and stuff like that, It's just infrastructure finance now. It has, it's available. It, the cost of capital is not high. You could argue it could be a little lower, but it's, it's infrastructure capital in the way that you want it to be, and you can get that stuff built and financed. I think what Benjamin's probably referring to in the part that, um, that has always been trickier is before you get to that point, so you're bringing some new technology to the market, it's not considered to, either it hasn't scaled up enough, or it's just generally not considered to be de-risked by big project finance providers, and that, you know, there have been a bunch of folks creating innovative project finance vehicles of one kind or another, but it real…

AI assessment note: “it hasn't scaled up enough, or it's just generally not considered to be de-risked”

Q on what's in the middle, is it a liquid like salt? Is it a solid that you're in aggregate that you're blowing air through? Or is it a something that's melting and freezing? The way it connects to the output process is somewhat different, but they all have those same characteristics. What's it cost to charge it? What's it cost to hold it? And what's it cost to deliver it?

A Greater than 95% efficiency is one of these key points that I think, like, is, is important to Let's reiterate, because we're used to thinking of storage. I mean, most people are used to thinking of storage as being batteries. Very few batteries can approach 95% efficiency. This is a different context where actually you do get, you don't lose a lot of the energy when you're just turning electricity into heat, storing it as heat, and delivering it as heat. Now, if you were to try to turn that heat back into electricity again and use thermal energy storage for electricity Storage purposes, which some companies have also done, you would lose more in that reconversion. Heat to power has a conversion efficiency loss. But again, what we're trying to do here is just electricity to heat stored as heat delivered as heat. And it's actually remarkably efficient when you do it that way. The other thing, as I learned about Rondo that, uh, I came to appreciate is that it also, you know, we, we talk, you said this is McKinsey's thing is the long duration energy storage council. Um, When we think about things like long duration energy storage, the other thing is it doesn't, you don't really lose a lot of heat over time. This is what you were describing with the, the rock that stays hot. Uh, if I put it in my oven, or if you're trying to create like a hot stone massage, you'll discover that the…

AI assessment note: “turning electricity into heat, storing it as heat, and delivering it as heat”

Q the infrastructure, on the labor force is to make that possible. We'll see if it materializes or not, but I'm not going to bet against it. You don't bet against tech. The tech gets better. So the really, the question becomes, what do you invest in? When do you invest in it? What's the smart play? I look forward to joining you in five or 10 years to discuss that.

A It's literally my day job is trying to answer that question. Um, The last thing I want to talk about just briefly is the infrastructure side of this. You, we talked about at the beginning how there are 5000 miles of existing CO₂ pipelines. Um, on the other side of it, though, you know, there have been a couple of pipelines that have been already proposed and in development in the Midwest, Summit and Navigator, mostly to take ethanol-based CO₂, um, to sequestration sites, and they've both faced a fair amount of opposition, local opposition, and all the challenges that we face in In any time we try to develop horizontal infrastructure in the United States, it's not dissimilar from what, from what happens when we try to build transmission lines, for example. So to what extent does that pose, you think, a real challenge for this CO two, whether utilization or sequestration future, our ability to build the infrastructure required, the mainstream required to get it from where we capture it to where we use or sequester it?

AI assessment note: “It's literally my day job is trying to answer that question. The last thing”

Q Oh, great. I had never heard of it. It sounds like a cool organization, and it's described, um, as a place that builds deep tech startups that slash emissions, remove carbon, and cool the planet. So my first question is, what is a deep tech startup? And to be clear, that's not the question from Benjamin. That's just from me. What is a deep tech startup?

A That's so funny. So that's often the term that I use in describing what I do, right? So I, my job is not this podcast. My job is at EIP, and I, I lead this fund we call our Frontier Fund, which when I'm describing it in shorthand, I say it's our Deep Tech Climate Fund, and then occasionally somebody asks me the question you just asked, which is, what is Deep Tech? Um, and Deep Tech is a term, I'd say, that has sort of taken hold in Silicon Valley circles, but often means something a little bit Different than what I mean it to be. I also don't think there's a perfect definition. In Silicon Valley world, you know, it means sort of like, you could think of it as being sort of frontier tech, but in Silicon Valley world, that can also mean things like next generation ML. Um, you know, I think some people were calling, like, some of the stuff that you're seeing now with generative AI, like six months ago, people called it deep tech, and now they probably don't. In my context.

AI assessment note: “In Silicon Valley world, you know, it means sort of like... frontier tech”

Q that we don't yet have technical solutions for. And This is gonna have to be a mix of both working to have the growth of livestock brought down, because the number of livestock is still dramatically growing, and also figuring out for whatever that massive remainder of livestock is that we will, we will almost certainly have in the twenty-thirty time frame. How can we better minimize these methane emissions?

A Right. So what you're saying in, in brief is there's things that there are practice changes that we can make and are already being made and those have an impact, but it's, it's honestly, it's sort of an incremental impact, right? If we're still going to have a billion cattle in the world, there's only so much we could do out of practice change alone. So the second question is, are we going to have a billion cattle in the world? And that's where things like alternative protein, alternative, um, dairy come into the mix, and there, it sounds like you're saying, yeah, absolutely, we should do that stuff, but let's be realistic about the timeframe in which it could theoretically replace this global population of livestock, especially given that currently it is continuing to grow, not shrink, along with global population growth and Uh, and economic development and all that kind of stuff, and so if you add those two things up, they will have a non-negligible impact, but they are not the solutions alone.

AI assessment note: “things like alternative protein, alternative, um, dairy come into the mix”

Q We should. Okay, so next question. Coming at us from Philip Healy, another fellow Californian, which clean energy sectors would be the best use of funds, and which of those areas overlap with delivering the most near-term cost decreases in electricity prices?

A Well, I like the second part of that question the most, right? Like, I, I don't know that I have The strongest opinion at the high level on which clean energy sectors are the best use of funds overall. But the second part of the question implies that the best place to place your funds right now within a clean energy world is to deliver near-term cost decreases for electricity. And you mentioned this, this guy also is a Californian like us, like we, we both appreciate, I think we're actually kind of at the vanguard of what's going to happen Nationally, to some extent. Maybe not for the same reasons. Like California, power prices in California are Ludicrous. They are ridiculously high. Everybody in California who's listening to this podcast is immediately going to agree. Um, now that is for a host of reasons in California, right? We have wildfire costs and all sorts of other things, um, from the utilities here. But in my mind, it's a harbinger of what's coming in the rest of the country. I'm very concerned about retail electricity prices. I just don't, I don't see a trend that points in the opposite direction right now. We've been living in a world for a while where T and D costs have been rising, but it's been offset largely by cheap natural gas prices. Right now we have the cost of new generation going up. Um, natural gas prices, natural gas new generation is getting more expen…

AI assessment note: “I don't know that I have The strongest opinion at the high level”

Q And, and therefore lower range for vehicles, right? So iron, iron-based vehicles have lower range than nickel-based vehicles today. They're really kind of marginal, but they're good enough for, you know, some markets, they're good enough for some consumers, but they're not great for everybody.

A That speaks to, like, a bigger thing for us to talk about, because I want to get to, like, what's happening in the EV market today, but it's an interesting, if you think about it in that frame, right, like, the trade that we've been making broadly over time, as you said, is toward lower cost, shorter range, all things equal, right, because we've, we've gone from cobalt to nickel, nickel to iron, so now, like, LFP is, On the ascendant path. And, and that presumes that that trade is worth making that like the, um, the downside, which is lower range or needing to put in a bigger battery in order to get the same range was worth doing in order to get to like a cost point that maybe brings more consumers into the market. So there's this implicit trade that the industry has been driving toward for a decade. Plus, do you think that that's a false trade, or do you think that Or do you think it was the right thing to do, I guess, for the industry?

AI assessment note: “That speaks to, like, a bigger thing for us to talk about”