The Exchanges

Q you enroll in a program, we, and as that enrollment, you know, we are going to send you a signal, or we're going to call you, as it's been historically, uh, a couple times a year at peak hours, and we're going to ask you to ramp down. This is kind of an extension of Demand response, right? Like, what do you think of as being the same and different?

A I, I completely agree that, first of all, there are, ah, there are natural pricing tiers. So if you're in ERCOT in Texas, you can pay more money for power right now, or you can ramp down and pay less money because power prices are high. Then there are demand response programs, as you mentioned. A utility might say, you can make some money if you agree to curtail during this period. You might own a Nest thermostat and be enrolled in a smart thermostat program, and they will pay you if you are willing to reduce your consumption. There are even mandatory programs where a prerequisite of your enrollment is your promise to curtail and a hefty penalty if you do not curtail. But what's missing from all of these is the ability to offer the service of curtailment at the large scale that data centers could theoretically provide it and get a real benefit out of it. And that benefit is not just a cheaper cost of power or a flexibility payment. That benefit is a larger power connection or faster access To the power grid, whether you're an existing data center seeking to increase your capacity or brand new data center seeking to connect, you should be allowed to harvest the existing stranded capacity on the grid that can be served to you if you are willing to curtail every so often. And that, that product doesn't exist today. And with good reason, you know, the electric utility industry for …

AI assessment note: “But what's missing from all of these is the ability to offer the service”

Q Somebody else might be running the workloads or actually being the customer. So can you walk me through, like, how you think about the stack of who needs to do what? If we're going to deliver on this promise and we're going to take advantage of the hundred gigawatts of latent capacity we've got on the grid by making data centers flexible, like who needs to sign off on what?

A Yeah, it is a wickedly complicated multi-party problem. I'm, I'm delighted, Shail. You got comfortable getting your hands around this, and now we can work together. Look, I go back to an earlier question you asked, which is, what's the most critical thing that has to happen? The most critical thing that has to happen is, Power utilities and system operators and regulators and governors saying, if you are willing to be power flexible, oh data center, we want you in our state, you get to skip the line, you get to connect faster as a flexible load fast track, you get a bigger data center, etc. If that happens, I believe everything else quickly falls in line. Now, you're right. The data center is not one monolithic entity. It comprises a lot of players. You might have, for example, a data center developer, Owner and Operator. I'll make one up. Digital Realty, a terrific one that we partner with, that is operating a data center within which they have a tenant, right? That tenant might be one of the many folks we've partnered with, like Nebius, for example, or Oracle, or Lambda. And within that cloud provider, by the way, it could be a hyperscaler as well. Within that cloud provider, you might then have a customer, And that customer, by the way, may not be the end customer. You might have together.ai or fireworks.ai, which is an inference serving service, which is then serving tokens…

AI assessment note: “You might have, for example, a data center developer, Owner and Operator.”

Q So let's mostly spend our time talking about the details. Um, first of all, at the high level, what is the First Movers Coalition? Like, fundamentally, is it, what is it? Is it an organization? Is it an initiative? Who is it led by? Who governs it? Like, what is it?

A Yeah, um, I'll try out my elevator pitch. Um, the First Movers Coalition is President Joe Biden's flagship public-private partnership to scale clean technologies. It's a partnership between the U.S. government and the World Economic Forum, and it assembles a coalition that today is 66 companies, some of the biggest companies in the world, each of which is making a purchase commitment, an advanced market commitment, to buy, ah, a technology in one of these hard-to-abate sectors in twenty-thirty. And that set of commitments amounts to about twelve billion dollars of purchasing power, purchasing commitments in 2030. So that's, in essence, what the First Movers Coalition is. And since we launched it, we've actually brought on additional government partners. So it's no longer just the United States and the World Economic Forum. Even though we created it, now we have nine government partners. India and the United Kingdom and Germany, et cetera, and Sweden. So, in effect, it's a, it's a consortium of governments, And companies, and the companies are the ones making the commitments to buy these emerging technologies.

AI assessment note: “The First Movers Coalition is President Joe Biden's flagship public-private partnership to scale clean technologies.”

Q process, I know I want to get into some individual sectors, but I'm too interested in this question. Um, to what extent is the recruiting process around companies sort of, uh, accelerated by a, like, keeping up with the Joneses sort of effect? Like, you're like, Ford said they're going to do it, so now you go to BMW and say, well, Ford's doing it, so like, why aren't you?

A It is, I mean, getting that initial seed of companies was so critical, right? Um, there's no way we could have gotten to 66 without the first 24, without the first 12, um, because peer pressure is, uh, is a large part of the recruiting process. Look, when we recruit, um, there's a few different things we can bring to bear. Um, the, the first is we make an appeal to companies' competitive advantage. We say, look, In the future, a decade, two decades from now, you're going to live in a world where policies are net zero aligned. You're going to be forced to buy, uh, clean steel. If you're a first mover and you secure those supplies of clean steel early on, you get some, you exercise those muscles about how to offtake, uh, these new materials. Well, you'll be ahead of the game and you'll have cornered the market and everybody else will pay a lot more for clean steel. So that's, that's one argument we make. Another argument we make is, you know, this is kind of the highest leverage thing you can possibly do, um, compared to all your other Commitments. Um, third is what you said, Shale, keep up with the Joneses, keep up with the, the Fords and GMs, right? Uh, and, and that's, that's proven to be effective. And then fourth is, look, I want to be unabashed about this. We will celebrate you. President Biden got up and said, you know, this is my flagship initiative at COP-VI. You know, t…

AI assessment note: “peer pressure is, uh, is a large part of the recruiting process.”

Q of various policy mechanisms, but also just like in the global north versus the global south where most of the, most of the growth in manufacturing is Is occurring. A lot of the demand growth is occurring, and this is inherently this, like, global effort that you're undertaking. So, um, like, what else needs to happen for this to play out concurrently in the global north and the global south?

A Totally, and this is a big fault line. You know, in, in the last two years of traveling to countries, you know, I led a delegation to South Africa, and what the South Africans want to do is they want competitive, clean industry. Um, so they want to invest in their Green hydrogen exports or green steel exports. Um, here comes the United States IRA, and basically the only thing we heard from countries in Egypt was, um, hey, you guys are now the most competitive producer of these clean products in the world. What about us? Um, we even heard from some emerging economies who asked, could you carve out some of the IRA funding for, you know, clean hydrogen projects in this developing country rather than only in the United States? Um, And, and so I think it's, it's super important that, uh, we, we be receptive to the, the requests of these countries. Obviously, we're all locked in this, you know, great global competition, but developing countries, it's, it's important for us to help decarbonize developing countries in these hard-to-abate sectors that will dominate emissions growth. So we have the Development Finance Corporation. It's a U.S. organization. It invests in development around the world. Um, and, and we have a partnership with them to invest specifically in projects, um, Uh, that will have FMC companies as their offtakers. So American, uh, investment capital, our loans, for e…

AI assessment note: “American, uh, investment capital, our loans, for example, will go toward these hard-to-abate sector projects”

Q is and does, uh, Let's talk about the premise behind it, which is it's a, you know, you didn't invent this idea. It's a form of an advanced market commitment. You talk a little bit about, like, what advanced market commitments are and why I know even before you joined the administration, you were sort of hung up on, like, can we leverage this tool to do something on climate?

A Yeah. Well, look, I, as you said, I definitely didn't invent the advanced market commitment. Um, and I was mostly, I was inspired by how successful it's been outside of, Our own field of clean energy technologies. Um, for example, in the global health space, advanced market commitments, which are basically promises to purchase a certain product that meets a set of specifications, even if those products don't exist today on the market. Those advanced market commitments have enabled us to get the pneumococcal vaccine a couple decades ago. Um, most recently, advanced market commitments brought COVID vaccines onto the market, right? It was the U.S. government saying, I promised to buy X doses of this As yet uninvented technology, if you bring it to market and it passes a certain performance threshold, right? It reduces the incidence of COVID by X percent. Um, and these things have been going on for even longer. Um, NASA has been doing this for, like, 70 years, right? In the 19 sixties, uh, NASA provided the early market for integrated circuits in the Apollo mission. And that's why we have Silicon Valley. Um, most recently, NASA provided these milestone payments for SpaceX. So any company that could actually meet these performance thresholds and eventually be a private manned space flight would get payments from NASA every time they met the performance specifications. So that was ki…

AI assessment note: “advanced market commitments, which are basically promises to purchase a certain product”

Q capital that will make the purchases directly on behalf of the parties who've been a part of it. First Movers Coalition is a little different, right? It's not a pool of capital, so the buyers are saying, I will buy X tons of this thing, but they're not putting money into a bank account or committing capital into a bank account for some third party to do the procurement, right?

A So, um, close. First, let me just say, the Frontier folks are fantastic. It's a great initiative they've developed. So, we launched the First Movers Coalition in November, A few months later, Frontier came to market focusing just on carbon dioxide removal. We focused on some other hard-to-abate sectors. After Frontier came to market, we then, we said, look, this, this model is wonderful. Um, and so we added the carbon dioxide removal sector as well, and several companies that have not joined Frontier were then able to join the First Movers Coalition, so together we have large coverage over the corporate world. And the way we structured the carbon dioxide removal commitment, at least for our three champions, um, Alphabet, Microsoft, and Salesforce is almost exactly as how Frontier has done it. It's a dollar commitment. Um, but you're right, Shail, for all of our other sectors, um, for steel or trucking, the commitment is, like Ford Motor Company says, in twenty-thirty, I will buy 10% of my steel using near zero carbon steel. Um, so it's, it's a percentage of their supply chain is the commitment they have made to purchase, uh, in, in twenty-thirty. So that's, you know, a qualitative difference from how Frontier is doing it, which is a dollar value.

AI assessment note: “you're right, Shail, for all of our other sectors, um, for steel or trucking”

Q So start with, um, who's committed to the steel, uh, steel commitments, who's made steel commitments. And then how have you gone about defining what the actual commitment is? Is like what defines clean steel? How much needs to be purchased? Like what are the, what, what is the, what is the nuance that needed to get baked in here for this to be a real kind of a commitment?

A Yeah. Well, so I'm glad you started with steel. Steel is so important. It's, you know, eight percent of global carbon emissions and rising. Like you look at India, um, India's largest source of emissions growth. It's not coal power. It's steel, iron and steel. So, critical for us to decarbonize, um, primary steel, the production of new steel. Uh, so we have 18 companies that have committed to, uh, the steel commitment. They include major automakers, um, I think I mentioned Ford, um, Volvo, Scania. Uh, they include utilities like Engie. They include, um, renewable energy producers who use big wind turbines that use steel, for example, um, Invenergy or Orsted or Renew Power. So the, it's a wide range of folks who use steel and are making this purchasing commitment, actually, therefore, for a range of different types of steel. And you asked a critical question, which is what goes into making the standard by which we call steel near zero carbon steel? We're walking this balance, right? We have to pick something that, uh, is feasible in 2030 for companies to meet 10% of their supply chain from. But, uh, We want to make sure that every single commitment is technology forcing, that we're bringing new technologies into the market. The commitment can't be weak enough that you can use an existing technology and get away with it, right? We want to bring to market these new technologies. S…

AI assessment note: “we have 18 companies that have committed to, uh, the steel commitment.”

Q amount of time, I think, in aviation world, because in some ways it's sort of like, Uh, it's like just far enough away that you could really focus on that next generation of things, but it's probably kind of tight if you're talking about real big volumes. So how did you think about that? And like, what does the First Movers Coalition commitment look like in the context of aviation?

A Look, I personally think the aviation nut is the toughest one to crack. Um, and I think we're so lucky that we've got some really incredible companies. Um, you know, there's United who goes out there and says, we're going to meet our sustainability commitments with no offsets. Big deal. Um, we got Delta, who's already signed an off-take agreement for half of their FMC-compliant, uh, clean fuels. Um, by the way, the, the FMC standard, the first movers coalition standard, is more stringent than the first generation of bio-based SAFs, right? Sustainable aviation fuels. That first generation will get you, as you mentioned, Shail, a life cycle reduction of greenhouse gas emissions of, let's say, 50%. Um, Our standard is 85% reduction or more. It doesn't mean all bio-based fuels are out. You could have next generation, fairly advanced ones, alcohol to jet, fissure tropes, um, processes, uh, that, that in many cases, um, have some benefits that the first generation doesn't. Today's generation is based on waste oils, fats, and greases. As you mentioned, Shale, they compete with food crops. They don't have great greenhouse gas reduction. Oh, and by the way, they don't scale. Um, we're just, there's no way we're going to get to, you know, Double-digit percentages of these, uh, fuels in our, uh, displacing our jet fuel mix. Um, the next generation of biofuels has a better shot at it, but …

AI assessment note: “the First Movers Coalition standard, is more stringent... Our standard is 85% reduction”

Q brief version of, and we'll get into a lot more detail on. How does that job interact with, like, the State Department, which is doing all of our international diplomacy, or the DOE, which is focused on energy? Like, what's the Where do you sit, and how do you know what you're able to and supposed to do if you are, you and John Kerry, trying to make this happen?

A Totally. I mean, look, it's such an important role precisely because so many different U.S. agencies touch on climate, right? Our Department of Energy is developing the technologies of the future. Um, uh, our Department of State, as you mentioned, has diplomats all over the world that, uh, conduct U.S. foreign policy. So it's been critical to have This new role, Secretary Kerry, the first, uh, U.S. principal to actually sit on the White House National Security Council, who is solely dedicated to climate. It's been critical to have him to play kind of this coordinating function. So, uh, when we go, for example, to, to the COP, COP-XVI in Glasgow last year, COP-XVII this year in Egypt, Secretary Kerry leads the entire U.S. delegation, which, uh, in both cases has involved several cabinet secretaries, and, um, we now speak with one voice. Right. We're able to coordinate all of the different things that the U.S. government is doing, both at home, the IRA, for example, and abroad, our, you know, development finance corporation investments in emerging economies. We coordinate all of that with one voice, one negotiating position, and a bunch of really cool announcements that just came out in Egypt.

AI assessment note: “It's been critical to have him to play kind of this coordinating function.”

Q there's the training versus inference distinction, and I'm interested in your perspective on that, but even within a category, even within inference, for example, like what have you learned having run all these demos now about like what types of workloads seem to be the ones where they're Is enough volume, they are a big enough portion of the overall workloads to matter, and where they have the most flexibility?

A Yeah, absolutely. There are many different AI workloads. It's more than just a simple dichotomy of training and inference. Each of these categories has lots and lots of different subtypes. I will say, ah, a good reference, there's a March, 20, 26 paper that was published actually by Emerald's chief scientist, Professor Aisha Koskun at Boston University with Two others, both at BU and at Emerald AI, that shows a 18 to 55% power flexibility opportunity across a range of different representative AI workloads, spanning training, inference, fine-tuning, and all of their subtypes. So, there's a lot of inherent flexibility is kind of my overarching point, and then I can go into the various kinds as you suggested. As you mentioned, We've now done these five demonstrations at data centers with NVIDIA, with EPRI's DC Flex Initiative, and with many other partners from Oracle to Nebius to National Grid. And in each of these, we've tried to reenact real production-grade actual workloads. So, for example, in London, we ran workloads from real models, whether they're, you know, OpenAI's models, Meta's open source model, even an Alibaba Model from China to showcase that we could achieve performance levels that real customers are, find acceptable. Making sure not to throttle workloads that are the customer labels as mission critical or that should not be throttled. While at the same time precis…

AI assessment note: “many of these will be fine-tuning, post-training, training type workloads... like batch inference”

Q starting to emerge. These, the, you know, unique tariffs that data center operators or developers are starting to sign with utilities, I think, Are interesting, and there's a bunch of different structures of those. Let's narrow in, though, on the portion of that that is most relevant to you, which is making data centers flexible assets. So on that front, over the past nine, 10 months, like, what has changed?

A A lot and a little has changed, I'd say. Um, let me put it this way. There is this deep divide that I observe between the level of flexibility Service levels and tiers that the compute industry offers, and the level of flexibility in the service tiers and level that the utility and grid operator ecosystem offers in terms of electric service. And that schism is one of the reasons that this is such a hard problem to solve. That even though, you know, I'm super excited and want to talk to you about the five commercial demonstrations we've done since you and I just last talked at Emerald AI, it's still a fundamental challenge to make sure that we can Take advantage of all of what I call the stranded power on the grid. So to back up, Shail, for those viewers who didn't hear the last episode, data center flexibility, why is it important? Data center flexibility is important because if these new AI factories, as Jensen Huang at NVIDIA calls him, if they can be flexible just a little bit of the time, they can utilize this vast amount of unused capacity on the grid. The grid is utilized at roughly 50% or less during most of the year, and therefore, We have all of this hundred plus gigawatts of stranded power capacity that could power AI factories connecting to the grid today if during those rare peak moments AI factories could ramp down, uh, partially for a limited amount of time. That'…

AI assessment note: “we've seen a lot of movement on the AI and compute side in terms of flexibility tiers.”