← Energy Gang
Energy Gang
AI could break the electricity grid. What do regulators and the industry need to do to keep the lights on?
00:59:47
Loading summary
Transcript76 lines
- [00:02]
A
Hello and welcome to the Energy Gang, a discussion show from Wood mackenzie about the fast changing world of energy. I'm Ed Crookes. I'm welcoming you now to a special edition of the Energy Gang, recorded live in front of an invited audience at the Council on Foreign Relations in New York. Hello, everyone in the audience. Today we're going to be talking about the reliability and resilience of the US Energy system and about the grid in particular. Very pressing subject at the moment, in particular, I think because of everything we've been talking about in terms of rapid growth in electricity demand in the United States, mostly because of AI and also because of new tensions, strains on the grid that had been created by changes in the generation mix, retirement of a lot of old generation, addition of a lot of new generation, which in recent years has been very largely wind and solar being added to the grid. And that's creating a lot of new issues. And that's really what we're going to be talking about today. To do that, it's a great pleasure to welcome my old friend from the Energy Gang, Amy Myers Jaffe, who's director of the Energy Climate justice and Sustainability Lab at nyu. Hi, Amy, how are you?
- [01:11]
B
I'm great, Ed. Great to be like launching off New York City Climate Week here at the Council of Foreign Relations.
- [01:18]
A
It's also a great pleasure to welcome Neil Chatterjee. Neil is one of the big names, I think it's fair to say, in US Energy policy had a number of different roles. You were previously chairman of ferc, Federal Energy Regulatory Commission, and now you're wearing a number of different hats, aren't you?
- [01:33]
C
Number of different hats. Most significantly, Energy Gang podcast participant. I'm also the Chief government Affairs Officer at Palmetto, a clean tech company, industry advisor at kkr, doing due diligence in the energy space. I'm on a number of boards and I may be breaking news here. I'm soon to be a distinguished visiting fellow at Columbia. So the singular throughput through all of my professional activities is that I'm a conservative who cares about decarbonization, with an emphasis on maintaining reliability and affordability, which I think will undergird our conversation today. I'm very much looking forward to it.
- [02:10]
A
Yeah. Well, thanks very much for joining us today. Great to have you here. And it's also a pleasure to welcome to the Energy Gang for the first time. Cecilia Velasco, who is a managing director for infrastructure at kkl, are of course one of the world's biggest investment firms. Cecilia, thanks so much for joining us.
- [02:24]
D
Well, thank you Ed for having me. It's a pleasure and look forward to comparing notes here with this group around the Ed topic at hand.
- [02:31]
A
Got power? At Hythium, we make sure the answer is always yes. Recognized seven times as a BNF Tier 1 best provider and ranked top two globally for battery shipments for 2025. Hythium delivers safe, reliable and profitable energy solutions that keep the clean energy transition powering forward. Let green energy benefit all trusted worldwide. Built to last. Discover more at www.Hythium.com and let's talk energy that works for good. Neil, I wonder if I could start with you perhaps to kind of frame things for us and to talk a little bit about your view of this issue. When you think about reliability of the grid, resilience of the grid to shocks and disturbances, what are the key issues as you see it?
- [03:16]
C
Yeah, so look, this was, I felt, my singular focus during my tenure as a commissioner and chairman of the Federal Energy Regulatory Commission. I think while the Commission has vast responsibilities, in my view our foremost obligation, the one that my colleagues and I had, was to ensure that when Americans hit the switch that the lights come on. But this is an exercise that quite frankly many of us take for granted. And it is one that used to be oversought primarily by engineers and economists. And over the course of time, particularly over maybe the last two decades when we in the US have been the beneficiaries of relatively flat demand for electricity, what we've seen is increasingly politics has started to infect this discussion around really complex, nuanced, wonky, technical things like resource adequacy. And we really struggled with it during my time at the commission and I for one bear some responsibility for it. When I first was seated at the Commission, I had just spent nearly a decade as energy policy advisor to my home state Senator, Senator Mitch McConnell of Kentucky, where I worked very closely on, on behalf of his constituents who largely represented coal interests in Kentucky. And very early on in my tenure we were presented with a notice of proposed rulemaking from then Secretary of Energy Rick Perry that sought to compensate certain forms of baseload power, namely coal plants and nuclear plants who had the attribute of having on site fuel, 90 day supply of fuel. And I'll be honest, I handled this very poorly. I think Secretary Perry raised a serious question about reliability and the resilience of the grid. I handled it the way that a politician would, not the way an independent regulator should. And in doing so, I think I started a period to where politics really unfortunately impacted in A negative way. This broader question around resource adequacy and reliability and resilience. Even the very term resilience. Some people call it resilience, some people call it resiliency. And your politics can be borne out in how you even pronounce the word. I wrestled with it during my time as chair of FERC in the first Trump administration, where we looked at things like the minimum offer price rule in the PJM capacity market, which we could get to. I thought my successors in the Biden administration went the other way and took steps to, in my view, artificially suppress capacity prices to accelerate the retirement of their sources of negative generation and to proffer up their preferred sources of generation. And I published an op ed in Fox last week where I expressed frustration with both administrations. I thought the first Trump administration was wrong to try and artificially impact markets to drive their preferred sources. I thought Biden made the same mistake and now I think Trump 2.0 is making the same mistake again, but just from different fuel sources. The reality is, and we're going to talk about it, and I don't mean to monopolize the time, but I think not just demand being driven by AI and data centers necessarily support AI, but everything from streaming services to cloud computing and quantum computing, vehicle electrification, building electrification, crypto and bitcoin mining, a reshoring of manufacturing in this country, all of these things are energy intense endeavors that are going to have to cause us to rethink our approach to resource adequacy and meeting this supply demand imbalance. And to me, I want to take the politics out of it. I genuinely believe we need every available electron. And so for the political left, I think that means coming to terms with the reality that in order to meet this coming surge in demand, we while maintaining reliability and affordability, we're gonna need fossil fuels, particularly natural gas, for the foreseeable future, which is going to mean more natural gas plants coming online and more natural gas pipelines being built. But it also means for the political right that we have to become aware of the fact that we cannot possibly win the AI race, meet this coming surge in demand and maintain affordability and reliability with fossil fuels alone. The that not only are we going to need solar and storage and geothermal and nuclear and wind, but we're also going to need negawatts, things like demand response and virtual power plants, grid enhancing technologies that can squeeze maximum efficiency out of our existing infrastructure, increased deployment of distributed energy resources. These are all going to be critical and, and so Amy and I were on a podcast earlier today where I joked she wrote a book. I'm not smart enough to write a book, but if I could get a tattoo somewhere on my body, if I didn't have such limited pain threshold tolerance, my tattoo would say solar plus storage. With gas peakers to balance the grid and investing in alternative technologies, it's not that hard. We just got to get the politics out of it. Thank you for listening to my TED talk.
- [09:21]
A
Very good. It's still quite a lot to get tattooed on you, I think, but still, even that much, I'm very interested. I want to come back to questions about the current administration and what they're doing just before we do, though, I just want to be clear about when you say you think you handled the first Trump administration wrongly or some of the energy policy moves under the first Trump administration, what did you do wrong? And what, in hindsight, do you think you could or should have done differently?
- [09:50]
C
It's simple. Should have kept my mouth shut. The problems that I created within the markets and within stakeholders who observe the commission and the markets that oversees is I was opining the way a politician would opine about what the commission may or may not do in regards to the specific policy. And it was having a direct market impact. People were making decisions based on the thought that, oh my God, this maniac may actually adopt this policy. And ultimately I voted with my colleagues unanimously to find that what was being proposed was not legally viable. But because I talked about kind of screwed things up, I learned my lesson. And a couple years later, there was a petition regarding net metering, about nationalizing kind of net metering and policy in the US that would have been very hostile to solar energy. And a lot of people assumed because of the political demeanor I had sort of adopted at the commission that I might have been sympathetic to this very hostile clean energy provision. And when we dismissed it 40 and rejected it, people were surprised. And it has a positive market impact. But the lesson I learned from my first exercise to that one, I kept my mouth shut the second time.
- [11:10]
A
Well, it's nice, obviously, to have left for now and to be able to talk as much as you want.
- [11:14]
C
Yeah, much easier now. So.
- [11:15]
A
Yeah, indeed.
- [11:16]
C
Much better for podcasts.
- [11:18]
A
Amy, so what's your take then on this issue of reliability, resilience on the grid? Where do you stand on what the issues are, what the key trends are, and what the administration is doing about them?
- [11:31]
B
I'm going to second Neil and up him in the sense that, you know, I'm a university professor People do, you know, power engineering modeling. We just have a new FERC order that all transmission planning processes need to be 20 years and need to be very forward looking, need to do scenarios. One of the scenarios needs to look at extreme weather, which is vitally important at this point. I actually have a group of students, graduate students at NYU who've been fooling around with the different data we've been improving. The data set on virtual power plants and where they're going in in the United just downloaded and started manipulating all the outage data for the entire United States for the past decade. And we've only done the annualized data first. So obviously the monthly data is going to be very different and state by state's going to be very different. But when we compared PGM with Caiso California. PGM by the way, is sort of like the Mid Atlantic and the center of the country. So we're talking like the Ohio's, the Pennsylvanias, et cetera. And we also compared it to ercot, which is Texas. And the outage data for Texas is pretty flat. And then there's this giant spike during Hurricane uri and the California data is surprisingly flat, but with a slight uphill climb. But the PJM data outage data is three to four to five times higher over the entire period and increasing, which shocked me because we hear a lot of talk about California's failing because of the fires and the bankruptcy of pge, the main utility for Northern California. We all have seen the whole giant as Neil's alluding to publicized political debate about Texas. But our real problem in the country is in pjm and that is where the most data centers are. And let me tell you another little factoid not to be partisan. It is where the highest concentration of natural gas peaking plants is. So when people tell you that there's a technology solution and it's for the whole country, that's clearly not right. Because here you have a place in the United States that has more natural gas peakers than anywhere else and they have the highest outage rate compared to places that, that have successfully deployed batteries. Texas is the largest market now in the United States for battery storage deployment. Nrg, the big utility down there, has literally just announced the 1 gigawatt virtual power plant. So you know, I agree there are going to be places where we have to use different fuels because of the nature of what's there and the nature of what we need to go and how much it's growing and so forth. But you know me, Ed, I'm like a broken record on the battery thing. And that's a challenge because we have to keep our investment in R and D and manufacturing for batteries in this country and we can't afford to have that be a political question.
- [14:47]
A
Celio, let's talk about the investor perspective. Then Neil talks about needing every electronic. If you're going to get significant increases in generation capacity onto the US grid, that means deploying a lot of capital to make that happen. I was just out a couple of days ago on the west coast talking to a lot of people in various types of low carbon energy, in particular wind and solar and storage. And something you heard from them quite a bit. Well, they talked about sort of two kind of key trends. The kind of the good news for them is growing electricity demand and data centers and some of the other things that Neil identified driving that growth in demand. So it's fantastic for them. But then they also say there are various obstacles and difficulties being created by Congress and administration. Crucially, they would identify one, the big beautiful bill and the changes to tax credits in particular, ending the tax credits for wind and solar, clearly creating problems for them. And also some of the new regulatory stuff we've seen coming out of the Department of the Interior, some of the new barriers and challenges being put in the way of wind development in particular. And so it's a kind of mixed picture for them. And certainly it's not kind of full steam ahead, all systems go, it's not disappearing. But you know, investment in renewables and storage definitely is facing some challenges. How do you see it?
- [16:11]
D
Yeah, absolutely. It's a good question. Firstly, I agree a lot of people always say we're, you know, talking about energy transition and in my mind we're talking more about energy expansion. Like if you look at us, you know, we had a period of over a decade where power demand was flat. That is not the case and has not been the case for the last couple years. And if you look at what's expected here from here to like 2030, you know, there is like maybe a two and a half ish, give or take. You know, different research outfits will put different numbers of power demand growth through the end of the decade and two and a half might not sound like much, but the US has such a big power denominator that when you look at the incremental amount of electrons that that represents from here through the end of 2030, that is more electrons than the like most countries in Europe use today. And so you just think about what does it take for the US to plug in like the amount of power that like most single country, more than what most single countries in Europe use today, that, that is a lot of investment. And as you said, that comes in the form of generation, that comes in the form of electrifying and fortifying the grid on batteries and sort of the whole landscape. And when you overlay a little bit like that need and you know, us wanting to make it happen, you know, and if you want to win the AI race, if you want to increase data centers and connectivity and you know, enable all the great things that are to come around, robots, self driving cars, you're going to need that electricity. And when you look at policy and around it, I mean as an investor, one of the hardest things is like the policy uncertainty. You know, these infrastructure projects, you know, you don't make decisions on a whim and turn it on and off within, you know, days, months, weeks, like they're multi year planning processes to get an asset in place. And so when policy kind of changes quickly and abruptly, you know, in short periods of time, it makes it really difficult to plan and deliver on those projects and it kind of grinds things to a halt and significantly delay things. So first and foremost I would say a little more stability and visibility on a longer term policy would be critical to really enable the sector and the infrastructure space to really kind of catch up and facilitate that energy expansion trend.
- [18:26]
A
And so do you think if we had that stability, we could get a lot more investment than we're seeing now?
- [18:31]
D
Yeah, absolutely. You know, I think just when, when the rules are changing, like it's very hard to plan and kind of work around those. So I would say first and foremost that stability would be the biggest enabler. You know, from there you could argue certain kinds of policies and you know, tax credits and whatnot that would facilitate things. Sure. But at the end of the day, look, I don't think there's a one solution fits all. You know, as Neil said, I think it's a little bit of, you know, if you're in the winter area, if you're in a sunny area, you know, if you're in an area with land, in an area with water, in an area with gas, like the solutions are different, like there's no silver bullet for it. And so you need a little bit of all of the above. And so, you know, the industry can accommodate to whatever the policies are and find the biggest solution, but we kind of need that visibility and that certainty.
- [19:17]
B
And do you think that, I mean, we're talking like private capital can really do this or do you think the federal government has to come in with capital? Is there a role where the federal government has to come in with capital? Or you think it's a private capital matter? Like if the different states and everybody has stable regimes, the money will appear and the practice will appear?
- [19:37]
D
Yeah, I mean, I think as long as there is that opportunity, like private capital will show up and you know, private capital can get it done. I mean, so I don't think there is a need or a dependency that government capital needs to make it happen. You know, if government capital wants to play a role and participate and accelerate like it can. But I don't think it needs to really.
- [19:57]
C
I mean, I think the challenge for private capital is to Cecilia's point is that stability, and I'm going to speak very parochial here to my experience, the Federal Energy Regulatory Commission. I think FERC has been a really important beacon of stability in the space over the better course of the last decade or so, with a few notable exemptions that I've already taken culpability for. If you look at the differences between FERC and other agencies in this space, look at US EPA for instance. EPA went one direction under the Obama administration, completely different direction under Trump 1.0, flip flopped again under Biden, and now has whipsawed again under Trump 2.0. For folks like KKR who are trying to deploy private capital, trying to pin those investments to the political pendulum swinging wildly back and forth is ma thing. The beauty of ferc independent agency, five commissioners, no more than three that can be from one political party, each serving a staggered five year term. So you could theoretically outlast the president that appointed you. The point of that is that when it comes to something as significant as electricity and reliability, we don't want politics in it and we want to create that investment certainty so that deployers of both government capital and private capital can have that certainty and stability. What I'm a little bit worried about, and again, I played a role in starting this and now you can't put the genie back in the bottle. FERC's getting more and more politicized and like the White House today has basically openly stated that they're going to take control of the commission. And my concern there is even in areas where I actually am supportive of policies that the administration is pursuing, I don't like the long term precedent it's setting. We're basically going to turn FERC into EPA and suddenly that beacon of stability will be lost and it'll be increasingly more and More difficult for the KKRS of the world to deploy capital with confidence because of that uncertainty in government.
- [22:13]
B
And Neil, let me ask you, do you think that Order 1920 and for the audience, you know, that's this order that tells the different big regional coordination agencies, like you know, like a PJM or you know, places where it's just a utility centered system that they have to do 20 year planning in 5 year increments. Have you seen that make a change now? Is that bringing a positive thing? Like to hear Cecilio's point of view on that as well. And would you urge the Trump administration to make sure that FERC stays the course on that?
- [22:48]
C
Yeah, I was very public. I would have, had I been on the commission, I would have voted for 1920. Honestly, part of the reason I didn't undertake some of the reforms that ultimately were comprised in 1920 is because I wanted to do other things like FERC Order 2222, which removed barriers to entry for aggregated distributed energy resources. And it would have taken so much bandwidth to do transmission planning reform that I couldn't have done some of the other things I wanted to do. But I would have voted for 1920 had I been on the commission. Would I have drafted it exactly the way it ultimately came out? Probably not, but I think I could have gotten there. And that's back to my point about like having that necessary stability. What I worry about now, and to answer your question, I actually don't think 1920 is going to provide the certainty that we need to get the necessary investment to build this transmission. That is absolutely essential if we're gonna win the AI race. We cannot possibly win it without building out this transmission. But these are really complicated questions and they're hard. I'll just be honest with you. They're hard when it comes to something like siting. I have had my state regulatory counterparts when I was at FERC tell me, well, Neil, we would love it if FERC would just overrule us and run over us and approve this transmission line because otherwise we can't get it appro in our community. But we don't want to say that out loud. We'll be eviscerated. And I'm sitting here like I like to go on vacation, like what if I want to visit your community? Why are you going to put it on me to approve the transmission line? Cost allocation. Cost allocation is really, really tricky when you think about who is going to pay for this transmission line. Think you got state A that has a tremendous amount of renewable capacity. You have State C that has huge demand for this energy. What happens to State B in the middle and to ratepayers in State B? FERC, and I agree with them, made a determination that ratepayers in State B get a genuinely significant benefit in the form of alleviated transmission congestion, which is one of the biggest issues that we have on our grid today. And that they ought to pay some portion of this long haul transmission line. As an academic exercise, that may seem like a totally defensible thing from a real world standpoint, going to ratepayers in State B and saying you're going to benefit neither from the generation nor the consumption of the power. We're going to build a transmission line through your backyard that you don't want and we're going to make you pay for part of it. That's really, really hard. And so FERC did the heavy lifting. Chairman Willie Phillips, building upon my former colleague Rich Glick, they did the heavy lifting. They got Republican Chairman Mark Christie's vote for it. This should be stable policy that will enable investment so that we can build out the grid of the future. But what I worry about now is you're talking about policies out there sunsetting all regulations, going back and kind of revisiting policy. Who's going to deploy capital based on 1920 if they're not confident that 1920 will be kept in place. This is the type of uncertainty that I'm talking about that is a problem for both Republicans and Democrats. I'm not trying to be partisan there.
- [26:18]
A
Okay. So I want to get into, though, a substantive part of the current administration's critique of a lot of what's going on in terms of energy policymaking, where essentially they're talking about wind and solar as having no value to the grid or maybe even negative value because they undercut the economics of dispatchable generation. And you read, I'm sure, the paper that came out of the Department of Energy back in July where they talked about trends in the electricity system at the moment that are pointing towards very steep increase in outages, in particular because of 100 gigawatts or so thereabouts of dispatchable generation, mostly coal being retired, about 200 gigawatts of generation being added to the grid, of which only 20 gigawatts is natural gas, great majority, 120 gigawatts or so is solar and so on. The argument then is being made that actually this is going to make the grid weaker, as I say, will lead to an increase in blackout and actually if you stop that solar and wind being built and that actually is going to make the grid more resilient. Do you think there's anything in that?
- [27:29]
C
Yeah, I want to start by one thing, and this is where I think we do and why I like coming on programs like this podcast and speaking openly and candidly is I think we've got to get past, you know, some of the top line narratives that people like to push out. So when that study was initially presented, folks dismissed it as this was done by a bunch of political hacks that don't know what they're doing. And this is they're trying to force a outcome that the administration prefers. And I'll tell you, I know the people that worked on this report, these are very serious people and the questions they raise are absolutely legitimate. And so I just want to start there as the baseline that we should take that report seriously and not so quickly dismiss it in the manner in which some people did. I do have, I think, some sense of where components of where I agree with what they're trying to do. I understand, because I faced this myself during my tenure at the commission, that there are probably some plants that are under economic pressure to retire that are probably needed for resource adequacy to meet the coming surge in demand that we're about to face, that we probably need to keep on line longer. Market proponents don't love that. But if the administration is going to use tactics like 202C to keep some of these plants on longer, I will note that some of the plants that were scheduled to be retired that they utilized 202C to keep open were ultimately dispatched in the middle of the heat wave we had this summer. Which kind of validates the point that the report and DOE was trying to make.
- [29:06]
A
And sorry, a footnote here. 202C is.
- [29:09]
C
It is a. It is an emergency authority that through the Federal Power act, that can be utilized to keep a plant operational that was otherwise scheduled to be retired. Thank you for clarifying that. But I also think, again, in my view, coming back to the core point, that we need every available electron. There are just realistic constraints that we've got to consider. Gas is basically sold out through 2032 because of supply chain constraints. We probably can't get new gas turbines built before maybe 2029 at the earliest. I'm a big believer in nuclear energy. I want so bad for advanced nuclear to be successful. And I've worked very closely with KKR and looking at making potential investments in this space. But I remain concerned about the reality that some of these plants will be built on the timeframe that we need. I keep coming back to my tattoo. It's solar plus storage with gas peakers. That is the quickest thing that we can get onto the grid in the coming years to meet this coming surge in demand. And I do think the DOE report did touch on that to some degree and gives, you know, sort of like some, some, some base credentials for it. And so I got to defend DOE there. They didn't, you know, they didn't blush over some of these issues. Now I do have my criticisms. I thought Secretary Wright made a comment the other day that I thought was absolutely absurd, which was basically if you wrapped the world in a solar panel, you would only get like 20% of generation and that's just not even like remotely close to being active. I wish he hadn't done that because he's a smart guy. James Danley, the Deputy Secretary of Energy, was my former General counsel, my former commissioner, ferc. These are smart people, they know better than that. I wish he hadn't said that because again, it now colors this DOE report that otherwise was pretty reasonable.
- [31:05]
B
The people who looked like they were geniuses were the people who stepped in and restarting three my island or you know, talking to Holtec about Palisades because, you know, there is sort of a timeline for SMRs. KKR has really been at the forefront of being at least in the conversation for some of these big infrastructure projects. Too risky. Risky. What have you put your money down in? What do you love and which would come to market?
- [31:33]
D
Yeah, yeah, definitely a good question, Amy. So look, reality, as I said, like we look at sort of everything across the board and there's no silver bullet, right? Like as an example, you think about solar plus storage, that is an area that we are extremely bullish. We think it makes a lot of sense in a lot of areas, but not everywhere, right? And just to take an example, like in your desert southwest of the United States, right, there's a lot of land, it's a desert, it's very sunny, you don't have snow. So solar and storage is a pretty easy, quick to deploy, applicable solution. Like pretty competitive on a time and cost basis. At the same time, you don't have a lot of gas, you don't have a lot of water. And so it's not straightforward to build gas plants, you know, in that area. But you know, if you look at other areas, like, if you look at like, you know, wintry and very snowy, like northeast, you know, Is solar going to be the most competitive solution there? Probably not.
- [32:30]
B
Well, you know, Vermont's doubling down though. They're doubling down. What do you think?
- [32:34]
D
I mean, I think it depends. Right.
- [32:36]
B
Like a lot of people say, they say, I talked to people in Vermont in the electricity industry. They say that they've eliminated the duck curve with the battery program, but that every now and then they get a price spike based on having to shovel the snow off of the panels.
- [32:53]
D
It's hard to shovel snow off panels. These things are massive. But no, look, jokes aside, I think when I think about it, there's energy security, energy affordability and some reasonable environmental responsibility. And I think as you look at the energy transition and the energy expansion, you're juggling some version of those characteristics and so on the energy security. There is something to be said about do you put all your eggs in one basket and you have one solution and if there's an issue there, that's all you had, or do you proactively kind of diversify to have some solar, have some batteries, have some gas and like have some mix of different solutions that when you run different shocks through the system, you know, they kind of diversify and kind of counterbalance each other, you know, affordability, again, like price is a big one, right. And so you go for solutions that kind of help plug into the system complementary and bring things on a cost efficient basis, you know, and environmental responsibility, like that is a thing. People live in those communities, you know, they breathe the air, they use the water, they, you know, sit next to whatever that plant is. And so they care about what's there and how it affects their community and the day to day life. And so you do have to balance all three. And so I think different states, different areas, you know, do that in slightly different ways.
- [34:10]
B
So you've got Bill Gates, he's cutting the ribbon on geothermal the other day, right? The Pentagon saying they're going to try geothermal and bases. Can private money ever go into nuclear? First of a kind seems like off the table. Or am I wrong?
- [34:23]
D
Yeah, look, it depends a little bit on, on the nature of the private money. Right. Like I'd say like, you know, for us, we're financial investors, you know, we're not, you know, builders of nuclear power plants or operators of nuclear plants. You know, you wouldn't want me doing that. I wouldn't want to be the one doing that. You know, but there's a lot of companies and power companies, IPBs that have been doing and running nuclear fleets for a very Long time, you know, and they have the expertise, you know, they have the track record and they know what they're doing. So ultimately, you know, when we look at nuclear opportunities, you know, we seek to be how can we be partners where we bring, you know, our capital or resources, our know how in the areas that are complimentary but you know, we will lean and on and partner with, you know, folks that have that, that industry, that operational expertise and can do that, you know, in a responsible and appropriate way. And so a lot of the nuclear restarts, you know, I think are a great thing. Right. Like it is a way to bring back online capacity, you know, into the country. I share Neil's view and I mean he said we've, we've worked on some of the nuclear opportunities together. Like, you know, when you kind of fast forward, you know, nuclear has some ways to go to kind of grow and develop. But I do see how it could be an important part of the energy mix. When we're looking, you know, medium and.
- [35:38]
A
Longer term and so new build, then that is something which potentially you could be interested in investing in.
- [35:43]
D
Yeah, and look to us specifically like we're a little more like infrastructure capital, so we're less so on the kind of early side venture type risk. So for us specifically it would be a little more on, you know, operating kind of restarting or whatnot. But you know, there's, there's plenty of other private capital pools out there, you know, that do some of those more earlier stage kind of nuclear facilities. And you know, we've seen that there's a couple, you know, public companies on the nuclear side that you know, are developing new technologies or starting first of a kind projects and have raised billions of dollars to go and do so. And similarly there's you know, a number of private companies that have been backed by, you know, wealthy investors as well as many private capital pools and a very significant money to do that. And I think like, you know, the private sector is doing like wouldn't be doing that if they didn't thought that there was a role and an outlook where you could see some hours being a real thing deployed commercially, you know, in big scale.
- [36:42]
B
And do you think the electricity trend is going to last long enough for that? I mean I saw a report this week that some of these private wealthy capital is actually going into fusion now. And I thought, oh my God, that's like super optimistic. Like, you know, do we think that the electricity trend, Neil Facilio, is like, it's not like for the next three years it's like forever.
- [37:04]
D
Yeah. Look, my personal perspective is, you know, electricity is kind of like the ultimate commodity, right? Like, an electron is completely undistinguishable from one or another. But, you know, it is a very, like, important commodity. It, like, fuels a lot of what we do, right? Like lights, elevators, transport, like machines, devices, phones, cell phones, like, everything. And when you kind of see the trend, we only use more and more of it, right? And as you see data centers growing, as you see AI taking a bigger part in life, as you start looking at some of those autonomous vehicles that are already kind of driving around, some pilots started here in the city recently. Like, my view is like, there's only going to be more and more. And of course there's, you know, some offset on efficiency. And we've seen that over time, you know, and light bulbs, insulation, like, you know, things getting more power efficient. But, you know, at the same time, in my view, it's going to keep growing. It's going to keep growing for a while, and it's not going away anytime soon.
- [37:59]
C
I mean, I think we do need to have a better handle on the nature of demand and the coming surge in demand. I thought it was notable earlier this year when the Chinese deep SEQ announcement came out, which in my view, I viewed with a great deal of skepticism. I thought it was propaganda designed to chill investment in AI deployment in the US but it had a significant impact. If you look at the publicly traded independent power producers, all of their stocks tanked in the immediate aftermath of that simple press release from the Chinese Communist Party because their projected growth was predicated on this idea that there was this coming huge surge in demand coming. And we really need to grasp that. I think part of it is those of us on the energy side, we need to learn a lot more about AI, and vice versa. I am told, credibly, that, you know, the rule of thumb today is that a simple ChatGPT search takes about 10 times as much power as a simple Google search. But that's with today's AI, that with advanced AI, it'll be a million times as much power. Well, what does that mean? And how do we deal with that? And can these AI data centers be more flexible? Can they find greater efficiencies? To Cecilia's point, can we utilize AI to maybe navigate the interconnection queue? Right. Like right now, One of the huge things I wrestled with at FERC was our inability to do studies in a feasible amount of time that led to this huge buildup and backlog in the interconnection queue. Can AI help us better navigate the interconnection queue? I work with a company right now that uses satellite imagery and AI to do vegetation management to better enable the mitigation of wildfire risk. That's another positive implication that we can utilize AI. But there are going to be broader questions societally. I'm telling you it's people are focused on workforce and what AI might mean for a changing workforce and they're focused on on climate. And will AI be a positive lever for our ability to mitigate carbon emissions or is it going to drive demand through the roof to such a degree that it's going to lead to an increase in emissions? These are huge societal questions that we need to answer. I actually think it is very unfortunate that most of these monumental decisions are falling on energy market regulators. My poor former colleagues at FERC having to deal with data center co location. This is such a monumentally significant issue that should not fall on these poor people and the staffs at the commission. We need global leadership to answer some of these questions.
- [40:38]
B
So let me just say cards on the table piece coming out in the Wall Street Journal looking at all across the academic world's work on whether the AI could reduce so much energy use for things like route optimization, across the transportation sector, for green identifying and getting in place green materials for buildings, having H VAC systems work effectively in buildings that we might actually see more energy savings from the AI than the AI itself actually uses. So really interesting when you look at the actual research that people have already done, it's not as horrifying as kind of what we're like hearing in the popular domain about the evils and how AI is just going to crash us into climate change catastrophe.
- [41:32]
D
I totally agree. I mean when I look at the sources of power demand, I think some of the hardest ones to predict is really data centers and AI. And within data centers you have the cloud use case, which is the biggest use case and that is more predictable. Like it's had a growth trend and you know, people are using the cloud more and more. But AI, you know, I can't help but to think about like those videos that Nvidia puts out and every year they're like, here's our new chip and it is like a gazillion times more efficient than the one like five years ago and five years ago. And so where does that kind of equation of people using more AI but then like compute getting much more efficient on the use of power, like where does that even out is a very.
- [42:13]
B
Come to NYU on Thursday because Nvidia will be here talking. It's precisely, precisely above that.
- [42:18]
A
Exactly. Listen out for this. On a future episode of the Energy Gang, we have the head of energy at Nvidia as one of our guests. So yeah, it's going to be very.
- [42:25]
D
Interesting to talk for his answer.
- [42:27]
A
Absolutely. Looking for greater resilience in energy systems without the risks of going it alone. At Hythium, we've got you covered. As an integrated energy storage solution provider, we offer customized storage systems as a service tailored to each project, whether utility scale, commercial or renewable integration. From design and financing to operations and maintenance, Hytheum manages the entire life cycle delivering sustainable profit, stronger resilience and lower emissions. Built to last, built for progress. Discover more at www.Hythium.com and let's empower the green future together. Now I should open it to the audience for questions, I think. Do we have a mic on the floor?
- [43:10]
D
Yes.
- [43:11]
A
If anyone, he's got a question they'd like to ask. You want to raise your hand? Yeah, I see one over there.
- [43:15]
D
Philip Ellison, you know, Cecilio and all.
- [43:19]
A
Of you really have raised the issues.
- [43:20]
C
About long term vision and how to.
- [43:24]
A
You know, get something like 1920 graven into stone. Well, of course a year ago we.
- [43:29]
D
Had the Loper Bright decision, which ironically came out of the response to Chevron, an oil producer.
- [43:36]
C
But so the question really is, isn't this an opportunity now for Congress to.
- [43:42]
D
Write the balance in terms of administrative.
- [43:45]
A
State and executive branch agencies and public policy written into the law that things like 1920 become a part of the.
- [43:56]
C
Law and that we are required to have a long term vision, if you will, a Magna Carta for energy.
- [44:04]
A
Yeah, very interesting thought. I think this is one for you, Neil.
- [44:06]
C
Yeah, I'm going to start with this one and I have a fairly optimistic view on this, which I know is rare in this moment in time where people have really lost confidence in Congress's ability to get serious things done. The reason I like it, and you raise a significant point about Loper Bright and the erosion of Chevron deference and really kind of removing some of the deference on executive branch agency authority. I used to joke around with my colleagues that I loathed Chevron deference when I was chairman of ferc, except when it came to my decisions then I wanted it. But I actually think, and I hope I'm not being pollyannishly naive here, I think that AI, and in particular we've been talking a lot about the energy component of AI. We haven't focused as much on the national security component of AI. Amy actually addressed this beautifully on the Political Climate podcast, which will also air next week, which I hope you will listen to, where she talked about AI and drones and the warfare implications of AI. I actually am totally bought in on the need to win the AI race against the Chinese Communist Party for national security purposes, with an understanding that in order to do so we're going to need a tremendous amount of energy and to win it. And I'm hoping that that will be the trigger that enables Congress to bust out of the sort of partisan corners that we've been locked into when it comes to energy. I would have thought and hoped that a reliability event, having been a former reliability regulator, would have been the thing that would have shaken us out of our partisan corners. It didn't happen. I would have hoped, as somebody who oversaw the competitive wholesale power markets, that huge surges in prices would lead us to break out of our political corners. Didn't happen either. You know what? I sadly think it's going to take the minute one of these big AI data centers that was projected to be built in PJM or ERCOT or MISO goes to Alberta where they have surplus capacity or overseas South Korea, Japan, China. That is, I think, what is going to awaken policymakers in America that we gotta bust out of this partisan logjam that we've sort of been locked into for the last kind of couple of decades. We haven't had a major energy bill done really since the Energy Policy act of 2005. You look at everything subsequent to then. Most energy policy has been done vis a vis the tax code or budget reconciliation, which is not stable. We need real bipartisan set in stone Magna Carta energy policy. And I'm hoping that we get there before it's like reactionary. I would like to be proactive and get there. I am optimistic that you are starting to see some bipartisan conversations amongst Republicans and Democrats who really don't necessarily see eye to eye on policy, talking about the need for permitting reform. That's a great place to start. I think permitting reform could be the fantastic basis for that larger energy policy. So I'm going to go ahead and choose to view the optimistic camp that we can get there because we need to get there.
- [47:38]
A
Thanks. Another question? Yeah, one down here in the front.
- [47:40]
C
Thank you. Suhelster. I'm a research energy at the McKinsey Global Institute. We speak a lot about the AI need for energy and it is a very interesting question. But AI's energy use now is quite a small energy bucket compared to these other end uses. We talk about electrifying road transport, all these other things that it's not 10% of the energy system of the electricity, it's going to be 2x or 3x. If we're having so much trouble now getting the electricity we need for AI, what does that mean for these other goals? How realistic can we actually be in terms of this transition? And what does it mean that China is doing this so much faster than us in the us?
- [48:21]
B
So let me take that on for a second. So I think, and Cecilia has already sort of made this point that a lot of things are locational specific. But I think one of the challenges of AI and inference and, and some of the other applications within the AI calculation algorithm process is that it can cause a sudden surge or sudden change in electricity demand in a sort of, I don't want to say moment to moment in a less predictable manner. And that's a challenge not only for our grid, but for China's grid. I mean, coal cannot go flexing up and down. And so thinking about how to take this old system that was built was 100 years ago or more.
- [49:11]
D
Buy more.
- [49:11]
B
Yeah. And you're going to try to get it to address this new kind of usage, especially in commodity prices where you're having a capacity, market prices are, we all have this expression in economics, prices are set at the margin. So that little 1% problem, we saw that take down all the electricity of Spain. Right. I mean, that little problem can be a really, really big problem. And to Neil's point, and this is really fundamental because as a person who's been struggling to really understand all of this, you know, who in Congress really understands about maintaining frequency and voltage? Does anybody in Congress know about Sean Cassen? Okay, does anybody know about spinning? Like, you know, it's. The education curve is also very steep. And so we really do need to actually take politics out of it. And really, you know, it's like, I mean, there are many areas where we really should let technical experts weigh in. I mean, that could take the whole rest of the day to talk about that. But in electricity, where making a mistake means none of us have service, it really is important to understand how the grid is going to have to respond to new kinds of conditions and what are the ways of achieving that. We haven't talked about, here's a fancy word, reconductoring. Right. Do we take the copper out and put some more high grade material that's less apt to heat. I don't know if that's an investable space for outside capital or whether that has to be utility Getting utilities to see, Neil, that that's a safe technology, even though it's been around for a long time. These things are challenges that an electricity specific bill on the Hill could be really helpful.
- [50:59]
D
Yeah, I totally agree with that and I think you raised a great point, which is like, what is the actual share of electricity that AI is taking? And it's small, you know, like when I think about kind of the power demand, data centers are less than half and of that the majority of it is cloud and you have more visibility. AI is a small percentage of that and it's going to get more and more important. But it doesn't make or break the story. You know, it is like addition and incremental to the already backbone of growth that you see when you kind of take a step back outside data centers and AI. The majority of power demand is just coming from good old commercial industrial demand, additional residential development, the Internet of things, people having more widgets and gadgets and like all those things are pretty strong, you know, secular trend lines like, you know, developers don't stop building a residential community because, you know, like power is going to be 10% more expensive or 20% more expensive in order to get the power to actually build the commercial development or the residential development. You know, you don't stop buying a second phone or an iPad, you know, just because, oh, like, you know, my power bill or my access to power. Right. Like, people just move forward with this.
- [52:10]
A
Right. Although if I can just try and reinflate the hype bubble just a little bit, I mean, it is the case, isn't it, that power demand in the United States was basically flat, let's say 2007, 08 through to the early 2000s, and now it is taking off. Now it is actually moving. And so I think it's worth thinking about that as being a significantly different thing that's happening, a significantly new development. And I think in terms of that incremental one, as you say, within data centers, there's a lot of other things that people use data centers for, not just AI. The world is definitely changing.
- [52:48]
D
Yeah, totally. I mean, like I said at the beginning, in my view, we are in an era of energy expansion. Right.
- [52:53]
A
Time for another question. I think might have time for one or two. Oh, yeah, one down the front here. Thanks.
- [52:59]
E
Thank you very much. I'm Ann Kathryn Mertz. I'm with First Ammonia. As we're talking about resiliency. I also wanted to ask you to comment about onshore and offshore wind. It's clear that there's a consensus that solar plus batteries has been really effective as a low cost, fast solution. We're seeing in ERCOT and Texas that we're getting much better at managing the spikes. But when we look at energy, of course the time that energy comes is available is very significant. And here onshore wind, when we look at Texas, you get wind at night. That's a huge advantage. And then when we look at offshore wind, with offshore, you get even higher capacity factors at different hours of the day, again uncorrelated. And so I'm curious what you see the path forward for wind in the United States, given that it has become very political, and I know we're weary about the politicization of these topics, but particularly when we look at onshore wind, it becoming so political has become a question mark for investment decisions. What does it mean if a president does not like a specific technology? And then also in terms of our allies, we're thinking many of these technologies are coming from European energy companies and potentially threatening their investments, then also hurts their energy industries. So I'm curious, this is a delicate topic, but what you see as a role for onshore and offshore wind in the United States, if any, or is the future really thinking of your tattoo, Neil, Is the future really solar batteries and gas?
- [54:29]
B
Wind is very commercial. Offshore wind, like you say, is a much steadier source, even of asset. It's definitely moving forward in other places and providing good measure to the grid. It was very challenged in the United States to start with because with the interest rates being high and some of the other supply chain issues with putting together such a big mega project, people were having to renegotiate terms as it was. And so I think it was sort of vulnerable already. And therefore some of it, of course, you know, you're stopping a project that's already being built. That's a totally different story. Hopefully they'll reissue the permit for that. Understanding all the different variables. What I would just say is, especially when you're doing giant infrastructure, and that goes to nuclear, right? And that goes to offshore wind. We have to absolutely get the technology right. I grew up in Boston, so what happened in Nantucket was really quite frightening. And it looks like, sorry, when you.
- [55:39]
A
Say what happened in Nantucket, I mean.
- [55:40]
B
There was this giant explosion of an offshore wind turbine. You know, it was touch and go for a couple of weeks about whether or not, I mean, it seems like everything's been worked out, but it was very dangerous in the first couple of days. I would liken it to sort of a Texas oil spill and its impact on the beaches in Nantucket. And we need to know. And it turned out there were some things that are now going to be fixed. But we need to know. It wasn't the first time that it happened. It's happened in other countries. And so we really need to know and make sure that everything is being done exactly correctly. It's an engineering thing. That's what I said about electricity. Right. We need to take the politics out of all of this stuff and we need to go with knowing that there's a regular process that's going to make sure that everything that gets built gets built safely, not because it's a pet of one person versus the pet of another person or that somebody's politically connected or not politically connected. And that goes for all of these technologies across the board. And so I think that that is really what I would say about offshore wind. Are we sure that every installation is going in properly? And then I would add to that, since we're not mentioning it all, you know, you're talking about, you know, Europe or New England or whatever. I mean, hydro is yet another, you know, untalked about resource. But when you're talking about data centers, you know, you mentioned Canada, are they going to go to places that have hydro? Because, you know, that's also like a really promising source for something like a data center and it has some flexibility to go up and down. Now let me just tell a two second anecdotal story because I know we're getting out of time, but I went to Europe this summer and to get a much cheaper airfare, I went to Iceland. And you know, they have these programs where you could get out of the plane and, you know, do a little sight thing in Iceland and then you go onto your destination. So my husband and I did that. And as we were driving past the geothermal plant that Iceland uses for all its hot water and electricity, the driver was explaining to us that this very dark brown and black soil that was around this area that had the geothermal plant was lava. And I started asking him, because I'm an energy expert, I started asking him questions about it because, you know, that seemed like pretty risky to me, like what if the lava hit the plant? And he admitted to me that the lava had knocked off the major hot water pipeline from that plant and people had no hot water in Iceland for several weeks. My Point is all of these facilities have risks associated with them and how we think about how to mitigate those risks. It doesn't mean you shouldn't do geothermal in Iceland. It doesn't mean we shouldn't do offshore wind. It just means that we need to make sure that in the process, and I understand, you know, speed, speed, speed, you know, we still have to have responsible human beings who know something about technology, who are going to actually look at the safety of all the things we use from oil and gas to coal, across the entire spectrum.
- [58:53]
C
If I could just very quickly on offshore wind and it kind of ties into a point I made earlier in the conversation about the politicians fictitization of all this wonky technical policy. In October of 2020, when I was the Trump appointed Republican chair of FERC, I led a technical conference at the Commission on Offshore Wind Integration because this was one of like the nuanced complexities that we needed to solve before we could move forward on it. This is the kind of record that independent agencies like FERC build that stay in place forever, that don't, or at least aren't supposed to flip and flop back and forth between different administrations. And that's where to the earlier question about Loper Bright and Chevron deference and the importance of these agencies. I want to take a moment to plug the importance of these critical agencies because like there was nothing political about this offshore wind interconnection technical conference. It was the nerdiest thing in the world, I'm certain. Maybe you were one of the only people that even understood what we talked about at it. But that record is critical for us to build upon to make progress in the future and that's why we need these things.
- [60:09]
A
Yeah, that is a great point and it's been fantastic talking to you all. Really great conversation. Unfortunately we do have to leave it there though. But thanks very much indeed, Neil. Thank you Amy. Thank you Cecilio. Great talking to you all. Thanks very much to our producers, Toby Biggins, Gilchrist and Dan Cottrell. Thanks very much to the Camera Council on Foreign Relations for hosting us in this fantastic venue. And above all thanks very much to all of you for listening. Thanks to everyone who's listening at home on the podcast, wherever you may be. Thanks to everyone who's been listening in the room as well. And we'll be back very soon with all the latest news and views on the future of energy. Until then, goodbye.