← Shift Key with Robinson Meyer
Shift Key with Robinson Meyer
Live From New York Climate Week: The AI and Electricity Moment
01:04:04
Rob interviews Representative Sean Casten about his new energy price bill, plus Emerald AI’s Arushi Sharma Frank.
Loading summary
Transcript130 lines
- [00:00]
A
This is a special edition of Shift Key recorded live at Heat Map House in Union Square during New York Climate Week 2025. On this week's show we look at a new Democratic bill called the Cheap Energy act that will try to put consumers at the center of the energy system. And we talk about how AI is turning tech companies into energy companies and forcing utilities to change way faster than they've ever needed to before with someone who's done both of those things several times in her career. It's my conversation with Democratic Congressman Sean Kastin and AR Rushie Sharma Frank of Emerald AI recorded yes live in New York in front of a studio audience on Wednesday, September 24th. It's all coming up on ShiftKey after this. This special edition of Shift Key is brought to you by Salesforce, the presenting sponsor of heatmap's live taping of Shift Key at New York Climate Week. Let's hear from them now.
- [00:51]
B
We really early on were hyper focused on where are impacts, where does this data sit and how can we automate some of the toughest or biggest data sets that are coming in?
- [01:01]
A
That was Amanda Von Allman, senior director of sustainability, Intelligence and decarbonization at Salesforce. She was chatting with heatmaps Gillian Goodman. Stay tuned at the end of this episode for their full conversation. Hi, I'm Robinson Meyer, the founding executive editor of heatmap News and you are listening to Shift Key Heat Maps weekly podcast about decarbonization and the shift over away from fossil fuels. On this week's show we are talking about AI and where electricity and energy markets are going with one of, I would say, to quote Heat Maps editor in chief and CEO, the most formidable climate minds in Congress. As a scientist, clean energy entrepreneur and CEO and now as a member of Congress representing the western suburbs of Chicago, he's dedicated his life to fighting climate change. He previously before Congress worked in market based solutions to reduce emissions. He's now both on the House Financial Services Committee and and the Joint Economic Committee. He also serves as Vice Chair of the Sustainable Energy and Environment Coalition. Congressman Sean Kastad, welcome to the show.
- [02:08]
C
Thank you so much. And it's delighted. Delightful to be in the Shift Key zone.
- [02:12]
A
Exactly. Now you're getting into it. So we are seeing this world changing. This paradigm shifting moment in electricity load growth has returned for the first time in two or three decades and a lot of it, not all of it, but a lot of it is driven by the AI boom and data centers. You released a bill actually just this morning and we covered it Here at Shift Key. Oh, sorry, we covered it here at Heat Map called the Cheap Energy act. And it's kind of a new mix on how you are thinking about electricity, a new pivot, how you're talking about it. Can you just describe your thinking behind the bill and what it tries to do?
- [02:49]
C
Sure. So we call this the Cheap Energy act because cheap energy is synonymous with clean energy. There is largely, there are very few people in this country who are opposed to clean energy. There are a lot of people who are opposed to cheap energy. And you have to understand that dynamic to understand why it is that we have so utterly failed to build things like grid enhancing technologies that would lower the distribution losses in the grid, like load sided efficiency that would allow us to stretch our electrons a little bit further and as Amory Lovin says, make our beer colder, our showers hotter and lower our energy bills. Why don't we do those things? And the headline answer is because American energy policy at its best has been schizophrenic on whether we should prioritize the interests of producers or consumers. More often than not it prioritizes producers. And so what we've tried to do in this bill is to say we have to deploy things vastly faster than we've been deploying them before. We have to deploy things that are reliable, that are cheap. We have to deploy things that markets want to deploy because at the end of the day government spending is only 25% of GDP. Right. Most of money is in the private sector. And we can either take an approach to build those that says we are going to castigate all of the actors in the industry who are not deploying fast enough and say I am going to compel you to do things that are opposed to your economic interests. Or we could change their economic interests. And so our philosophy is if you get the first P right, that is the profit incentive. The second P permitting is easy.
- [04:28]
A
Do you think it's easier to get the first P right to change how utilities are regulated at this moment of suddenly electricity demand going up, of energy demand going up, is it like a particularly right moment for it or is it just like we should have done this long ago and now is just a good time to do it?
- [04:46]
C
Well, look, I mean it's always the right time to exercise. Maybe it's more compelling when you're just met your cardiac surgeon. The yes, the surge in prices right now I suppose makes this more resonant. But to some degree that's only because it's painful now because we didn't do it so long ago. We're sitting here right now. There are 2,000 gigawatts of generation wanting to be interconnected to the US grid. Right now of that generation, 1 gigawatt is coal, 85 gigawatts is gas. All the rest is solar, wind and battery storage. That's what markets want to build. The reason why markets want to build that is because it has no marginal operating cost. And if you could build something that doesn't cost anything to operate and sell it at some uncertain price, that's a much better investment thesis than building a coal plant where you're saying, boy, what's the price of coal going to be two years from now? I don't know. Am I going to be able to make money? I don't know. So if we had fixed that problem before and gotten that stuff out of the queue, we wouldn't be facing rising costs now. So no time like the present.
- [05:47]
A
Can you just like orient us to what the policies in the bill that we should kind of keep in mind is that the most important changes are.
- [05:56]
C
So some of the. I think the big headlines at the start are let's make utilities partners rather than adversaries. As it sits right now, we've hugely jammed up. The word is performance based rate making. That's the kind of word that only a nerd could love. But the idea of performance based rate making is if a utility is helping you to save money, they should be able to make more money. They shouldn't be forced to say every dollar that I save for you is a dollar that comes out of my pocket. And so we've got a hugely amp performance based rate making section that says let's look at utilities. Let's say are your investments serving to to take congestion out of your node? Are your investments really, for the first time in forever building wires that would lower cost instead of just providing reliability at a higher cost, Are your investments helping your state meet some of their own economic targets? If you're doing that, let's allow you to earn a little bit more return on your rate adjusted return. We've also had a similar provision in that says if you build out, invest in grid enhancing technologies so more efficient superconducting wires, smarter dispatch of your grid. Let's structure a shared savings model so that instead of saying, boy, if I do this, my regulator is going to say you have to pass all the savings to the customer to say, I keep a couple bucks, you get a couple bucks. And now we get utilities and consumers fighting for the same End I guess the last thing that I'd say is we restore in this bill the IRA tax credits that were taken away. Now, a part of that is an equity question, because there isn't a person on this planet who would not like to pay nothing for electricity. But there are a lot of people on this planet who can't necessarily afford to come up with the money to put a solar panel on their roof. So why are we only giving that luxury to people who've got spare capital? The IRA was the way to get through that. The other thing that the IRA did, which goes away in this Republican bill, is that we had very intentionally prioritized in the IRA to make those investments in communities that are left behind in this transition. There is a reason why the West Virginia economy is not doing well, and that's because it's hitched itself to an expensive fuel that can't compete in today's economy. West Virginians are Americans who deserve dignity, who deserve the same opportunity that we do. And part of what the IRA was doing was saying, as we make these investments, let's prioritize those communities that are left behind so that we depoliticize this and so we treat everybody with decency because everybody deserves cheap electricity, not just people who voted one way or the other.
- [08:24]
A
Do you think I want to get back to the utility performance based rate setting a bit A thrilling topic that I know makes everyone's pulse in this room quicken with passion and thrill, but I want to just get at that for a second. Do you think that the ira. Let's talk about the IRA before we go back to the bill. Was the IRA succeeding in that most ambitious of its goals, which was to bring in fossil fuel communities and places that had kind of been dependent on the old economy into the new economy? Or if we had really wanted to realize that goal, would Democrats have had to pass different legislation or add new powers?
- [09:02]
C
I think it was unambiguously succeeding on that goal. And the proof of that is that when you look at where the preponderance of IRA money went, it overwhelmingly went to the politically redder parts of our country. It didn't go to those politically redder parts of our country because there was some political calculus about how to deploy the money. It went there because we said we need to prioritize energy transition communities. And it is whether this is good or bad, it just is. The way that American politics has sorted out over the last two decades is that if you live in an extractive region of the country, you are probably represented by Republicans. If you live in a part of the country that turns lower value materials into higher value goods and services, you are probably represented by Democrats. And so if we didn't have those priorities in there, you would see this giant sucking sound of money out of red America and into blue America. And that doesn't matter that it's red or blue, you'd see this sucking sound because what do you say to someone whose family has worked in the coal mines all their life, who had dignity, who had respect, that that led to a community and first they lost their job, the coal mine invested in labor productivity enhancements like longwall mining. Then they lost more jobs because gas was under competing coal. Now they're losing more jobs because to the extent that fracking came along, it's getting out competed by renewables. All those people deserve dignity. And the IRA money was flowing to those communities with intentionality. And so one of the tragic effects of turning the IRA off is markets still prefer cheap, but it's going to exaggerate that wealth transfer and make a political problem where one shouldn't exist.
- [10:41]
A
Can I ask, I mean, the IRA was funding all the funding redder areas, red states, red districts. And I think when, you know, before oba, before the one big beautiful bill, there were like a couple theories about why that was. One of them, which I didn't really agree with, was that it was going to like bring those red districts more into the Democratic fold. Yeah, it's a hard thing to do. The second thing, the second argument, which I think was maybe had more weight, was that it was going to protect parts of the IRA from repeal and that the representatives for those areas weren't going to want to repeal, say EV policy that was going to employ their district, employ a lot of workers in their district and create investment in their district. But then it's those very parts or some of those parts that got repealed. Why do you think that that didn't work? Or was it just a matter of time?
- [11:29]
C
I mean, it's. Some of that answer I think is into psychology and the people who are involved in the room. And it's tempting to tell a neat sidey story and it's probably a messy story.
- [11:38]
A
Yeah.
- [11:39]
C
But I think a part of the problem is that you have members of the Republican Party, and I use Iowa as an example. Iowa is long on biofuels, long on wind. They have a strong vested interest in getting their renewable energy to market in the rest of the country. There's no one in Iowa who's in a Meaningful position of leadership in the Republican Party. Mike Johnson represents the Gulf Coast. That really depends on oil and gas exports. John Thune represents the Bakken Oil Shell. That really depends on oil production. The folks in the Trump White House who are nominally in charge of energy policy, you know, Chris Wright is a fracker, Doug Burgum, you know, they made their money in these extractive industries. And so you have at the leadership level of the Republican Party this very, I suppose narrowly self interested, but understandably self interested reaction that says I need to make sure A that we continue to consume fossil energy and B that the prices of fossil energy are high. So you know, look at what they were doing as the demand for their products was going down. They were saying we have to build more LNG export terminals because I want to make sure that American gas producers can sell to higher priced markets overseas, which has the added benefit of reducing supply in the United States, which raises the price of gas. The that's a logical thing to do if you're looking out for the interests of gas producers. It's a terrible thing to do if you're looking out for the interests of consumers. But maybe that's good for your local politics. And I think at a headline level it's a gap between where the leadership of the Republican Party is and where the people who live in Republican districts writ large are.
- [13:17]
A
It's so funny because the last big bill that we talked about was the Inflation Reduction Act. It was called the Inflation Reduction act and now Democrats are pivoting on energy issues, I mean, as demonstrated by this bill and we can talk more about it to affordability. But the last bill was called the Inflation Reduction Act. Do you think we talked? And by we I mean I like very much implicate myself as a climate journalist who was running around after the IRA passed and was like, look at this, Democrats passed a big climate bill. They said they were going to do it, they did it. It's like this huge deal that the U.S. congress, U.S. senate did. It. Was there too much talk about climate change after the IRA passed in a way that then kind of wound up dooming the IRA because Republicans who have a particularly punitive outlook, at least the current administration came in and was like, we gotta tear all that stuff out.
- [14:03]
C
There's a lot of second guessing we can do about polling and the right way to talk. And I always hesitate a little bit with that because I feel like the job of a leader is to persuade people to follow you not to do poll tested messaging. That's the definition of following what I've personally always found, and this is going back to being in the energy industry for 20 years, is that I was going out, I had businesses that had a mission to profitably reduce CO2 emissions. I never went into a company and said, you should do this because if you don't, I will be more moral than you are. That's a really dumb sales pitch right now. Maybe if you share those values, okay, we can talk about a values based reason for doing it. But ultimately there's nothing wrong with appealing to people's self interest. It's kind of Marketing 101. And what I always found, even in the private sector, selling cheap energy, AKA selling clean energy, is a much better pitch. If you can say, I'm going to, I'm going to build this project on your premises that's going to save a bunch of money and I'd like to keep 100% of the money for myself. And then they say, I want some money. All right, well, we're having a negotiation now, right? And I think this, I think, and I think the message is much stronger if you say this is about providing people with access to cheap energy. It's also true. It's also why there's so much pushback. I mean, look, the coal industry has lost 50% of their market share in the last 10 years. They're not winning. Oil demand today is the same as it was 15 years ago. They're not winning. They are petrified about the fact that so much clean energy is being deployed that people prefer electric vehicles, that they prefer cheap energy. We are not the good guys are not winning nearly as quickly as we should because of these incentives. But we are seeing somebody who is losing, pushing back against markets, pushing back against competition, pushing back against cheap. And that's the ground we can win on.
- [15:54]
A
It is interesting though because at the same time that clean energy is cheaper, it produces at peak, it produces extremely cheap electricity. We're also seeing in PJM for instance, in the country's largest electricity grid and in Pennsylvania specifically, like coal plants that were uneconomical before because they're only economical. Like when power is 3x, the gas price are now just like turning on and operating every day because electricity is so expensive in PJM now partially, and this is, I'm not disagreeing, this is partially because PJM has really struggled to connect new renewables to its grid fast enough. But it's like we're seeing high power prices kind of bail out coal at the moment.
- [16:34]
C
I think there's a Couple different dynamics. You've. There's one question around, is anybody building new coal plants? The answer is no, because that's a lousy investment. You can't raise money to do it. There's a separate question of if you have a grid where you have a gas plant that's running at half of its capacity and a coal plant that's running at half of its capacity and there's a little more demand on the grid, which do you pull from? And what's happened on the coal side recently? It goes back to this LNG export. But because of the surge to increase LNG exports, natural gas prices have gone up in the United States, which means that gas assets which were cheaper on the margin than coal six months ago, are now more expensive on the margin on coal. So if you look at the rise in coal in the US over the last couple of months is exactly matched by the fallen gas. It's not swapping with renewables. Renewables still kick both their butts. Yeah, but if you haven't built those assets yet, I mean, there is not a solar plant in the country that only runs when it's affordable to run. They run every hour the sun is shine. There is no wind turbine in the country that says, boy, I better turn off. Windmills are getting a little expensive right now. Like they run all the time once you build them. So as you build those, you squeeze the other stuff out. But those, the coal, the gas that sits in dispatches on fuel in the margin. That's where the swap is happening right now.
- [17:51]
A
Earlier you said something that I want to go back to, which was that our energy system doesn't reward cheap energy and it hasn't been set up to reward cheap energy. What did you mean by that?
- [18:02]
C
So at a high level, no market left to its own devices will reward cheap things. Because if I'm a buyer, I want to buy things for cheap. If you're a seller, you want to sell things for a lot of money. I remember my dad when I was a kid had a little paperweight on his desk. There was an oil barrel and on one side it said, relax, the price will go down. And on the other side it said, relax, the price will go up. And depending on which side of a negotiation you were on, that was how you formed the oil barrel. What's happened in the energy sector that has made that hard is that we have, because it is such a highly regulated sector, we vastly over advantage the producers in what would otherwise be an even negotiation. So for example, if you as a Consumer want to put a solar panel on the roof of your house, you have to get the permission from your local utility, who's going to lose the revenue, who can raise all sorts of technical objections and do that. If you have a solar panel and you say, boy, there's hours when I'm making more power than I want or than I need, maybe my neighbor would like to have some of my excess, well, you're not a regulated utility. You're not allowed to do that. Your neighbor can't buy it from you. These are because of laws we've set up that says only that utility has the right to do it outside of the electric space. We've had these things like, there's a law that's been on the books since 1935, the Natural Gas act, that says that you cannot build gas export facilities in the United States unless it is in the national interest. Is it in the national interest to raise people's price of gas? That was never specified in the act. And so when the Trump administration went through and approved all those assets, which, by the way, the Biden administration had shut down in part because they said it's not in the national interest, they said, well, we think it's in the national interest to look out for our gas producers. Somewhat more recently than that, when the price of oil collapsed during COVID In April of 2020, Trump called the Saudis and said, we are going to withhold military aid from Saudi Arabia unless you raise the price of oil. The Saudis flinched and the price of oil went up. And he was praised on the COVID of all the business magazines as saving our oil industry. Why didn't we do the same thing two years later when everybody was complaining about the price of oil being so high? And we had a Democrat in the White House, right? So we've always had this feeling like, I need to look out for producers because the producers have had more political clout. We've connected those things together. And you can be angry about that, you can be embarrassed about that, or you can see it as an unbelievable opportunity to generate a tremendous amount of wealth, to lower energy costs and, oh, by the way, cut a bunch of CO2 emissions.
- [20:31]
A
What are the structures in the utility industry that keep us from realizing cheaper electricity? Because I think one thing my colleague Matthew Zeitlin has written about the formation of electricity markets. One of the ideas behind electricity markets, when they were built, was like, this is incredible. We're going to lower costs for consumers because the utility model, the monopoly utility model, isn't working. And utilities are overbuilding. They're charging everyone for too much. Like what structures exist today that are keeping us from realizing cheap electricity.
- [21:01]
C
So the monopoly utility model largely still exists. Most of the utility sector in the country runs on a cost plus basis. I'm somewhat oversimplifying, but basically you're allowed.
- [21:11]
A
On this podcast if you want to wonk out on cost plus.
- [21:14]
C
You're very much allowed if you just understand that if you broadly speaking, if a utility lowers its cost of production and they have a good regulator, that regulator is going to compel them to pass those savings loan to their customer. If their costs go up, they pass them through. And so there's a disincentive that's there. The when we deregulated markets in the 90s, the idea was to bring in a bunch of private capital that would compete against those. And those efforts were actually phenomenally successful. You saw in fairly short order the nuclear fleet in the United States that was only running about 60% of the time all of a sudden went to 90, 95% of the time, displaced a ton of carbon out of the system and it ran more often because it was cheaper. You saw a huge amount of investment in combined cycle gas turbines that are much more efficient than simple cycle gas turbines because they were more efficient, would use less gas and they'd compete out. You then saw a whole lot of those private actors go bankrupt because in a market that prices power to the and this is also true in oil markets, you price to the sort of the highest marginal cost provider that clears the market. And I'm trying to find some way to and non economics jargon, but how much do you need? Who can provide you with just enough to meet the supply and then they set the clearing price for everybody else. What that meant practically is that as people deployed cheaper energy, it lowered the price, which meant that all of a sudden people weren't getting the kinds of returns on their investment that they wanted. And so there was a real clawback in the energy industry where the private sector guys were saying, these assets I built that are still running are not giving me enough to pay off the debt that I had. So that they're good assets, they're still a really important part of our grid. But capital providers got a little bit spooked. And we've not really gone through the effort to say how do we restructure our markets to still use market efficiency to still provide those incentives. But to get people to invest in a world where we could have a world where every single generator has Zero marginal operating cost. If every single generator had zero marginal operating costs, no one would have an incentive to build anything under the way our markets are structured right now. That's a fixable problem. Good creative people can solve that with more time. I could explain to you how we solved it back in my own life as a regulated utility CEO, but it just takes a somewhat more creative approach to the problem that we've not done because it was easier to say, you know what, let's just keep the producers wealthy as long as they maintain our reliability, which they're very good at doing. The electric good is an amazing thing. Amazing engineers who run it, we have incentivized them to provide reliability. We have disincentivized them to provide cheap.
- [23:45]
A
Well, I was going to say that your description of investment being crowded out or investment fleeing the space because a lowering power price can no longer support investments made under a higher price environment does sound a lot like problems faced kind of chronically by the solar industry. Just in that if you build a lot of solar, eventually you're just. Your solar pants are cranking in the middle of the day. It's awesome. And the market price is zero or negative. You have to export a lot.
- [24:11]
C
Yeah, no, it's. Right. And yet the consumers all benefit because the price is falling.
- [24:15]
A
Right.
- [24:15]
C
So just how do you get that incentive? And that's, I mean, I think that's a problem in any capital intensive commodity industry. People have come up with creative ways to solve that. We're just overdue to do it in the energy sector.
- [24:25]
A
You were describing a rate based utility model and I'm actually going to ask just you to walk out a little bit on that. What would we need to change in the US So that. For instance, one thing we recently discussed on a recent episode of Chipski was that in the uk, in the European Union, if you are, as a utility want to build a big piece of infrastructure, you kind of get handed a cost target. And if you come in below that cost target, you get to save the money. But if you come in. But here in the US it's a cost plus model, as you were saying. Just like what needs to change about, like how we regulate utilities legally or at the federal level could change to make this happen.
- [25:04]
C
So two different pieces of the cost puzzle. One is what's your cost of construction for an asset? And I struggle a little bit with that one. Just because, you know, when the Seabrook nuclear plant was incapable of coming in on time or on budget and public service, New Hampshire was allowed to go bankrupt. That basically ended nuclear investment in the United States because the Public Service New Hampshire said, your costs are too high from a climate perspective, that was bad. And we need to have the whole nuclear conversation here. I think there's a separate question about what are you doing to lower your operating costs and what we did, what I told you when I was a utility CEO, we convinced the New York State Public Service Commission to craft a rate for us where we were structured as a for profit co op, where we looked at all of the historic energy efficiency of the assets we were running and said, if we will bill the consumers at those historic efficiency targets, but if we can invest capital to drive the efficiency up, we'll get a preferred return on our capital and then we'll share 50 cents of every additional dollar with our customers. And all of a sudden our customers had an incentive to bring to us ideas that they had about how they could run better. Maybe they wanted to put the solar panel on their roof and they could say, I don't really have the capital, but I've got a good place for this. Would you put up the capital to do this? Sure, we'll share in the savings. Within four years of putting that in place, we had doubled our profits and cut our revenues by 20%, which is to say the, the customers got a 20% rate cut and we got a doubling of our profits because all of a sudden we were aligned. And that was looking at those at your marginal operating costs, which, by the way, is the thing that generates the CO2. Right. Because that's where all your fuel costs are and say, let's just go in and steadily drive that down. Instead of saying, I got this locked in and if I save money and the utility finds out, I'm going to get busted, or the utility regulator finds out I'm going to get busted.
- [26:51]
A
One thing that's fascinating about this moment is that electricity has emerged as this big political pain point. We're so used to oil and gasoline being the big pain points, I think, and being the big, like politically salient energy prices. But oil and gas are like pretty cheap at the moment. And yet we're having an energy crisis like conversation because electricity bills are going so up, are going up so fast. Just as someone who like, thinks about electricity thinks about it in this bill, thinking about it in your career, like, what's going to be different about a world where electricity is like the political pain point input rather than oil and gasoline? What makes electricity different?
- [27:30]
C
I'm not sure which is the chicken and which is the egg. There's this weird dynamic that maybe someone smarter than me has answered this question. But for 100 years, our electricity consumption in the United States grew pretty linearly with GDP. 2008, economic downturn comes and electricity load totally flatlines. And for 15 years or so, electricity load didn't grow. I've never heard a really compelling story about why that happened, but it got us accustomed to a world where you didn't have to build much new stuff to meet load. And so the cost of serving that load, there was less and less debt, less and less equity to amortize. And so we were just stretching old assets out farther. And it got cheaper for the same reason that a car you've paid off is cheaper to run than a new car. Starting as we rebounded from COVID we just saw this surge of growth in electricity. Some of that is data centers. Some of that is that electric vehicles are way more fun to drive. It's the fastest growing vehicle segment. And so that's a direct shift from oil to electricity. Some of that is some of the backlash against globalization that's brought manufacturing back. But our economy is just becoming more electric intensive and now needs a lot of stuff built. And the reason I say, I don't know if it's sick in the egg, that's the reason why we're suddenly having all this economic pressure, because it's not just the generation, it's the wires that are not connecting. And you mentioned pjm, they're not getting the wires to decongest the nodes that are out there. And part of the reason they're not building those wires is because the people who ultimately have the approval. Right. To build that wire are making a lot of money at the current power price and would prefer to. To make a lot of money to a little bit of money. I understand that. I feel the same way.
- [29:11]
A
Yeah, right. Same. Yes.
- [29:13]
C
But that congestion that's there and figuring out, okay, if that's our new reality, we're going to have to build stuff and we're going to have to build cheap stuff. So let's take off the barriers that exist to building cheap stuff, because that's the new reality we have.
- [29:26]
A
I want to just lean into something you were saying earlier, which is that the Natural Gas act doesn't discriminate on price. Right. It just says you can export or you can build gas infrastructure if it's in the public interest.
- [29:38]
C
With our bill, it will.
- [29:39]
A
What?
- [29:39]
C
With our bill, it will.
- [29:40]
A
With your bill, it will. Exactly. So in the cheap energy act, it basically does discriminate on price, which effectively would shut down liquefied natural gas exports from the US or at least keep us from building more terminals. Is that a fair description of what the effective outcome of the bill would be?
- [29:58]
C
Well, in the current moment, yes, although I would say that wasn't our goal. What we said is the Natural Gas act says you can't build these assets unless they're national interest. Doesn't define it. So every administration over their politics can decide what the national interest is. What we've said in this bill is it is in the national interest to lower consumers prices. It is in the national interest to look out for our national security in the world. And it is in our national interest to lower CO2 emissions and pass along a stable planet. One could imagine scenarios like say right after Russia's invasion of Ukraine where there was national security value in us sending liquefied natural gas to Europe and helping them wean themselves off Russian gas. That is not this moment. The Europeans have weaned off that. They've moved on. All of the marginal LNG that's being built is being sold into commodity markets for a higher price. But what we're doing is not saying thou shalt never build another LNG export terminal. What we're saying is if you can't justify this on climate grounds and you can't justify it on domestic price grounds, and you can't justify it on national security grounds, then keep that good red, white and blue American natural gas in America underground if possible.
- [31:11]
A
Last question. We've been talking a lot about the Cheap Energy Act. I just want to raise the specter of another piece of legislation that could come up this term, which is a permitting bill, perhaps a bipartisan permitting bill. I guess it would need to be bipartisan in the Senate for it to get through. Do you think that their Democrats should embrace a bipartisan permitting bill this term in Congress? What would it need to include to get you to the table?
- [31:34]
C
Well, so first of all, we have talked all about the economic incentives in our bill. Our bill actually is, I think in terms of word count is probably mostly about permitting.
- [31:43]
A
And it has incredible transmission.
- [31:47]
C
Pieces making sure that we bolster interregional connections, making sure we enhance minimum connections between different regions of the country, doing things to standardize some of our environmental compliance. So if you provide environmental data to nepa, CEQ also gets it. EPA also gets it. You only have to do that at once. One of the most important things I think it does for permitting is to give the federal Energy Regulatory Commission, the same authority over transmission that they have over natural gas, which is to say you are the sole source party to regulate this. Everybody who supports objects to a project can submit their objection. But you can't lose in one county and then pick the same fight in the next county and then pick the same fight in the next county which strings these transmission projects out interminably. So all of that is in there. I think on the question of bipartisan permitting reform, there was an effort on bipartisan permitting reform in the last Senate, which was the Manchin Barrasso package that included some good things for transmission. I would submit to you that it didn't deal with any of the profit incentive problems and it didn't deal with some of the really hard political questions. We've got to solve cost allocation. Our bill has a way to solve cost allocation. But if you're going to build a wire from a wind excessive part of the country to a high priced part of the country, who should be responsible for building how much of that wire? That's a political problem. We've got that in there. Mansion Brasso didn't deal with that, didn't deal with the interregional.
- [33:07]
A
It's a huge deal. It's a huge deal and it's all in the bill. I should credit that.
- [33:10]
C
But Mansion Brasso also had huge gifts to the expensive energy sector to lower their costs as well, many of which were already granted in the big ugly bill. So for example, in the big ugly bill they wanted to cut judicial review for dirty energy projects. In the big ugly bill it says that if you pay a million dollars, you can basically buy your way out of judicial review. So if you're going to talk about like we should have a bipartisan negotiation, sure, but they just picked our pocket and they got a lot of money out of our pocket when they picked our pocket. It is to argue that it's hard to build fossil assets when we are building such a huge ramp up in LNG facilities. When in any given year we build 10 to 100 times as many miles of natural gas pipeline as we build transmission, they're already getting plenty and they're losing market share. Right. And remember what I said before, of all the stuff that's in the pipeline that people want to build, it's overwhelmingly the wind, solar and clean stuff. So you're basically saying like is there a bipartisan play where we could prop up a failing industry rather than allowing people to build the stuff that they want that's cheaper? And if the answer is yes. Because that's the only way to get the votes. Okay. But let's at least acknowledge that they already got a lot of that stuff. And if they're saying that's not enough, maybe it's time for them to eat at the little kids table for a while.
- [34:34]
A
There's so many more things to talk about. There's so many threads to pick up. You have to go to an event across town, and so we're gonna have to leave it there. But Congressman Castin, thank you so much for joining us.
- [34:43]
C
Thank you so much. Pleasure to be on the show.
- [34:48]
A
Thank you. Thank you. And with that, I want to bring in someone who you hear only in the luckiest. Thank you so much. Of shift key episodes, our producer and my friend Gillian Goodman here to have a conversation with our wonderful sponsor for the evening, Salesforce.
- [35:13]
D
Thank you so much. I'm very pleased to be here today with Amanda Von Allman, the senior director of sustainability intelligence and decarbonization at Salesforce. So I want to start actually by talking about Salesforce's own experience with decarbonization and own experience with a product you call the Net Zero Cloud. Where did that originate? Like, what is it? And where did that originate?
- [35:38]
B
So, you know, Net Zero cloud is our technology platform that really enables our customers to measure, manage, and really track progress against their sustainability goals. And it was something actually born out of our own sustainability team. So I sit on our corporate sustainability team. I've been there a decade. We are no different than any other corporate sustainability team. We were toiling away in spreadsheets, right. And just getting lost in them. And it was all of us as sustainability practitioners, right. Got into this because we want to have impact and we want to actually institute change, not just measure and measure and measure constantly. And so we thought, there's got to be a better way. And so we had this sort of light bulb.
- [36:20]
A
Aha.
- [36:20]
B
Moment of, well, we work for a tech company that has a platform. Like, why aren't we leveraging this really cool technology? I think, in part two, right. And I think we get later to it around skills, Right. As a sustainability practitioner, I wasn't really entirely sure where to start, but we did. We sort of built the first, you know, I say version, sort of duct tape it together of using our digital platform, our core technology. And then over time, we had customers, peers, friends just asking us, like, what are you using? Right. And we'd share our experience. And eventually it became our product, which is Net Zero Cloud, which is really cool because it's now Part of our core platform. So if you already, like, are a Salesforce customer, you can pull in sales service all this other tableau data as well through the platform, which is really cool.
- [37:04]
E
Yeah.
- [37:04]
D
So speaking of beating your heads against carbon accounting, we actually heard earlier from Katie Ross of Microsoft who said that she spends a lot of time thinking about carbon accounting, which made me think of you and the conversation we were about to have tonight. And so how can AI help sustainability teams work more efficiently and effectively? How can it sort of cushion the blow, as it were?
- [37:26]
B
Yeah, well, I love Katie Ross, she's amazing. Known her a long time and I do think a lot about carbon, but I do think it's important not to get, as they say, lost in the sauce. Right. It's important to measure, but you can also kind of go swing the pendulum too far in the other direction where you're kind of chasing perfection, everything. So, yeah, back it up a little bit. So after the, after it became a product and not something that we were sort of managing, we then embarked on a really cool journey, which is we're customer zero of this product now, so we're using it day to day to manage our own emissions. But what's really cool about that, right, is we're giving that feedback loop to our own product team. We're saying this is the way practitioners would actually use this product. Not necessarily sort of a product team who may have initially thought, oh, this would be something that would be good for our customers. And so it's been this really awesome feedback loop. I've learned a ton about product development alone and just really all the different tools, technology at our fingertips. And that investment, that focus on customer zero has really started with this investment in our data strategy, which is important to AI really. We really early on were hyper focused on where are impacts, where does this data sit and how can we automate some of the toughest or biggest data sets that are coming in. So one of the sort of first kind of at the time seemed like a weird move, but has paid dividends is I stopped hiring Carpet Encounters and I hired a software engineer on my team and was like, tell me what you know about how we employ data lakes, data quality and piping automated data in. And he really helped us sit down and think about a really thoughtful architecture strategy for the way we get data in. And at the time it just seemed to make sense to me because admittedly I need to do reporting, I need to make sure it's accurate. But then AI hit the scene and agents and we were so ready for that moment because our data was in this really centralized place, it was trusted, it's coming in real time. And so pretty much instantly we were able to start deploying AI solutions. Whether that was having it do data checking and gap filling or even just querying data in natural language to make it easier to get some of the data. Because even though you have a beautiful big repository of data, you still have to know where it is, you have to look for it. And so then about a year ago, Salesforce launched Agent Force that really made it easy out of the box to build an agent. And you know, it's no surprise, like I don't know how to build that stuff. Right. And so some of the ways that made that really easy is we instantly built an agent on top of our instance of Net Zero Cloud and started just querying it like, hey, how many. What is the biggest source of my short haul travel emissions? Which data center is using the most energy? You still would have to live. But it was able to pull all that instantly for us, really saving my team tons of time because we could ask these questions over again and frankly, when other parts of the business needed this information, we could just say, hey, log in to Net Zero Cloud and ask the agent yourself. Right?
- [40:21]
E
Yeah.
- [40:21]
B
Freeing up our time.
- [40:22]
D
And you're also talking about being more proactive than reactive.
- [40:25]
B
Absolutely, yeah.
- [40:27]
D
So obviously, you know, as has come up today, when we talk about sustainability in AI, often we're talking about the power demands of AI and the mix on the grid. How are you thinking about the cost benefit of AI products from a sustainability perspective?
- [40:43]
B
Yeah, of course. I'm glad you asked that question. I'm so excited about how, how much AI can really augment the way sustainability teams do work, really free us up to do much more strategy and proactive thinking. I think there's like so many cool possibilities there and I'm excited. But I'm also in charge of thinking about decarbonization and I have to think about the other side of the house of what do these tool, this new technology will potentially do to power demand and to the environment. I don't just measure all of these emissions and then lob it over the fence for somebody else to figure it out.
- [41:14]
E
Right.
- [41:14]
B
We have to take it all the way through. And so we do think about the other side of that. And we recently put out our sustainability, our AI sustainability outlook. You know, definitely encourage you all to check that out. But it's really centered on three sort of primary principles when we're evaluating AI and its sustainability. Right. The first is really just looking at are we using the right size models? We don't need a sledgehammer to crack a nut. Right. If they were doing simple calculations, let's use the right size model to do that. Let's make sure that we are asking ourselves the question, like, is this the right application of AI? Are we just doing it to say we did AI? Right. Like, let's be really smart about that. And then efficiency, right?
- [41:54]
C
There's.
- [41:55]
B
We've learned a lot from. In general, data center efficiency has been a really, I think, in general, a success story over the years of just really focusing on that. But how do we ensure that we're taking that to AI? So, you know, chip efficiency, making sure, obviously that we're focused on data center efficiency. The way we code and build these products actually has a tremendous. So green coding is emerging and something that you're seeing a lot of organizations focused on. And then of course, clean supply. Right. We have even the most efficient AI on the planet uses resources and we need to be really cognizant that those resources are coming. So energy and water. Water's a big one.
- [42:30]
E
Right.
- [42:30]
B
We don't want to forget that one either. But, you know, are we using renewable energy? How can we add more renewable energy to the grid? How can we make sure that these GPU clusters, data centers, are not located in water stress regions to begin with? Right. How can we recycle water?
- [42:43]
E
Right.
- [42:43]
B
So we think about all of those in harmony with. Also, again, what are the tremendous benefits that AI could potentially give us?
- [42:51]
D
So you've been at Salesforce for a decade. You've been doing this work for a decade. How have you seen attitudes towards corporate sustainability change? Or I guess, first of all, have you seen attitudes toward corporate sustainability change in that time?
- [43:04]
B
Yeah, yeah. It's hard to believe it's been 10 years. I mean, change is constant. Right. Things are always ebbing and flowing. But I would say is that Salesforce, what's been really fantastic is pretty early on we baked in sustainability as a core value of our company. And that has made it that sort of no matter what's going on, politically or economically, whatever's going on, we can really band together as a company that this is something that matters to us may look different, we may have different areas that we're surging, but we're all united on this fact that sustainability matters to us as a core value value of our company. Our values don't change. And so it's been the sort of no matter what's going on? The thing we hold on to and really make sure is permeates all parts of our work. So I work a lot with our business partners across the business infrastructure, you know, data centers and real estate, business travel, and then of course, supply chain, which is its own. You could have your own talk just on that alone. And these are a lot of disparate groups with very competing priorities, Right. They're all trying to manage their own departments. And we can unite on this thing of sustainability that we do care, that we do need to prioritize this. And so this is how we're able to really sort of change is constant, accept that. But, you know, sort of anchor yourself on something that we all agree, which is we don't want to pollute the air. We want a sustainable world.
- [44:19]
D
So for companies that are just starting to think about how to prioritize this, what would your advice be? Like, where should a company start integrating sustainability into their data process processes?
- [44:29]
B
Sustainability or like AI on top of sustainability?
- [44:32]
E
Oo, both.
- [44:33]
B
I feel like hopefully most people have already started implementing some version of sustainability. But when we talk about how do we can we use these cool AI tools and technology for my team? Well, first off, again, invest in data, right? AI is only as good as the data you pointed at and have available to you, so invest there. But for my team, again, I kept all my team, our traditional, with the exception of one, sustainability practitioners, and we've really committed to continuous learning and skilling up of all of this new technology and skills that are at our fingertips. So my team, you know, I have somebody on my team who's a traditional carbon accounter analyst. She's been spending time using AI coding assist tools like Gemini to build scripts to clean up data sets to save herself hundreds of hours of time. And she had no experience with that, Right? So. So these things are now more accessible than ever, but they require you not to sit on the sidelines, Right? You have to spend the time, skill up, learn, get outside of your comfort zone. We have things like Trailhead, which is a free online learning platform that Salesforce has. But there's like a million other amazing resources out there that will teach you how to do basic Python coding and other things that are actually really valuable when you're trying to automate your work. And again, focus on the stuff that really makes you happy with is the strategy and really getting into that next level. Yeah.
- [45:52]
D
Last question before we wrap up. For someone who wants to do this kind of work with the benefit of AI, like what skills should they be trying to develop and you know, on the hiring side, what skills should people be looking for?
- [46:02]
B
There's some basics of like, hopefully all of you have taken, like basic prompt engineering courses. That's one of the first I've done. Because how you prompt and get data out of AI is also an art in itself sometimes. So, yeah, I think just learning. What do all these terms mean? You know, nowadays when I'm interviewing and I say, what kind? How are you using AI, AI tools? I think saying using chat GPT occasionally is not enough. Right. Like I said, how can you understand prompt engineering? And you know, I said writing basic scripts to clean up data. There's a lot of different tools and out there, but I think it's just being really aware. I mean, the other thing is this stuff is changing hourly. Right. I feel like I wake up every day and there's new tools and resources and things to read. And I think we're in this time where we need to really mind share. We need to share as sustainability professionals what's been working, what's not, what have we learned, what cool resources have we leveraged to skill up and really share all this stuff with each other. Because I don't think I've seen like a. This pace of change in terms of what's become available is the fastest I've seen.
- [47:02]
D
Yeah, it's really exciting. I think that'll do it for tonight. Amanda Von Allman, thank you so much for joining us on Shift Key. And now we'll welcome back to the stage Rob Meyer with our final conversation of the evening.
- [47:22]
A
Y', all, we're almost there. I am the last thing standing between you and an open bar, but we have a really fun conversation for you. I want to welcome Arushi Sharma. Frank Arushi, you began your career in gas and infrastructure permitting. You worked in generation at Constellation and Exelon when you were swept up by a small and economically insignificant company called Tesla, where you worked on the thesis and strategy and execution for basically what Tesla virtual power plant and the wholesale electricity business in Texas.
- [47:58]
E
The first wholesale company the Tesla ever had, the first retail energy provider they ever made, and all the stuff in.
- [48:03]
A
Between, which is very important. And we're going to talk about. And now you are an advisor to a number of companies, including Emerald AI. It's so great to have you here on Shift Key. Welcome to the podcast.
- [48:15]
E
Thank you.
- [48:16]
A
Let's. So I want to just start here and then we're going to talk about your bio and different parts of it. But like what's your role at Emerald AI right now? And, and what is Emerald AI doing? Why is it important?
- [48:27]
E
Yeah. So you're, you've asked a lot of folks about this. You asked Congressman Casten about this, about utilities and the power grid and it's so like weird and slow and old school. Like what are we, what are we doing with it? My role as an advisor to Emerald AI is to basically be like the excuse, like the random ethnic reference, but the shadi.com matchmaker. Have you been on the website?
- [48:47]
A
I have done a profile. I have been on shoddy.com. can you, can you demystify shoddy.com?
- [48:51]
E
Yeah. Like imagine you took Bumble or whatever and turned into a full on resume of yourself and you stuck put it on a website and then like people come and find you and try to make the deal happen. I'm like the auntie at the back of it. The no is that the two of you have to get married. These are two of you are like the tech company in the grid. It used to be like Elon and Patty. Poppy too. It could be anybody. But they're just extremes. But they actually can and do and should work together. So my role for Emerald and any other company I've advised since leaving Tesla has pretty much been make them work together. And that's real. Right. So the coolest thing that to know about Emerald right now in this very moment is that one of their next demos is going to happen in Illinois, smack in the middle of the PJM service area, where they will get a data center to demo its actual flexibility and grid response.
- [49:37]
A
And just so I think I'm aware of this to some people what Emerald AI basically does the data center flexing is this right? That we've been told is possible, but that so far I think only Google has really said yes, we think we can do this. We can imagine it being feasible. But right now it's not deployed in many places. But Emerald AI is beginning to play.
- [49:59]
E
Yeah. And look, there's lots of little weddings happening on the sides too. Yeah. So the way that like data centers and you read, you can read like Varun Chevron has this or you can listen to his pod with Shael Khan, recently CEO of Emerald, who is Varun is the CEO. Yes. Of Emerald and Shell is at Energy impact. And one of the things that they say on that pod that I say a lot, a lot, all the time is even the grid, there's this like mismatch that needs to be fixed with a nice little marriage which is the way that we plan the grid is not the same way we operate it. Same way the way we are planning for how data centers are going to behave is not how they actually operate. And there's certainly not homogeneity there. And there's even going to be less homogeneity with every new cool thing every one of the hyperscalers does. And so you're solving for two different sets of problems. And I'm a very busy auntie.
- [50:48]
A
What is an example of the way that data centers are planned for and then the way they actually operate on the ground grid?
- [50:55]
E
So inside the data center, the kind of workload, like the actual jobs that are being done are just one little piece of it. There's all the other stuff in front of it that is going to fundamentally change how much power consistently is happening in the actual compute workload and how much is happening in the balance of plant as they call it. And every time you see a new innovation get announced about Nvidia and AMD and whatever, basically that balance of where the power is actually getting consumed the most is changing and it's changing super rapidly.
- [51:24]
A
And is AI managing that? Are you basically using AI to manage the power consumption from AI?
- [51:30]
E
Emerald is using an AI based orchestration tool, conductor to do that. Yes.
- [51:35]
A
Your career has been at this really interesting and I think rare intersection so far, which is the tech industry, Tesla and utilities. The engine of economic growth in the United States and what has historically been our slowest, among our slowest but most beloved utilities people, industry. What have you learned at the intersection of these two industries so far? And like, as more people begin to pile into that nexus, like what should they know about these two industries and how they work together?
- [52:11]
E
Yeah. So one of them is that the utility industry has a lot of slow capital behind it on purpose. So a lot of things like Congressman Cast and talked about like how markets work and what they're incentivizing and all that comes down to fact that like insurance funds decided originally they're going to fund these very low, slow old school returns. So like a poll is the kind of cool investment they think is interesting. Like a poll, like an actual poll?
- [52:37]
A
Yeah.
- [52:38]
E
You are excited as an investor in a utility if the thing that the utility is doing is buying lots of poles, putting them up and waiting 50 years for them to degrade into the ground, which is called running to failure.
- [52:50]
A
That's like the coolest investment you can get as a utility investor. Yeah.
- [52:54]
E
And the more poles and stuff you have, the better pole they Love it. And they also want it, by the way, from an investment standpoint, you want those poles to fail. You don't want to worry about maintaining them. The last thing you need is to spend a bunch of money going out there on trucks like maintaining all your poles and spending money and making them composite. Like these cool companies are doing composite storm resistant poles and infrastructure. Right. For utilities. From like the investment standpoint of what that thesis was and still is to a large degree, that's not cool. I want to spend more capital. I want to make a return on the capital cost of this thing. If you start maintaining it and extending its life, like what's even happening? Like what?
- [53:31]
A
Yeah, right, right. Then you're just, you're operating costs, utilities.
- [53:34]
E
Right. And it's like that's crazy. Right Here in New York there are some really great util managers who get angry and call me sometimes when I'm like in a different state or country. And they're like Arushi can get my, get my financial accounting and planning people to understand that we have to do ds, we have to do distributed energy. And it's because like the way you value that stuff, it's fundamentally about. Yeah. First of all, capital is coming from the private sector, not from the utilities funded rates. And then on top of that these things last forever. Like from like a standpoint of they don't really depreciate, they just keep providing lots and lots of value. So FP&A folks looking like, what is this? I can't work with this. It's not breaking fast enough.
- [54:15]
A
And just can you like contrast that with like when tech people show up, do they understand any of this? Why not? I feel like we understand but just like spell it out.
- [54:27]
E
Yeah. So you know, the utility side, the tech side is quite the opposite, right? Yeah. I want you to imagine for a second what it's like to be the strategy. Head over energy at a company that does car Stu. That's the mismatch. It's the same mismatch. Yeah. If you want to make money on vehicles, you roll the darn things off the lot as fast as possible. Near end of quarter before your earnings call. That's tech too. Vehicles is tech with energy. It takes me a year and a half, which is like seven years faster than everybody else to get a giant megapack installed and correctly permitted and then launched in Texas. The mismatch is so huge, it's like tech capital does get deployed quickly. And when it comes to products you and I can buy in our homes and Put in our businesses, we're pretty fast about revealing our preferences and saying we want it. But we've never actually asked for energy and energy related stuff, including compute as a value form of energy the same way we haven't asked for it. You didn't have to call your utility to buy your laptop, but your utility is now weirdly in the middle of your laptop doing cooler things like running the next notion agentic version. So those mismatches are not just substantial, it's also the mismatches are perpetuated by the fact that we seek tech returns as humans in a way that has never thought about utilities being in the middle.
- [55:44]
A
What did you learn at Tesla working with utilities?
- [55:49]
E
Nothing at all.
- [55:49]
A
Nothing at all. Nothing you can talk about.
- [55:51]
E
Yeah.
- [55:53]
A
Okay, well, let me re ask that from your career now. Surely you remember things that you experienced at Tesla and then you extract learnings from them in your current life. What's an example of an insight you have in your life now that you don't think you would have had had you not had the three years you can't remember at all at Tesla or the several years it was four. Four.
- [56:16]
E
Yeah. It's a long time. I'm alive. So let's see. I think the biggest takeaway is that I often found myself doing the work that could have been done adjacent to or inside a large electric company that wasn't, which could have been a utility, but it could have been another retailer. It could have been one of the regulators. It could have been like a third party ngo. It's like everyone needs help. And at some point, if you're at a tech company trying to do cool stuff in energy and AI and tech, you have to what's called in at least in the legal industry, second yourself or create yourself as a second as a person that has to do the stuff that other people should exist to do, but they just might not exist yet. And closing the gap of existence. Right. Sounds so existential. But putting yourself in that shoe, giving, wearing that other person's like shoe is what lets you walk.
- [57:12]
A
How should climate folks think about AI? I mean, you have a great view into the AI industry and the AI boom at Emerald. How should climate folks be thinking about it? What strengths should they be using or what opportunities should they be seizing that right now really aren't being taken advantage of?
- [57:27]
E
Yeah. So in like conventional energy world. Okay. Or energy and climate, it is very hard to align how much us as people really value a climate solution to actually shoving the institutional capital behind the solution. You and Me are kind of boring to an investment thesis because we don't actually have the money as people to put into our utility programs, to put into our regulated infrastructure to then actually make the climate outcome real. With data centers being the vehicle for AI, there is so much money just falling from the sky into AI and they have what is called a undetermined infinite value of loss load in the AI world. They will pay absolutely anything to not lose power. And of course the other part of it is to be able to do it sustainably. Because if you go to one out of every three data center conferences right now in this calendar year, the number one issue that executives talk about in this space is zoning and community resistance. If now the climate issue, the value thesis has moved to a there's a party that's willing to dump capital falling from the sky behind it. And not only that, one of the things they're most worried about is that that funnel slims and becomes like a martini glass stick instead of being the nice big cup they want it to be because there's a zoning issue in the middle that's tied to climate. Suddenly they should become your climate advocates. Like they have to be. Which is nice because then you and I, we're looking for the same social outcome. But it's not our bank accounts and our, you know, blood and toilet sweat that has to fund it. So I, I really need to see that in my next matchmaking session. That I do like after I get Emerald to like work at scale is we have to put move capital to where it needs to go because the climate vision that works for everybody is the socialized good. It still needs to be funded by private interests.
- [59:26]
A
I feel like this is one of the things we've learned from like the China example too is it's actually demand growth. It's like growth, growth, growth that drives a lot of the innovation and you kind of need to hitch to that to do some of the cooler things. You were describing utilities as a slow patient industry financed by very slow patient investors. Do they need to change to handle growth from AI or from electrification in order to meet this moment? Or is it kind of futile to expect them to change? They're always going to be financed by these patient investors and we need other companies to come in and do the work around them.
- [59:59]
E
There has to be more special purpose vehicles of various kinds like Constellation Tech Ventures or MJV Natural Grid Ventures here in New York that bring private capital from a different angle and attack it after the utility sits down and realizes that the Things that are the ambitions can't be done with utility capital. So that's one that's huge. Yeah, you already know about Blackstone and other firms seeking to actually buy out small utilities. So that's two. The third is coming up with new ways to ring fence where utility capital goes by grouping customers differently. And so like that's happening at a co op in Virginia. Some co ops in Texas already do that pretty well.
- [60:38]
A
And what like categories would they use to.
- [60:40]
E
Yeah, so the room has different sets of chairs in here. So sectioning it off so that the rate base, which is these chairs, which is a bunch of rich cats and AI and everybody else who is like, I just want power, they're treated differently and then the infrastructure for these things is funded differently. So you don't have these mismatched cost and benefit problems that we're seeing at least in the news.
- [61:01]
A
Is that something a utility can do by itself or does that need regulators coming to them and being like, hey, you got to start ring fencing your rate payers.
- [61:08]
E
The regulators can achieve it with performance based rate making and other mandates that usually show up in like large load tariffs. And that can't be the only thing though. And I say that because a regulator just says, you guys on this side of the room here, you're stuck with a higher rate for power. All that does, it just tells you, oh, we need to go find more power. All this stuff that we're talking about that is so climate adjacent, like the flexibility solutions, batteries, solar, behind the meter and all that, you need more, you need co created investment outcomes. So this side can't just be regulators yelling at folks. It also needs to be utilities reaching out to hyperscalers, developers, all of the builders and saying, bring your money this way, yes, you'll make your return, but also we can give you what you want if you help us co finance.
- [61:53]
A
Is it wrong to think about this moment? I mean Tesla already is the classic version of a tech company that's become an energy company. Is it wrong to think that in this moment all the tech companies, or at least all the tech companies with AI exposure, are becoming energy companies almost by force?
- [62:09]
E
So the thesis from our friends at Core, Weave and Oracle yesterday was that the hyperscalers are likely to become wholesale energy transaction entities too? Meta already has market based rate authority, application or filing done at ferc. These MBR filings are public of course, so anytime you see any of those, you know. Yep, that's also also it's interesting they become energy companies in the sense that they need to be able to de risk risk the future even though everyone is rushing to build for the present.
- [62:40]
A
Last question. You've described yourself as an energy matchmaker. Are there any recent exciting matches you've made at Emerald?
- [62:47]
E
AI Yeah. So we all know about PJM and the high prices.
- [62:52]
A
Yes.
- [62:52]
E
Yeah, everyone knows.
- [62:54]
A
Everyone knows so much. Grid neighbor here in New York on the verge of war.
- [62:58]
E
So Emerald did two demos that actually followed and helped the data center curtail load, which reduces system level prices. In Arizona. Those results are out and the next demo is actually happening. As if what is today? I'm lost.
- [63:12]
A
It's climate September 24th Friday.
- [63:15]
E
This Friday they begin their demo at a heretofore undisclosed live data center site in the PJM Service territory in Illinois.
- [63:23]
A
Well, that's really exciting. Arushi Jeremy Frank, thank you so much for joining us on Shipkeep. And I want to thank thank as well our live studio audience for this version of Shift Key. Thank you so much for sticking it out with us. You can find us on any podcast app ShiftKey. If you enjoyed this episode, please like and subscribe. I want to thank Nico Loricella, Gillian Goodman and I want to thank Salesforce, our sponsor for the night. Multimedia editing and audio engineering is by Jacob Lambert and by Nick Woodbury. Our music is by Adam Komalow. Thank you so much for listening and see you next week. Week.