The Exchanges

Q Apple had just acquired Tesla from the start? What if Apple had just acquired Tesla from the start? So let's start at the beginning, like the way, way beginning. Steve Jobs actually proposed the idea that Apple could build a car in the 2000. Apple even thought about acquiring General Motors in the wake of the financial crisis. Why was a car attractive to Apple even in the early days?

A The car was attractive to Apple because of the idea that a lot of people spend a lot of their time and a lot of their days, uh, driving, right? They felt like that there was something they can contribute to that portion of people's lives. They wanted to own the on-the-go. Obviously, the iPod and the iPhone was an on-the-go device. They wanted to own the living room. They wanted to own home and work and business. And The car is a big part of, of the on-the-go experience, so that was something they took a brief look at, ah, when Steve Jobs was around, and then that project re-kicked into, into gear, ah, so to speak, 2013, 20 14 time frame. Initially, the, the first part of the car project was examining what it would take to acquire Tesla, integrate Tesla, and take Tesla to a new level under Apple ownership. And so they spent a lot of time exploring the possibility of acquiring Tesla. Apple's head of mergers and acquisitions had met with Tesla's Elon Musk and JB Straubel multiple times, and Tim Cook, Luca Maestri, the company's CFO, decided not to, you know, move forward with such an acquisition. And so that was a big moment. Instead, Apple allocated budget, resources, et cetera, to Going ahead and engineering and developing its own car. So Dan Riccio, the head of hardware engineering at that moment, uh, started hiring people from within the company, program managers, software man…

AI assessment note: “The car was attractive to Apple because of the idea that a lot of people spend”

Q So now you're at Snowflake after Myst AI was acquired last year. What is Snowflake's approach to building out the data cloud, and how does energy actually fit into that?

A Yes. So in companies of all types, data gets siloed as the company grows. It's a bit like entropy. It happens naturally unless you try to prevent it. And Snowflake's aim and vision is to mobilize that data so that Business leaders can ask questions that span across the business. For example, which parts of our grid should we do maintenance first to minimize the chance that there is an outage? Or how much capital should we allocate to make sure that we meet the risk requirements of a market like ERCOT? The Snowflake Data Cloud sits on top of the three cloud hyperscalers, so we don't own our own data centers. And as companies Use this data cloud and store all their data in the cloud. That then unlocks workloads like AI and machine learning, or applications that run on their data, or visualizations. What we really unlock is the ability to combine IT data, so data around, ah, unit sales or supply chain data, with OT data, which is data from devices. So data from Wind turbines or data from the grid. And by combining those two things, that really unlocks a lot of these valuable insights for executives to take action on.

AI assessment note: “combine IT data, so data around, ah, unit sales... with OT data, which is data from devices”

Q to apply it to the power sector, and so there, you have a number of different Approaches inside of utilities. Like, what are the most common approaches now, and are there any that you think are particularly effective when it comes to team building and then eventually, like, testing out products and collaborating and building this sandbox approach so that they can figure out how to actually scale an application?

A Yes. I think that there are a few key components. One is, as you mentioned, unlocking a data foundation in which all the data is there so that you can actually Run models. The second one is around, uh, experimentation. So something specific I've seen work very well is to embed teams of data scientists and software engineers with the lines of business. So to make sure that somebody who is a data scientist is actually sitting with a trader for a week, or sitting with the grid management team for a week, or sitting with line crew for a week, and maybe even going with them to get a, uh, Clear a sense of the issues that they grapple with. And then part two of that is setting up those technical teams to experiment quickly. So for example, creating sandbox environments that do not have the same requirements as the fully skilled production models that power the utilities and customers in which they can, within a matter of days, deploy new applications that are initially used by just a small subset of users to see what works and what doesn't. The final thing I'll say on accelerating AI adoption is that I think sharing data in the utility industry is not yet delivering on its full promise. So one of the amazing things about some parts of utilities is that they're actually not competitive. And as an example, in California, the smart meter data that the three big investor on utilities have…

AI assessment note: “embed teams of data scientists and software engineers with the lines of business.”

Q What's the second utility doing that is investing in distributed power plants or virtual power plants? Walk me through that potential scenario.

A Sure. So a utility who wants to explore the potential for a virtual power plant to serve that increase in demand, they would take a look at the distributed energy resources that the DERs that are already on their grid, and they may also forecast adoption rates And what that looks like is, you know, taking a look at your commercial industrial base, who has flexible load, that's, you know, pretty well understood, kind of sector by sector. You would look at EV adoption rates. You'd look at rooftop solar. You know, the DOE's national labs have tools to allow you to kind of forecast adoption rates and understand what types of assets are on the grid. And they would look at What kind of capacity can I get from these distributed resources? And then they would consider, okay, well, how am I going to orchestrate that? So utility could, you know, do what National Grid and Eversource are doing in New England and through their Connected Solutions program. And this is, you know, what's known as a bring-your-own-device program. They say, hey, as long as you have this set of brands of smart thermostat or of water heater or of battery, you can enroll in our program. And we'll pay you for shaving peak. And through connected solutions, those, uh, utilities have access to megawatts of capacity that help them shave peak so that they are less reliant. In this case, it would be, you know, on ISO New …

AI assessment note: “do what National Grid and Eversource are doing in New England and through their Connected Solutions”

Q And how much do we actually need to meet the expected load growth that's coming?

A Well, we have the need to serve 200 gigawatts of new peak demand between now and 2030. And at the same time, we have an incredible amount of DER capacity coming online. So, and I can put some numbers to that in a second. You know, what that adds up to is an enormous opportunity for that DER capacity to be aggregated into VPPs to serve That peak demand. Now, we're not saying we're going to get every last smart thermostat, every last battery, right? But if we can triple the capacity of VPPs between now and 2030, we could serve 10 to 20% of peak load nationally through virtual power plants. And if we did that, because they are a lower cost solution, we'd be saving about ten billion dollars per year.

AI assessment note: “we have the need to serve 200 gigawatts of new peak demand between now and 2030”

Q Okay, so then do we actually have the distributed energy resource capacity to support all that?

A We will. We are about to experience a tsunami of DER adoption across the United States, and the way we think about that is across three different categories of DERs. Those that generate electricity, so think, you know, rooftop solar or, uh, fuel-based generators. DERs that consume electricity at flexible times, so think, you know, EV chargers or commercial industrial loads or electric water heaters, and then DERs that store electricity, so behind-the-meter batteries or even EVs. With these DERs, we are experiencing a higher and higher capacity addition every year, so with DERs that generate electricity, we're getting about 20 gigawatts added to the grid each year in twenty-twenty-five. That's going to increase to about 35 gigawatts added to the grid by twenty-thirty. If you look about, and that's nameplate, right? If you look at flexible demand of electricity, we're adding about four to six gigawatts of flexible demand to the grid each year in the form of smart thermostats, water heaters, and commercial and industrial load. Now, Roughly, you know, four to six gigawatts of flexible demand is the equivalent of about 50 peakers worth of flexible supply per year added. And then you have about five to 15 growing to about 25 gigawatt hours of behind-the-meter battery capacity added to the grid each year between now and twenty-thirty. Now, those are enormous numbers, But they pale in …

AI assessment note: “We will. We are about to experience a tsunami of DER adoption”

Q as clean as possible, but as we'll hear, it is potentially stalling the market, and, um, there's a real debate over, uh, how stringent these rules need to be, and how much hydrogen, and, and whether they'll hold back hydrogen production, and, um, we're going to talk about that now. First, let's talk about these treasury rules. Shalini What did Treasury propose about how these hydrogen tax credits should work?

A So this guidance has been hotly anticipated. As you said, it came out, uh, near the end of last year, right before Christmas, I think. That's my memory of it. And, uh, it's described as having three pillars. So, um, and, and this all goes to what a project has to do to qualify for the, um, the highest amount, um, of the hydrogen production tax credit. The most you can get is three dollars a kilogram. And for, to do that, you have to be emitting less than point Four or five kilograms of CO₂ equivalent. Um, so it, you know, that's, it's a high bar, and if you hit, if you aren't quite that clean, then you get, um, you can still qualify for the PTC, but you may not get the full three dollars a kg. So the three pillars to get you to, um, you know, that, that higher, um, PTC amount that you definitely want to get, um, are additionality, so clean energy resources, um, Have to begin operation within three years of the hydrogen plant, um, being operational, so that you're actually incenting new generation and not just, um, using the renewables that are already built and maybe taking it away from a data center. Um, uh, the second pillar is deliverability, so the clean energy resource has to be in the same region, um, as the, as the green hydrogen project, and that's to prevent, Um, actors from just buying, overbuying in one market, and then kind of spilling that power on the grid where i…

AI assessment note: “it's described as having three pillars... are additionality... deliverability... hourly matching”

Q ton of the new activity that we've seen as a result of, uh, the, the new EV tax credits has brought a lot of battery manufacturing, retooled factories, new, new production factories. How has this impacted solar? So there are also specific tax credit requirements for solar cell manufacturing. Are Inflation Reduction Act incentives creating the same level of activity in domestic solar production as they are in battery production?

A That's a good question. I would say the IRA has generated a lot of interest in module manufacturing. There's, I think we're, we're tracking something on the order of over a 120 gigawatts of module manufacturing announcements. Not all of those will materialize of course, but that's, you know, 10 times what the U S currently has today. I think the difference with the, uh, potentially a difference with the battery manufacturing industry is that The U.S. would virtually have to build cell manufacturing, which is just, you know, modules made up of, um, a bunch of cells. They'd have to build a cell manufacturing, uh, industry basically from scratch. There's no real, there's, there's no cell manufacturing in the U.S. to speak of. There have been some announcements as a result of the IRA, but we expect it'll probably take at least another few years for that to actually manifest. And without cell manufacturing here in the U.S., it makes, you know, building out that domestic, that fully domestic supply chain and truly taking advantage of some of the domestic content incentives in the IRA a lot harder.

AI assessment note: “I would say the IRA has generated a lot of interest in module manufacturing.”

Q There is some biggie guidance coming out on EV tax credits and hydrogen tax credits. Jeff Maria, I know you've been following this. Any commentary on, um, How those landed and what's to come now that we know how those tax credits are going to be structured?

A Yeah, I think that the EV tax credit was largely anticipated, but it, it will put a lot of pressure on automakers selling vehicles in the U.S. to try to find a way to get not just the, uh, batteries that go into those EVs made in U.S. or U.S. free trade partner countries, but to ensure that all the Critical minerals and materials that are going into those batteries don't come from foreign entities of concern, and for us, that term means largely one country, which is China. Uh, we have guidance from DOE and Treasury on just what that means, that foreign entity of concern guidance, and, um, right now it's, it's looking like, frankly, um, with China dominating so much, between 70 to 90% of key precursor, you know, Uh, materials and, uh, kind of components of lithium ion batteries that very few, if any, EVs sold in the U.S. will be eligible for that 7500 per EV tax credit, which is seen as really important for, uh, making these vehicles more affordable and really upping their adoption to hit that Biden administration goal of having half of all new car sales in the U.S. be EVs by 2030. So, on the other hand, it's Driving an enormous amount of investment in domestic manufacturing of batteries, and what I think is probably accurately described as kind of a lithium gold rush here, seeking out sources of the raw material for these batteries in the U.S., and investment in technologies th…

AI assessment note: “EV tax credit was largely anticipated, but it, it will put a lot of pressure”

Q All right, Maria, what is your rising star?

A I think a rising star in clean energy this year would be geothermal energy, especially sort of the next generation technologies that try to access heat from unconventional places, kind of deeper into the ground, or maybe where there aren't hot springs or geysers available. And one standout's Uh, startup in particular would be Fervo Energy. They use horizontal drilling techniques, fiber optic sensing tools, um, mainly developed by the oil and gas industry for fracking, and now they're using it to extract heat from the earth. And late last month, uh, Fervo and Google successfully started operating a first-of-its-kind power plant in Nevada using these advanced, uh, geothermal techniques. And it's now delivering electricity to the grid in Nevada, which is where Google operates some of its massive data centers.

AI assessment note: “I think a rising star in clean energy this year would be geothermal energy”

Q Yeah, the DOE piece is fascinating. So, so first, let's just talk about FERC. Critics call FERC a rubber stamp for these projects. Why?

A Yeah, that, that accusation of FERC being a rubber stamp for the fossil industry, fossil fuel industry, uh, is repeated a lot, and part of that comes from the fact that FERC has only ever rejected one LNG export proposal, the Jordan Cove project in Oregon, uh, because it didn't meet the, uh, commission standards, and notably Jordan Cove was never built, but that's not because of FERC. FERC ultimately did go back and approve the project. Um, it was, um, It ended up being canceled because of opposition at the state level. However, environmental groups keep challenging FERC's approvals of these LNG projects, mostly under the National Environmental Policy Act. Federal courts keep on finding that FERC needs to evaluate these projects' climate and environmental justice impacts a lot more thoroughly than it has been.

AI assessment note: “part of that comes from the fact that FERC has only ever rejected one LNG export proposal”

Q more question on FERC. So, Nicole, there was, there were some attempts, some proposed changes to factor in the climate cost to infrastructure within FERC, and there was a bunch of backlash from Republicans and also from Democratic Senator Joe Manchin, and that backlash caused FERC to step back from scrutinizing LNG, the climate impact of LNG proposals. How did that play out, and how has that impacted the commission?

A I said this already, but I think it's important to repeat. Um, so FERC is an independent agency, which means that it's a little bit insulated from pressure from the administration directly, but importantly, it's still subject to political pressure, particularly from Congress, which has to approve people who are nominated to the commission. Almost two years ago now, um, Richard Glick, who was the chairman of FERC at the time, um, and the two other Democrats on the commission, uh, voted to adopt a policy that would basically Put more scrutiny on the greenhouse gas emissions and environmental justice impacts, um, from LNG projects and a lot of other gas projects as well. Um, and they basically said that that was to make sure that FERC's decisions were holding up better in court, uh, amid all of these challenges that FERC kept losing. Uh, and that really didn't go over very well. Um, Republicans and also, uh, Joe Manchin, the Democrat, Democratic Senator from West Virginia, uh, were not happy. And the timing was terrible because it happened about a week before Russia invaded Ukraine, which then put LNG in a huge global and national spotlight, often very positively, at least for a while. So, ultimately, Manchin refused to vote for Biden's renomination of Glick to the commission, um, and Glick ended up having to leave at the end of Since then, his seat still hasn't been filled. So FE…

AI assessment note: “Manchin refused to vote for Biden's renomination of Glick to the commission”

Q on a sectoral level? So Catherine, um, Fatih Biral, who's the executive director of IEA, wrote in an op-ed that the world is on the cusp of a historic turning point, and that this is the first time that a peak in demand is visible for, uh, each fossil fuel this decade earlier than anticipated. So what does IEA say is pushing us toward peak fossil fuel consumption by 2030?

A Yeah, for decades we've been stuck at about 80% fossil fuels no matter what, um, but it looks like based on the steps, which are the stated policy scenarios for all of the countries, um, that, you know, internal combustion engine sales are down, um, they're below where they were before COVID, um, additions of coal and gas plants are down 50%, sales of residential gas boilers is down, heat pumps are up, there's a real momentum toward electrification. So the clean energy transition has really taken hold, and when we talk about a peak, that means after 2030, we'll still be at about 73% fossil. It just means we will then start the decline. Doesn't necessarily mean we're going to meet our net zero energy goals at all, or that we will meet the 1.5 degrees centigrade targets, but it does mean that we're on a trajectory where eventually The clean energy transition will take over where the fossil fuel industry was.

AI assessment note: “internal combustion engine sales are down... additions of coal and gas plants are down 50%”

Q have been grabbing because of how IEA has shifted in its projections. And, you know, they've, they've been historically very conservative about the growth of renewables and, and batteries and EVs. And now they've really caught up with and are very proactive about what the growth of those sectors looks like. Um, Mike, to you first, like, does this represent a pretty significant shift in how IEA views the world?

A Yeah, I think it is significant. The shift has been happening over time. IEA did its net zero pathway analysis, which to me was a real groundbreaking report that an organization that was created in 1974 to help coordinate a collective response to major disruptions in the supply of oil is now looking at how to phase out the use of oil and giving direction and leadership to Developed and developing countries on, on the pathway they need to take. I mean, IEA's role is not to be on the bleeding edge of technology, and this pivot means that the data is just staring them right in the face. It's, it's undeniable. Um, Some of the figures about electric vehicle deployment and solar power investment in particular, these are called out in their report as exponential trends that are now forecast to meaningfully eat into the growth of fossil fuels. I think their projections may even still be conservative, especially when it comes to the continued use of coal power at scale, particularly in China and developing countries. Not only is doing so untenable, Under a country's Paris Agreement commitments, but China in particular is adding clean energy at an absolutely mind-boggling pace. They're expected to add a 150 gigawatts of solar in twenty-twenty-three. That's roughly five times what the U.S. is projected to add, and they're projected to top a terawatt of total installations by twenty-twenty…

AI assessment note: “Yeah, I think it is significant. The shift has been happening over time.”

Q And so in the chemical space, for ethylene in particular, we're not necessarily talking about a change in chemistry. We're mostly talking about electrification. Is that right?

A Right. So in my story, I focused on sort of the most emissions intensive part of ethylene production, kind of within the, the walls of the plant itself, and that is, uh, generating high heat to crack ethane gas into ethylene molecules, which are then processed into other chemicals. And so this sort of, this process of generating heat to crack ethane accounts for 90% of the carbon dioxide emissions associated with an ethylene steam cracking facility. So I focused on initiatives to electrify that portion, but you're completely right that, um, a huge part of sort of the life cycle emissions of ethylene and all chemicals come from the fact that they are oil and gas feedstocks, and they, even electrification can't fully address some of these other issues that come with ethylene and plastics more generally, and that is they generate a lot of toxic air pollution, they generate a lot of plastic waste that ends up In waterways, in our bloodstreams, etc.

AI assessment note: “Right. So in my story, I focused on sort of the most emissions intensive”

Q Yeah, so Maria, over to you. Why don't we talk about steel, since we've covered the others?

A Sure. So I would say in general, the steel industry, uh, is characterized by sort of these large legacy players. Um, in the United States, U.S. Steel and Cleveland Cliffs are actually the only two companies that still operate blast furnace and basic oxygen furnaces, but there are a handful of other companies that operate these electric arc furnaces using scrap metal, Nucor being one of them, and Nucor is actually America's biggest steel supplier. Uh, and then you have kind of a whole range of equipment manufacturers. Midrex is a company that makes these direct reduced iron plants. And in terms of sort of the startup space, obviously Boston Metal, Electra are ones that, that stand out. But I, my sense is that right now that a lot of the innovation is happening right now among sort of these established companies with support from Government programs. If you're talking about Europe, especially a lot of those projects are funded through European Union funding, or kind of state-sponsored initiatives.

AI assessment note: “in general, the steel industry, uh, is characterized by sort of these large legacy players”

Q be a really valuable resource, uh, but properly managing that solar is, is critical, and we're now in a place where we have the battery and the control systems to make them super valuable on the grid. So, what a transition that's been, huh, Jeff? You know, you, I remember you were covering the Hawaii, uh, solar story when, when there were concerns about it. Uh, completely overloading the grid.

A Well, I, I remember there was a presentation, uh, that Dora Nakafuji at, at, uh, Hawaiian Electric put out, uh, you know how California is the duck curve where all that rooftop solar, you know, brings down the net demand on the grid real low, and then it spikes in the evening when solar goes away. In, in Hawaii, they had the Nessie curve, which is like the duck curve, except it goes underwater, per se. Uh, you know, that, that midday solar Is actually exporting more power, uh, than the grid as a whole might need, and that's a big problem for utilities who are trying to manage those distribution circuits. They weren't made to take power going the other direction. You got to kind of reconfigure a bunch of equipment there, but if you can just shift that production from that belly, uh, to that neck, you know, that time in the evening, When all the solar's going away, all the load remains, you've got this enormously valuable resource, and, you know, I guess necessity is the mother of invention. Hawaii has to deal with this on its own. It's a bunch of islands. They can't get power from anywhere else, and land is limited. That's a lot different from continental grids, but The same logic is driving, you know, changes on the mainland. California's, you know, change from a net metering system to a net billing system, which really radically reduce the value of solar that's fed back onto t…

AI assessment note: “I remember there was a presentation, uh, that Dora Nakafuji at, at, uh, Hawaiian Electric”

Q Created this new tax credit for clean hydrogen, and the bill did create some language around how to calculate emissions and what classifies clean hydrogen, but that calculation raised concerns as well, and it quickly spawned this debate between industry and environmentalists and, and, and people within the industry. So as officials start to craft and finalize the language for these tax credits, how did the debate start picking up?

A The debate really picked up, uh, first in Europe, uh, about a year ago. It's because the realization is that just producing hydrogen using an electrolyzer doesn't guarantee it's clean or, or low carbon, right? Because if in the US, for example, you plug an electrolyzer into the electricity grid and don't, don't do anything to that to, and don't set any standards, that hydrogen has double the emissions intensity of just producing hydrogen from natural gas directly. So you need some more requirements than just Procuring hydrogen from, uh, electricity from, from the grid, right? So the U.S. said that clean hydrogen needs to be produced, uh, at a low carbon way, and that those emissions need to be measured over the life cycle of that car, of that production of that plant. And while that's already a good standard, what it doesn't account for is the sort of the consequential emissions that hydrogen produces by taking renewable electrons away from other uses on the grid. And that's really what this new hydrogen guidance that should be issued by mid-August is trying to regulate. And that's where most of the debate is on.

AI assessment note: “The debate really picked up, uh, first in Europe, uh, about a year ago.”

Q And there are definitely clear lines between environmental groups and the hydrogen industry, but it's not as simple as that. There is debate within the hydrogen industry itself. So how do we see those lines separating the different sides of the debate?

A Yeah, you're absolutely right. In general, environmental groups are more in favor of stricter regulation on these criteria, while most of industry is not But there's many exceptions to that rule. Companies were really interested in producing large volumes of clean hydrogen, like air products, and manufacturers of electrolyzers like electric hydrogen have come out in favor of very stringent standards on, on hydrogen production. Whereas others, like Plug Power, which is also an electrolyzer manufacturer, doesn't really favor very strict standards on hydrogen production. And then there's a bunch of companies in the middle of this debate which want some rules on clean hydrogen production from the grid, like NextEra Energy in the US, but are arguing for a phase in of these requirements over time. So giving some sort of transition period to most companies to be able to adopt these rules later.

AI assessment note: “Companies were really interested in producing large volumes of clean hydrogen, like air products”

Q Why do we need clean hydrogen? I mean, when you look at the European economy, the US economy, why is hydrogen so important for cleaning up the industrial economies in those regions?

A Yeah, we would argue that hydrogen is essential to reaching a net zero emissions. You can get very far in terms of emissions abatement just with clean power and electrification of end uses alone. That gets you, like, 80, 90% of the way. Then adding carbon capture and storage also gets you much closer. But the final five, six percent or so of emissions abatement that you need to reaching net zero by 20 50 really relies on hydrogen being used widely in the sectors that is already being used today. Hydrogen is already a chemical feedstock that is being used for things like ammonia production, which is used for fertilizers. And in sectors that have no other alternative to using hydrogen, for example, as a fuel and heavy transport like ships and planes, uh, for steel production, hydrogen is one of the few options to produce primary steel. So it's absolutely essential that hydrogen gets scaled up if you want to reach net zero.

AI assessment note: “hydrogen is essential to reaching a net zero emissions. You can get very far”

Q Yeah. So how does Carbon Aware Compute work and how has it evolved since when we caught up with Anna, with Anna, it was sort of at its early stages and it's been a couple of years. So how does it work? Where are you at?

A Yeah, so Carbon Aware Compute is a really exciting platform that allows us to shift our flexible workloads, both in time and in space now, uh, in order to have those workloads be run when the wind is blowing or where the sun is shining. Um, so we're now rolling this out across our global data center fleet. Um, and the way that it works, one of the primary inputs that we need is a day ahead, uh, carbon intensity forecast. And we get this data from a partner of ours called Electricity Maps. They are continuously running machine learning models, uh, looking at the historical data they've collected, looking at weather data to generate these predictions. Those forecasts, in addition to load forecasts, which are created internally from statistical models, are put through a robust optimization model, and that optimization model essentially outputs a schedule That minimizes the carbon footprint of our flexible workloads across the fleet.

AI assessment note: “allows us to shift our flexible workloads, both in time and in space”

Q Got it. And so where does that technology fit into the other machine learning and data science tools that you're developing, that you're working on specifically?

A Yeah, so we, um, are looking at, uh, AI and machine learning across a number of different, uh, technologies. So the Carbon Aware compute platform is really internal to data, Google's, um, data center operations, but we are looking at opportunities beyond, you know, Google's footprint. So let me walk through three different examples where we found AI to be particularly useful in the energy space. Uh, so the, one of the, the problems that, you know, we were trying to solve, Um, was to make our wind and solar assets more economically efficient, and, um, in order to do this, we created a solution called Grid Intelligence, and that leverages ML and AI models to generate solar and wind production forecasts using lots of historical data, again, weather data, other factors, and we run this, again, through an optimization model, so you can kind of see there's a pattern, machine learning inputs, optimization model, and that outputs an optimal dispatch Schedule, essentially, or, um, delivery commitments for the wind and solar assets, and that can therefore increase market participation revenue, making the projects, uh, more economic. So we're using that for our own renewable energy fleet, but we're also, what's really exciting is we've started to work with Engie to extend that technology such that they can use it for their own portfolio of assets. Um, the second problem I wanted to highli…

AI assessment note: “let me walk through three different examples where we found AI to be particularly useful”

Q And those two guys set in motion this boom in Texas for renewables that surpassed all expectations, and this was not until recently considered part of the woke agenda. What has happened in the years since?

A Well, it's, it's the energy marketplace, right? Because in the late 19 nineties, early 2000, it appeared that the Permian Basin was running dry. We weren't going to be able to produce that much oil. The natural gas was running out. A company called Chenier was building a natural gas import facility on the Gulf Coast to bring natural gas from Qatar. Uh, it was gonna cost Texans a huge amount of money, and we realized that the one thing Texas has is a lot of wind and a lot of, ah, sunlight. And wind, of course, was the first technology to be able to produce energy at a reasonable rate. You also had the federal, ah, tax incentives, the production tax credit that helped to bring down the cost of, ah, wind energy. So, Sure. I mean, it was a Republican national security issue to produce energy at home and not rely on imports. But a guy named George Mitchell from here in Houston perfected, uh, hydraulic fracturing, which was able to go into the source rock Shatter the shale that, that created oil and gas and bring even more oil and gas out of the ground than anyone ever thought was imaginable. Uh, that crashed the price of both oil and natural gas in Texas and set off a fierce competition between, uh, natural gas and renewables to supply electricity to the state of Texas.

AI assessment note: “crashed the price of both oil and natural gas in Texas and set off a fierce competition”

Q And then there are some other pieces of the legislation around interconnection and capacity requirements. Can you explain those as well?

A Sure. So one measure that, two measures that passed that are going to hurt the renewable industry. One is the cost of transmission lines. Up until now, um, connecting a wind farm to the grid through transmission lines was covered by all the rate payers everywhere in the state. Now, if your wind facility or solar facility is more than a certain distance from From, from the interconnect, you're, you're going to have to pay for those extra miles to connect. And it's about a million dollars a mile to run a transmission line in Texas. Uh, that's going to put significant cost on new facilities, because frankly, all the places convenient to interconnection, they're already taken. Um, the second thing that's going to happen with every new facility after January one, Is something called a firming requirement, and what this says is that when supplies are super, super tight for electricity, the Public Utilities Commission is going to sign every generator by portfolio, not by individual facility, but by their portfolio of facilities, a certain amount of electricity they're expected to produce in an emergency. And if you're a wind or solar facility, and you're not confident you're going to be able to supply that quota of power in an emergency, you're expected to go find it somewhere else. You can either build a battery facility and provide your quota that way, or you have to go out and cont…

AI assessment note: “One is the cost of transmission lines... The second... is something called a firming requirement”

Q So let's break down a few of these. Let's go to battery storage first. So what are you seeing in terms of cost and deployment trends in the short and medium duration battery space that will, will make batteries an important part of some of these hubs?

A One of the great things about, uh, batteries, lithium ion in particular, is it's commercial, it's scalable. We deployed a couple of large-scale, Each battery system's on Long Island several years ago, and they operate well, uh, and, and operate like peak power plants. That's a great thing. We'll see a lot more, right? So New York has a six gigawatt goal. You know, our modeling suggests, you know, New York will need anywhere from 20 to 40 gigawatts of, of just basically four hour duration batteries. I think we'll continue to see, see deployment of, of short and medium duration. It's really more about getting the, The commercial framework, the regulatory framework, right to do that. We think it can play, uh, play a role on the transmission system as well as look like an energy generation resource. You know, the, the price, obviously the cost of, of batteries has come down substantially over the last decade. That trend has stopped in the last, uh, year or so, um, given supply chain challenges like we've seen in the macro economy generally. But, uh, overall we expect that to continue, uh, over the long term and make it a, uh, an economic resource. And then things like the, the IRA are important, uh, important, uh, to provide the investment tax credit for storage to make those even more economic.

AI assessment note: “the cost of, of batteries has come down substantially over the last decade.”

Q What about the union path? We talked about that earlier. Don't, don't the unions train people as well, like through the apprenticeship programs?

A That's another really interesting point. So the unions have this robust training pipeline. They offer free classes, paid apprenticeships, and they can offer those because of union dues, but they can only admit so many people. And so I spoke to labor advocates who think we need to strengthen the unions. Right now, union electricians, they don't really do that much work on houses because they're undercut by non-union electricians who can pay their workers less. And I was told that If some of the public funding that was coming out for electrification had labor standards attached to it, that that could sort of level the playing field for the unions and help them compete for those jobs, and then they could train more workers.

AI assessment note: “unions have this robust training pipeline. They offer free classes, paid apprenticeships”

Q And what about people like Boren at the top of the show? He was talking about how he was doing some worker training too, right?

A Right. And so I heard from a lot of people that solving this also really hinges on involvement from people like Boren. And that's partly because, you know, they already know how to do this work. They're the people that will ultimately hire these workers. The obstacle there, as Boren talked about, is that training people eats into his profits. That's part of the reason he can't hire and grow right now. Panama Bartholome, who you heard from earlier, he told me that one way to solve this issue might be to provide funding for this kind of training. Like, for example, the Inflation Reduction Act, it has funding for state energy offices to do workforce development, and some of that could be used to pay, let's say, half or three-quarters of the wages of a trainee so that contractors like Boren can afford to take them on.

AI assessment note: “solving this also really hinges on involvement from people like Boren.”

Q So what are the other bottlenecks in the engineered space?

A So there's three things that keep me up at night about the engineered pathways, and it is not cost. We're going to innovate, the cost will come down. But there's three things that keep me up at night. The first of those is human capital. We probably have enough engineers, but they're not in the right place yet, and they're not trained. We certainly don't have anybody else. We don't have specialty welders or electricians. We don't have the people who will permit these projects. We don't have tax equity lawyers who know what I'm talking about. We, we do not have the full complement of business and worker skills to deploy. And until we do, we're going to be rate limited by human capital first. If you know something about Carbonville today, You can change jobs every six months for a higher salary because there's no people out there. Thing two is infrastructure. You're not moving electrons. You're not moving memes. You're moving molecules. So you need the infrastructure to do that. And some of that stuff like transmission grids, some of that stuff like ports that can ship and receive and move CO two and biomass. Some of that are pipelines, whether they're hydrogen pipelines or CO two pipelines and CO two storage sites, that infrastructure largely doesn't exist. And it's hard to deploy until we have more of it. The last thing that keeps me up at night is the permitting. And permittin…

AI assessment note: “The first of those is human capital... Thing two is infrastructure... The last thing... is the permitting.”

Q electric vehicle charging stations, and that is the demand charge. It's when a utility charges more at a certain time of the day for a lot of demand from a particular facility. It can hit commercial facilities pretty hard depending on the rate. This is a potential major problem for gas station owners, folks who want to operate these networks. How does it work, and why is it so problematic?

A Right. Well, the demand charge has been around for a really long time. It's been one of the main ways that utilities have made money off big power customers for a long time, and it's basically, from the way they look at it, kind of an equitable thing. If you're going to use a lot of power, um, doing so requires beefy electric infrastructure to provide that to you, and the way that the utility recoups that cost is by charging you this demand charge, and basically what it is is For the few minutes of the month, for the few minutes of your billing cycle, that you are using the most power, they, they add a multiplier of a certain number of dollars to your bill, and so if you use a lot of power, it's high, and if you use less power, then it's low, and it's generally not all that controversial because big power users like factories or a farm that has a lot of irrigation pumps, a data center, have the ability to kind of throttle their power use and avoid That sting of the demand charge. But the EV charging station is another thing altogether, because it doesn't have the ability to plan when its big power use is going to come up. They're just sitting there, and then at some point when they can't estimate, can't guess, it may be that every single one of their four or eight plugs gets plugged into, and their power use goes through the roof. And then at the end of the month, they wind up …

AI assessment note: “it doesn't have the ability to plan when its big power use is going to come up”

Q Okay, so he's got this idea that was considered radical when proposed, but over the last 13 years we have seen extreme weather increase, hurricanes intensify. What happens next? Why did this catch on?

A So for 13 years, Dr. Merrill slowly chipped away at his idea. He called it the Ike Dyke, named for the hurricane that inspired it, and slowly it started to gain steam. Based on Dr. Merrill's original design, the Army Corps and the Texas General Land Office put together something called a feasibility study for a storm barrier. That study itself took years and years, and then it ended up in a plan that finally made its way to Congress as a part of the Water Resources and Development Act of twenty-twenty-two. That piece of legislation passed both the House and the Senate this summer. It's still waiting to go to Joe Biden's desk, but once it's fully authorized, then the hard part starts. Construction. As it stands right now, it's a thirty-one billion dollar project that's going to take at least 18 years to build.

AI assessment note: “ended up in a plan that finally made its way to Congress”