# Wired for the Future: Reinventing the US Power Grid

An episode of [Switched On](https://invisiblenetwork.ai/podcast/1469286286), 1 Oct 2025, 36 min.

Audio: https://podtrac.com/pts/redirect.mp3/traffic.omny.fm/d/clips/e73c998e-6e60-432f-8610-ae210140c5b1/01a10738-3608-4dec-883c-ae3c00026b36/7824ddcb-5d12-4e78-ae29-b3680175080b/audio.mp3?utm_source=Podcast&in_playlist=6d114f63-833d-4b78-bbf5-ae3c00026b44

The US power grid is under mounting strain from aging infrastructure, slow interconnections and regional bottlenecks, while demand, complexity and variability are on the rise. Building a future-ready grid will require major investment, smarter infrastructure and policy reform. So just what can be done to expand transmission capabilities to meet oncoming energy challenges in the US? At the BloombergNEF Summit Houston, Ethan Zindler, BloombergNEF’s head of country and policy research, moderated a panel “Revolutionizing the Grid.” This episode brings listeners that panel, which featured Page Crahan, general manager of Tapestry at X, Alphabet; Michael Skelly, co-founder and chief executive officer of Grid United; Frank Kreikebaum, senior vice president and chief of engineering at Smart Wires; and Rina Harris, vice president of strategic business growth and engagement at CenterPoint Energy. To learn more about BNEF’s Summits taking place around the world and to see recordings of BNEF Talks at previous Summits, head to https://about.bnef.com/events/type/summit/ See omnystudio.com/listener for privacy information.

## Transcript

[0:00] Speaker 1: This is Standra Perkins, and you're listening to Switch on
[0:03] Speaker 1: the podcast brought to you by BNF. The aging US
[0:07] Speaker 1: grid was built at a simpler time, meeting the needs
[0:10] Speaker 1: of a less diverse supply and demand mix on the grid,
[0:14] Speaker 1: but today it's struggling to keep up with the variety
[0:17] Speaker 1: of new energy projects coming online alongside new demand sources
[0:22] Speaker 1: like data centers and electric vehicle charging, leading to interconnection queues.
[0:27] Speaker 1: Upgrading the grid is essential as expectations for reliable and
[0:31] Speaker 1: flexible energy increase alongside what is seemingly endless energy demand.
[0:37] Speaker 1: On today's show, we bring you a discussion about the
[0:40] Speaker 1: future of the US energy grid from BNF's summit in Houston,
[0:44] Speaker 1: where BNF Head of Country and Policy Research Ethan Zindler
[0:48] Speaker 1: spoke with Paige Crahan, general manager Tapestry at x Alphabet,
[0:54] Speaker 1: Michael Skelly, co founder and CEO of Grid United, Frank Kreikenbaum,
[0:59] Speaker 1: c your, Vice president and chief of Engineering at Smart Wires,
[1:04] Speaker 1: and Rena Harris, vice president of Strategic Business, Growth and
[1:07] Speaker 1: Engagement for Centerpoint Energy. During the discussion, the panel addressed
[1:11] Speaker 1: what can be done not only to facilitate upgrading existing
[1:15] Speaker 1: grids and who should pay for it, but also what
[1:18] Speaker 1: efficiencies can be found to allow for the better use
[1:21] Speaker 1: of infrastructure that's already placed. They also highlight what can
[1:25] Speaker 1: be learned from outside of the US, from policy to
[1:29] Speaker 1: infrastructure upgrades, alongside new smart management technologies. If you'd like
[1:34] Speaker 1: to see agendas and videos from BNF summits taking place
[1:37] Speaker 1: around the world, including the summit in Houston, go to
[1:41] Speaker 1: BNAF dot com and check us out at our events tab.
[1:45] Speaker 1: Right now, let's head to Ethan's panel from BNF summ
[1:48] Speaker 1: at Houston, titled Revolutionizing the Grid.
[2:01] Speaker 2: Hi, Everybody, Ethan Zendlaer with Bloomberg and EF.
[2:03] Speaker 3: So.
[2:04] Speaker 4: Through the morning, we've been hearing a great deal about
[2:07] Speaker 4: all the exciting developments around the expansion of AI and
[2:11] Speaker 4: the potential that that provides, and of course, as the
[2:14] Speaker 4: top line, after least twenty years or so of no
[2:19] Speaker 4: real electricity demand growth, that is now obviously changing. But
[2:24] Speaker 4: what this panel is going to do is talk a
[2:26] Speaker 4: little bit about how we actually accommodate getting all those
[2:29] Speaker 4: electrons from where they're produced to where they're needed, and
[2:33] Speaker 4: what smart ways there might be to kind of organize the.
[2:36] Speaker 2: Delivery of those in a better way.
[2:39] Speaker 4: Ultimately, our grid at the moment is essentially overmatched and
[2:45] Speaker 4: dramatically underfunded. We have about two terrawatts of new projects
[2:50] Speaker 4: and various stages that want to get onto the grid.
[2:52] Speaker 4: For those who aren't electricity people, that's about twice as
[2:55] Speaker 4: much capacity as we actually have on the grid at
[2:58] Speaker 4: the moment. According to recent studies in the state of
[3:02] Speaker 4: the New York ISO, about forty percent of all of
[3:06] Speaker 4: the lines there are over thirty years old. And we've
[3:10] Speaker 4: seen similar stories around the world. So we have an
[3:12] Speaker 4: aging grid, and we have a great deal of additional
[3:15] Speaker 4: demand that's coming, and I have a fantastic panel that's
[3:18] Speaker 4: going to provide all the solutions to that. So I
[3:21] Speaker 4: want to start, though, first with Paige from Tapestry at
[3:25] Speaker 4: Google to talk really at a high level about some
[3:28] Speaker 4: of the work that you guys are doing in terms
[3:30] Speaker 4: of thinking about the system from a macro perspective.
[3:33] Speaker 5: Thank you so much, and perhaps I'll first share a
[3:36] Speaker 5: bit about what Tapestry is. So Alphabet is Google's parents.
[3:40] Speaker 5: Company has an innovation lab. It's called Google X. Our
[3:45] Speaker 5: job is to invent and launch radical technologies that will
[3:48] Speaker 5: make the world a better place and build businesses from
[3:50] Speaker 5: those technologies. Probably one of the best known graduates is
[3:56] Speaker 5: a company called Waimo that does self driving cars, and
[4:00] Speaker 5: so it is of that ilk that I started working
[4:03] Speaker 5: on this effort called Tapestry, which is our moonshot for
[4:07] Speaker 5: the electric rid. So taking that same thinking, you know
[4:12] Speaker 5: something that is not always present for us, but the
[4:15] Speaker 5: electric rid itself is humankind's largest machine, the biggest invention,
[4:20] Speaker 5: the most expensive, massive thing we've ever built as a species,
[4:25] Speaker 5: from how you generate, how you transmit, how you distribute,
[4:29] Speaker 5: to sort of what we might call the last mile,
[4:31] Speaker 5: how you consume electricity, and it is not.
[4:36] Speaker 6: Managed as a single system.
[4:39] Speaker 5: And that really is what Tapestry thinks about is Okay,
[4:41] Speaker 5: we have great things happening at any one of those
[4:45] Speaker 5: specific areas of our system. We have innovation happening in generation,
[4:51] Speaker 5: we have innovation happening in transmission and components, in critical minerals,
[4:55] Speaker 5: all the things that go into an innovation happening. On
[4:58] Speaker 5: the demand side, what are we using you know, when
[5:01] Speaker 5: should I charge my electric car if you have one?
[5:03] Speaker 5: But there is a gap in systems thinking technology, which
[5:09] Speaker 5: is what we're seeking to fill. And what are the
[5:11] Speaker 5: technologies that Tapestry and our software and machine learning team
[5:17] Speaker 5: can bring to bear to meet the needs of this
[5:20] Speaker 5: large Humankind's largest machine that needs to double in size.
[5:26] Speaker 5: Just for a perspective, we have about eighty million kilometers
[5:30] Speaker 5: of electric rid on planet Earth. It took us one
[5:33] Speaker 5: hundred and thirty years as a species to build about
[5:36] Speaker 5: eighty million kilometers, and we need to build eighty million
[5:39] Speaker 5: more kilometers in the next fifteen years to meet.
[5:42] Speaker 6: Our energy needs.
[5:43] Speaker 5: So in our mind, it is not just going to
[5:46] Speaker 5: be keep doing what you're doing to meet that need.
[5:50] Speaker 5: We need to do things differently, more intelligently, and in
[5:52] Speaker 5: a more coordinated fashion.
[5:54] Speaker 4: So thanks for that for and I think that eighty
[5:58] Speaker 4: million figure is both transmission and distribution. Distribution for those
[6:03] Speaker 4: is the last lines to the people's store. But my
[6:06] Speaker 4: understanding is that last year, I think we built three
[6:08] Speaker 4: hundred miles of high voltage transmission in the US. Michael,
[6:13] Speaker 4: I know you're trying to build a whole lot more
[6:15] Speaker 4: than that. Maybe tell us a little bit about Grid
[6:17] Speaker 4: United and the work that you're doing.
[6:18] Speaker 7: Great, thank you.
[6:20] Speaker 3: So we're working on big transmission lines to connect the grids.
[6:26] Speaker 3: So we have three grids in the continent of the
[6:28] Speaker 3: United states, Eastern, Western, and or cut and these are
[6:33] Speaker 3: very lightly or poorly connected. And if you want to
[6:39] Speaker 3: improve reliability, create more efficient markets, serve new load, and
[6:45] Speaker 3: integrate new sources of supply, one of the most cost
[6:50] Speaker 3: effective things that you can do is connect these grids
[6:52] Speaker 3: together like a lot of networks. Networks work better when
[6:57] Speaker 3: they're better connected, and so that's what we do. And
[7:00] Speaker 3: the reason that we are doing this is because traditionally
[7:05] Speaker 3: the way we plan the grid, well, we used to
[7:06] Speaker 3: plan it sort of by city. The city Houston would
[7:09] Speaker 3: plan a grid back in the last century. And then
[7:13] Speaker 3: over time we connected the grids and we started planning
[7:15] Speaker 3: the grid by states, and then we came up with
[7:17] Speaker 3: these regional transmission organizations to plan it by multi state
[7:22] Speaker 3: regions now, but there's never been sort of a national
[7:26] Speaker 3: plan to connect these grids. So that's what we're working on.
[7:32] Speaker 3: And our business model is to identify projects that we
[7:38] Speaker 3: think are promising and valuable to the system and then
[7:42] Speaker 3: work to get the rights away and permits and work
[7:46] Speaker 3: with like the millions of stakeholders that you need to
[7:48] Speaker 3: touch base with before you attempt something like this. Put
[7:52] Speaker 3: the pieces together, and then we will involve utilities on
[7:56] Speaker 3: either side of our line as the old owners of
[8:01] Speaker 3: some or all of the projects that we put together,
[8:04] Speaker 3: and so we're sort of agnostic with respect to outcomes
[8:08] Speaker 3: in terms of who owns the projects. But we know
[8:12] Speaker 3: that you know that that new big lines are incredibly helpful,
[8:18] Speaker 3: and so that's what we're trying to make happen.
[8:19] Speaker 4: Just real quick, Michael, maybe can you give a couple
[8:22] Speaker 4: of states where you're operating or in what state lines
[8:25] Speaker 4: you're trying to cross.
[8:26] Speaker 3: Yeah, so we're active in about almost a dozen states,
[8:29] Speaker 3: and so we're connecting the Eastern grid near Bismarck with
[8:35] Speaker 3: the Western grid near a town called colst of Montana Pueblo,
[8:40] Speaker 3: Colorado in the west, to Guymon, Oklahoma, and the east
[8:45] Speaker 3: and so on, and the mechanisms. The reason that these
[8:50] Speaker 3: just to explain a little bit more that these connections
[8:52] Speaker 3: are so valuable, is weather events are typically hundreds of
[8:58] Speaker 3: miles wide, but not.
[9:00] Speaker 7: Thousands of miles wide.
[9:01] Speaker 3: So if you connect the grids, like the Eastern grid
[9:03] Speaker 3: is getting hit with one weather event and there's power
[9:06] Speaker 3: in the west and vice versa. Similarly, you have different
[9:10] Speaker 3: generation patterns, so in the spring, there's lots of access
[9:15] Speaker 3: hydroelectric power in the Pacific Northwest, we provide an outlet
[9:18] Speaker 3: to the east. In fall nights or winter nights in
[9:24] Speaker 3: the Midwest, there's typically an excess of wind, and so
[9:28] Speaker 3: we give an access.
[9:29] Speaker 7: An outlet to that capacity going to the west.
[9:33] Speaker 4: Beskay, I want to come back on sort of how
[9:35] Speaker 4: well we're actually able to take advantage of those kind
[9:38] Speaker 4: of differences given the lack of interconnectedness at the moment.
[9:41] Speaker 2: But I want to move on to Frank.
[9:43] Speaker 4: To hear a little bit more about smart Wires and
[9:46] Speaker 4: the work that you're doing, because I think it's interesting.
[9:48] Speaker 4: I feel like we have two folks here on this panel.
[9:50] Speaker 4: We're very much involved in the actual delivery of electricity,
[9:54] Speaker 4: and then two were thinking about how to do the
[9:55] Speaker 4: whole job maybe a little more efficiently, including you, Fanks.
[9:58] Speaker 4: So maybe tell us a little bit about smart Wires.
[10:01] Speaker 7: Sure, thanks for having me.
[10:02] Speaker 8: And smart Wires is a fifteen year old company that
[10:06] Speaker 8: effectively is on a mission to better utilize the existing
[10:09] Speaker 8: capacity of the network. And when we talk about a
[10:14] Speaker 8: constraint limiting a data center connection or a new generation,
[10:17] Speaker 8: it always comes back in the form of in a study,
[10:20] Speaker 8: some line was thermally overloaded. But what we find in
[10:24] Speaker 8: the transmission network even under the peak condition peak top
[10:29] Speaker 8: most warm day with lines out of service, the rest
[10:33] Speaker 8: of the system is on average thirty percent utilized. So
[10:36] Speaker 8: it's such a tremendous amount of latent capacity in that network.
[10:40] Speaker 8: But until now there are not effective ways to route
[10:43] Speaker 8: power away from those overloaded assets and onto underutilized assets.
[10:48] Speaker 7: And for the last.
[10:49] Speaker 8: Fifteen years we've been commercializing and advancing that technology and
[10:53] Speaker 8: have now done commercial projects delivering a total four gigawats
[10:57] Speaker 8: of capacity globally. Recently that work was mostly based in
[11:02] Speaker 8: Europe and Latin America, but for some of the drivers
[11:05] Speaker 8: we've talked about today, the US market is suddenly very
[11:09] Speaker 8: interesting and we're seeing a lot of up to we've
[11:12] Speaker 8: seen primarily renewable connection, generation connection, and most importantly recently,
[11:17] Speaker 8: data center connection is driving this and we've done some
[11:20] Speaker 8: projects where we've helped connect data centers in half the
[11:24] Speaker 8: time of other.
[11:25] Speaker 7: Solutions that utilities have at their disposal.
[11:28] Speaker 8: And so for US, it's about being one solution of
[11:31] Speaker 8: many that utilities can use to bridge and then also
[11:37] Speaker 8: make their longer term capital plans more effective and more
[11:42] Speaker 8: compelling from a value perspective.
[11:44] Speaker 4: And that's one one quick clarifying question, which is just
[11:47] Speaker 4: to say four gigaots of capacity, but you're not building
[11:50] Speaker 4: new transmissions, So what does that mean that you're able
[11:53] Speaker 4: to sort of reroute four gigaotts of electrons and various
[11:56] Speaker 4: directions so that that accommodates for it.
[11:59] Speaker 8: So we may we make physical hardware that sits inside
[12:01] Speaker 8: utility substations, fully owned and controlled by utilities, and in
[12:06] Speaker 8: real time that hardware will direct power onto or off
[12:10] Speaker 8: of utility assets, usually overheadlines, but could be an underground cable,
[12:14] Speaker 8: could be a transformer that's overloaded that takes too long
[12:16] Speaker 8: to replace. And so what that means is when you
[12:21] Speaker 8: do the studies and you say how much capacity can
[12:23] Speaker 8: I import from North Texas into Houston. If the number
[12:26] Speaker 8: is eleven point two gigawats, you make some modifications to
[12:30] Speaker 8: the flow of the system and that number could go
[12:33] Speaker 8: to twelve, fourteen, sixteen.
[12:34] Speaker 7: We've done some work in the UK.
[12:36] Speaker 8: We literally had a constraint at eleven gigawatts and we
[12:39] Speaker 8: took that number up to sixteen gigwats just putting these
[12:42] Speaker 8: devices in substations.
[12:45] Speaker 4: Thanks for that, Andriena, I want to bring you into
[12:47] Speaker 4: the conversation. Centerpoint Obviously you do a bunch of things,
[12:51] Speaker 4: and maybe I come to you last only because you
[12:55] Speaker 4: literally do the last mile on distribution that I know
[12:57] Speaker 4: you're also involved in transmission as well. So maybe tell
[13:01] Speaker 4: us a little bit about the role that you play
[13:03] Speaker 4: in all of this and maybe some of the unique
[13:05] Speaker 4: circumstances that you have in URCOT that maybe the rest
[13:09] Speaker 4: of us outside of this region don't have.
[13:11] Speaker 9: Yeah, absolutely, well for small, good afternoon everyone. So you know,
[13:16] Speaker 9: Centerpoint Energy is headquartered here in Houston. We operate in
[13:21] Speaker 9: four states, serve natural gas across Minnesota, Indiana, Ohio, and Texas,
[13:26] Speaker 9: but Texas is our biggest market for electricity, so we
[13:29] Speaker 9: only provide electricity in Texas and Indiana and Indiana. We're
[13:33] Speaker 9: vertically integrated, so we do generation and we deliver in Texas,
[13:39] Speaker 9: specifically around the twelfth County Houston area. We're kind of
[13:42] Speaker 9: the Amazon of power, right, so we don't generate it,
[13:45] Speaker 9: but we do deliver it. So our role in the
[13:48] Speaker 9: ERCOT market is quite unique. One of the things that
[13:53] Speaker 9: I will say is that over the past few years,
[13:55] Speaker 9: and I'll admit that I've moved to Houston a couple
[13:57] Speaker 9: of years ago. So I have some you know, lineup
[14:00] Speaker 9: side to how other markets operate in the growth that
[14:03] Speaker 9: other areas experience. In the Greater Houston area, the growth
[14:07] Speaker 9: that we see here in Texas and then the air
[14:10] Speaker 9: market is just absolutely unbelievable. Just over the past year,
[14:15] Speaker 9: you know, like our interconnection queue has doubled with a
[14:18] Speaker 9: lot of the onset of AI and advanced manufacturing life sciences.
[14:24] Speaker 9: Houston traditionally has been known for growth.
[14:29] Speaker 6: We grow at two percent year over year.
[14:32] Speaker 9: So you know, the grid in Texas, in the Greater
[14:36] Speaker 9: Houston area, you know, is capable of handling really great
[14:41] Speaker 9: amounts of power, but it took us one hundred years
[14:44] Speaker 9: to get here, right, so we support twenty two viigoatts
[14:47] Speaker 9: of peak demand power today. Took us one hundred years
[14:51] Speaker 9: to get here, and then over the next ten years,
[14:54] Speaker 9: we're projecting that that will double with the onset of
[14:57] Speaker 9: all of these new technologies. So we're doing a lot
[15:01] Speaker 9: of different things, I think collaboration and partnerships in a
[15:06] Speaker 9: number of ways. Obviously ARCOT working with ARCOT, working with
[15:09] Speaker 9: our regulators on policy and everything we need to do
[15:12] Speaker 9: with them to ensure that we have the right tools
[15:15] Speaker 9: in place to really move a light speed speed to
[15:19] Speaker 9: market is the name of the game, right, And so
[15:21] Speaker 9: I have the pleasure of being up here with you know,
[15:24] Speaker 9: all of these colleagues that are going to help us
[15:27] Speaker 9: be even better. So they're part of the solution in Texas.
[15:32] Speaker 9: And then you know, obviously we have an amazing ecosystem
[15:35] Speaker 9: with the partners that we have, whether they're you know,
[15:40] Speaker 9: energy advisors.
[15:41] Speaker 6: I know new markets here.
[15:42] Speaker 9: They were helpful in getting center point up here on
[15:46] Speaker 9: the panel. But we have a lot of economic development coalitions,
[15:49] Speaker 9: we have land developers, we have state local so it
[15:54] Speaker 9: really is a regional approach to be able to really
[15:57] Speaker 9: deliver power and advocate for the.
[16:00] Speaker 6: Right policies to go fast.
[16:03] Speaker 9: And so that's what we're here to do, is just
[16:05] Speaker 9: to figure out how do we support this awesome ecosystem
[16:09] Speaker 9: to continue to grow at the pace that is in
[16:11] Speaker 9: front of us.
[16:13] Speaker 4: Look, I think if we established that the need for
[16:16] Speaker 4: more electricity and supply and more transmission distribution to get
[16:20] Speaker 4: it there is going.
[16:22] Speaker 2: To be required.
[16:23] Speaker 4: I guess the next question I have is who's going
[16:26] Speaker 4: to pay for this and how does that work? I mean,
[16:31] Speaker 4: Michael may be starting with you on the assumption that
[16:33] Speaker 4: we have some number of people who are maybe not
[16:35] Speaker 4: electricity people, how to train how do transmission operators get compensated?
[16:40] Speaker 4: And maybe also how would you like to get compensated
[16:43] Speaker 4: if the way it's set up right now isn't the best?
[16:46] Speaker 3: Yeah, okay, so we so Traditionally the way we pay
[16:50] Speaker 3: for the grid i e.
[16:52] Speaker 2: All the wires that you see.
[16:54] Speaker 3: As you drive around is that we you have grid planners,
[16:59] Speaker 3: like in let's take an example of Arcot because we're
[17:02] Speaker 3: a comeback unit here, and the system planners look at, Okay,
[17:09] Speaker 3: how much power are we going to need, where do
[17:10] Speaker 3: we need it, how we're going to get it, where's
[17:12] Speaker 3: generation likely to come from? And then the archive plans
[17:17] Speaker 3: a grid to meet those needs, and episodically, the legislature
[17:23] Speaker 3: and you heard about SB six on a panel on
[17:26] Speaker 3: an earlier panel, but episodically the legislature says.
[17:29] Speaker 2: Hey, we want a little more of this or a
[17:32] Speaker 2: little bit less.
[17:33] Speaker 3: And so in the two thousands there was something called
[17:36] Speaker 3: KRES which was designed to get wires built out to
[17:40] Speaker 3: rural areas to tap renal energy. And that happened and
[17:46] Speaker 3: that was a big burst. And then recently there was
[17:50] Speaker 3: a direction for the legislature to help electrify the premium basin,
[17:55] Speaker 3: so they stepped in.
[17:56] Speaker 7: And you see these types of interventions elsewhere.
[17:58] Speaker 3: But generally speaking, the grid planners look at what the
[18:01] Speaker 3: needs are and then they plan the grid, and sometimes
[18:04] Speaker 3: they adjust like the maximum voltage because hire the voltage
[18:07] Speaker 3: and the more power than you can use, and then
[18:09] Speaker 3: the costs for all that when they decide to build
[18:12] Speaker 3: a new line, those get spread across all customers. And
[18:16] Speaker 3: that's how we've built the grids for the last you know,
[18:19] Speaker 3: one hundred and some years. And then occasionally we have times,
[18:23] Speaker 3: for example, during World War Two and during the sixties
[18:26] Speaker 3: and seventies when we had big grid expansions, and those
[18:29] Speaker 3: grid expansions were the political consensus was, yeah, let's all
[18:34] Speaker 3: pay for them because we all benefit for them, either
[18:38] Speaker 3: for national defense or economic growth reasons. And now we
[18:42] Speaker 3: have it, we're sort of at a new moment, and
[18:45] Speaker 3: it's a little bit of a different question because we
[18:49] Speaker 3: need to build out the grid to support large, very
[18:54] Speaker 3: concentrated loads. Okay, and so under the inciple of sort
[19:00] Speaker 3: of causality, people, some people are commissions. Public utility commissions
[19:06] Speaker 3: will ultimately decide who's going to pay and what's going
[19:09] Speaker 3: to get author us. They look at these these new
[19:12] Speaker 3: loads that cause impacts to the system, that require upgrades
[19:17] Speaker 3: to the system. A few years ago they say, this
[19:19] Speaker 3: is great, we get this AI center. Let's all pay
[19:22] Speaker 3: for that. This is economic development. They're going to pay taxes.
[19:25] Speaker 3: And furthermore, this new load is going to be paying
[19:29] Speaker 3: for this, paying into the system because they're going to
[19:32] Speaker 3: use the grid for many, many years to come. That
[19:35] Speaker 3: consensus has shifted in the last couple of years, and
[19:40] Speaker 3: there's big questions where the you know, regulators and so
[19:43] Speaker 3: on are asking, well, okay, we're going to pay for
[19:45] Speaker 3: these upgrades, but what if you know, Meta goes away
[19:49] Speaker 3: in five years and then we got this whole system
[19:51] Speaker 3: that you know, we're all paying for and they're not
[19:54] Speaker 3: there to contribute their share.
[19:56] Speaker 2: So there's been sort of a revisiting.
[19:59] Speaker 3: And in general there's a movement toward large new loads
[20:07] Speaker 3: paying for they're required, not just punctual, you know, point
[20:13] Speaker 3: source upgrades, but large upgrades to the grid. And that
[20:18] Speaker 3: discussion is unfolding really across the country at the level
[20:22] Speaker 3: of pretty much every public utility commission around the country.
[20:25] Speaker 7: We're trying to sort this out.
[20:27] Speaker 2: So that's one sort of pay for question.
[20:31] Speaker 3: The other really big pay for question that we're trying
[20:34] Speaker 3: to sort out. And Toby Rice referred to some of
[20:36] Speaker 3: the discussions around permit reform and EPRA. If you're a
[20:41] Speaker 3: nerd like me and you followed the permit bill in
[20:45] Speaker 3: the last Congress, there were provisions in there that said, hey,
[20:48] Speaker 3: you know what, the grid is not keeping up while
[20:50] Speaker 3: we're upgrading, while we're figuring out this permitting, let's come
[20:54] Speaker 3: up with mechanisms to expand the grid, and to come
[20:59] Speaker 3: up with mechanism so that you know, beneficiaries who might
[21:02] Speaker 3: be a few states away but who might benefit from
[21:05] Speaker 3: grid upbradge, they get to contribute to that as well.
[21:08] Speaker 3: So break down some of the regional barriers that I
[21:11] Speaker 3: mentioned and come up with a national mechanism to get
[21:15] Speaker 3: the grid that we that we really need.
[21:17] Speaker 4: So clear, that's legislation that has not passed, that has
[21:20] Speaker 4: not passed and may never pass and may never pass.
[21:23] Speaker 3: But look, we talked about it before the show here
[21:26] Speaker 3: and we're all octivists up here, so we think good
[21:29] Speaker 3: things like that are going to happen.
[21:30] Speaker 2: I mean, I see you nodding over there.
[21:32] Speaker 4: I mean, are you having similar conversations with data center
[21:35] Speaker 4: providers and and I guess the question is who's paying
[21:40] Speaker 4: for for all this new new infrastructure required to accommodate them, Yes,
[21:46] Speaker 4: and is it going to be rate payers? Because I
[21:48] Speaker 4: think that's a question a lot of us are wondering about.
[21:50] Speaker 9: So affordability is always something that utilities have to think
[21:53] Speaker 9: about because we're highly regulated and so you know, all
[21:57] Speaker 9: of our costs have to be prudent and justify us
[22:00] Speaker 9: and useful, and you know, whenever there's not a lot
[22:04] Speaker 9: of jobs that are maybe brought to the community and
[22:06] Speaker 9: small community, you know, all those costs go back to
[22:08] Speaker 9: the rate payers. So depending on what kind of market
[22:10] Speaker 9: you're in and how large your service territory is and
[22:12] Speaker 9: how it impacts the residential rate the commercial rate industrials, right,
[22:18] Speaker 9: so we just have to create models, which there are
[22:22] Speaker 9: several that are out there. You know, each jurisdiction probably
[22:26] Speaker 9: does a little bit differently based on their tariffs and
[22:29] Speaker 9: their regulations. But you know, a lot of the costs
[22:33] Speaker 9: are being born back onto the data centers and oftentimes
[22:38] Speaker 9: again it's speed to market, so they're willing to pay
[22:42] Speaker 9: not saying that cost isn't important. Cost obviously is important
[22:45] Speaker 9: to everyone, but sometimes pay to play is the name
[22:49] Speaker 9: of the game, but not always the case.
[22:52] Speaker 4: Frank one to bring you in and ask you how
[22:54] Speaker 4: your business model works, because historically, energy efficiency within the
[22:58] Speaker 4: grand site of things is something that's not always easy
[23:00] Speaker 4: to get compensated for. If you're providing an efficiency service
[23:04] Speaker 4: to the grid, how do you guys lay money?
[23:06] Speaker 8: Yeah, so we're not an efficiency provider. We're not out
[23:08] Speaker 8: there managing people's thermostats, are lowering voltage to get more efficiency,
[23:12] Speaker 8: and we fundamentally allow transmission owners to get more out
[23:17] Speaker 8: of their system. And for most of our projects, those
[23:21] Speaker 8: have been rate based projects.
[23:23] Speaker 7: But I do have a couple of reflections.
[23:25] Speaker 8: I started my career as a consultant to renewable developers
[23:29] Speaker 8: and so I'm very familiar with pay to play. And
[23:31] Speaker 8: you get a study back from your ISO, except for
[23:33] Speaker 8: in Texas that says your proposed generation project is going
[23:40] Speaker 8: to increase the cost of this system because these three
[23:44] Speaker 8: lines have to be reconductor and win those costs for
[23:48] Speaker 8: minimal And you put in a five hundred megal application,
[23:52] Speaker 8: you got one three mile line that.
[23:53] Speaker 7: Need to be reconducted.
[23:55] Speaker 8: You were fine with pay to play, right, but the
[23:57] Speaker 8: generation businesses had to deal with this for a long time.
[24:00] Speaker 8: So the data center businesses could learn a lot from
[24:02] Speaker 8: that experience and the pain that those entities had. The
[24:06] Speaker 8: reality is today, it's not you come back with, oh,
[24:10] Speaker 8: it's a couple million dollars upgrade, we'll get it done
[24:12] Speaker 8: in twelve months. It's everything is so close to its limit.
[24:15] Speaker 8: Oh we think it's at its limit that you're getting
[24:18] Speaker 8: these huge sums back. And what we're seeing in the
[24:21] Speaker 8: projects we've dealt with data centers and helping data centers
[24:23] Speaker 8: get online, it's not about the dollars per se, even
[24:26] Speaker 8: though there are trade offs because they're building portfolios of
[24:29] Speaker 8: projects across multiple sites, so they will pick the best cost.
[24:32] Speaker 8: It's about the timeline. They cannot afford the timeline. So
[24:36] Speaker 8: what we need to do as an industry is find ways,
[24:39] Speaker 8: whether it's tablishry, whether it's us, whether it's other solutions
[24:42] Speaker 8: that allow you to get more out of that existing network.
[24:46] Speaker 8: So the ticket that we're handing over to the developer,
[24:49] Speaker 8: whether it's a manufacturer, data center generator, is not the
[24:53] Speaker 8: number of today, but the number of ten to fifteen
[24:55] Speaker 8: years ago, where business could proceed at a timescale that
[24:59] Speaker 8: was managed and a cost that was manageable. So that's
[25:02] Speaker 8: the first side of it. It's rate based business. But
[25:05] Speaker 8: we still have better offerings to our interconnecting partners for
[25:09] Speaker 8: progress to proceed.
[25:09] Speaker 7: We heard about that a couple panels back.
[25:11] Speaker 8: If we don't do that, we're going to lose some
[25:13] Speaker 8: of these global races. The other side of it is
[25:17] Speaker 8: in the UK market specifically, and this is only the
[25:19] Speaker 8: work that we've done. In the UK, they do have
[25:20] Speaker 8: a different regulatory model and they basically say, this is
[25:24] Speaker 8: the rate of return that we're going to allow utilities
[25:26] Speaker 8: to do to achieve for a period five ten year period.
[25:31] Speaker 8: But if you're able to achieve better performance of that
[25:34] Speaker 8: than that, because you deploy a technology that's one third
[25:37] Speaker 8: of the cost of your best in class previously, you
[25:41] Speaker 8: as a utility get to keep half of that value
[25:44] Speaker 8: and society gets to keep the other half. And what
[25:47] Speaker 8: we've seen is that the regulator will set price points
[25:51] Speaker 8: for this particular boundary Nor Texas to Central Texas. We
[25:55] Speaker 8: assume that to Adam Meg out there, it's a five
[25:58] Speaker 8: thousand dollars for that maywa if they can do it
[26:01] Speaker 8: for two thousand. They keep that delta right, and so
[26:03] Speaker 8: that's a model that has encouraged a lot of activity.
[26:06] Speaker 8: I know that's not the right model for every jurisdiction,
[26:08] Speaker 8: but I just we need to learn from others when
[26:11] Speaker 8: it makes sense that there are other models available if
[26:13] Speaker 8: you really want to incent action, and that's one of
[26:15] Speaker 8: them that has been very effective.
[26:17] Speaker 3: Yeah, that's a great point. I mean, the way we
[26:22] Speaker 3: pay for the grid in the US is typically pay
[26:28] Speaker 3: you know, you invest x dollars, you get x percent
[26:30] Speaker 3: ready to return on that and overdone. And we were
[26:33] Speaker 3: not very creative. And Frank's referring to off Chen of
[26:37] Speaker 3: the UK. They have very innovative mechanisms to where by
[26:42] Speaker 3: transmission owner bear some of the risk and get some
[26:45] Speaker 3: of the upside and that results in a system which
[26:48] Speaker 3: is less costly.
[26:49] Speaker 7: So part of the pay for is.
[26:52] Speaker 2: Figuring out who's going to pay.
[26:54] Speaker 3: And then another part of the equation that we sort
[26:57] Speaker 3: of ignore, I think at our peril is how do
[27:01] Speaker 3: we comble with with pay for mechanisms that put more
[27:06] Speaker 3: of the risk on well, on people like us, for example.
[27:09] Speaker 4: I want to bring you into the conversation as well
[27:12] Speaker 4: on this one. We're just talking about the UK. I
[27:15] Speaker 4: know tapipe's been reactive in Chile. I believe New Zealand
[27:19] Speaker 4: as well, maybe less so in the US.
[27:22] Speaker 2: Is there a reason for that?
[27:23] Speaker 6: Yeah, Well, we are US cased, so we would like
[27:25] Speaker 6: to be very active in the US.
[27:27] Speaker 5: And there are reasons, I think when we think about
[27:32] Speaker 5: the big problem, which is how do we deliver electricity
[27:35] Speaker 5: to meet humanities needs? Now we have load growth, but
[27:38] Speaker 5: we also needed to be reliable and affordable.
[27:40] Speaker 6: We have to do both of those things.
[27:42] Speaker 5: That's global certainly happening for a lot of reasons. Big
[27:45] Speaker 5: problem also means big opportunity. If you fix it, great
[27:48] Speaker 5: things will happen. There are some places in the world
[27:51] Speaker 5: that allow a company like US that starts from the
[27:56] Speaker 5: technical like, what is the technology that could unlock what
[27:59] Speaker 5: is really a systems problem? Economics, the social license, regulatory
[28:03] Speaker 5: you know, how does this business model work?
[28:05] Speaker 6: It is truly one of the deepest systems problems.
[28:08] Speaker 5: But when we thought about is there a technology that
[28:11] Speaker 5: would sort of allow us to solve this thing more efficiently?
[28:14] Speaker 5: There are places in the world that we're far easier
[28:18] Speaker 5: for Tapestry to start learning as quickly as.
[28:20] Speaker 6: We could if our technology was useful.
[28:22] Speaker 5: I say all the time, you know, the physics are
[28:24] Speaker 5: the physics when I simulate a grid, the laws of
[28:27] Speaker 5: physics in New Zealand, we work in the UK, in
[28:30] Speaker 5: New Jersey, in Chile, those are the laws of physics.
[28:33] Speaker 5: And so the faster that I can learn what is
[28:36] Speaker 5: going to happen when you know, three wind farms are
[28:39] Speaker 5: located next to each other.
[28:40] Speaker 6: And it's a hot day and we have a lot
[28:42] Speaker 6: of electric.
[28:42] Speaker 5: Vehicles, then the next grid doesn't have to learn that
[28:45] Speaker 5: same lesson or wait and simulate and test.
[28:49] Speaker 6: We just say this is what happens. We've seen this,
[28:51] Speaker 6: we model this.
[28:52] Speaker 5: And so the reason Tapestry, you know, looked at, you know,
[28:56] Speaker 5: outside of the US to start is what is the
[28:58] Speaker 5: fastest place that I can learn and how can I
[29:01] Speaker 5: scale that learning? And Chile was an incredible example. We
[29:05] Speaker 5: work with the National System Operator there. They have a
[29:09] Speaker 5: system that is similar to URKAT, which where interconnection ques aren't.
[29:13] Speaker 6: As big of an issue.
[29:14] Speaker 5: It's it's you show up, you connect to the grid,
[29:17] Speaker 5: and you manage what happens afterwards.
[29:20] Speaker 6: Their issue was planning, expansion planning, and so.
[29:25] Speaker 5: The data about their national grid is publicly available. I'm
[29:30] Speaker 5: not advocating for or against. It is just a statement
[29:32] Speaker 5: of fact. When I looked at how could I learn
[29:34] Speaker 5: as quickly as possible, we were like, wow, we can
[29:36] Speaker 5: start simulating this grid tomorrow, you know, loosely, and then
[29:39] Speaker 5: what I can learn, you know, benefits that the Chilean
[29:42] Speaker 5: you know, citizens and national and grid. But also I
[29:44] Speaker 5: can take those lessons, those technology lessons outside, and so
[29:48] Speaker 5: when I think about things that we can do in
[29:50] Speaker 5: the US, we have to make it easier to innovate.
[29:53] Speaker 5: We've got incredibly brilliant people building amazing things that are
[29:57] Speaker 5: getting stuck with, you know, even.
[30:00] Speaker 6: Something as simple and road as the procurement.
[30:02] Speaker 5: Process like the by the time you fill out your
[30:05] Speaker 5: form of how you're going to apply AI to transform
[30:08] Speaker 5: the network, it's two years after you fill out the form.
[30:11] Speaker 6: And the technology is already outdated.
[30:13] Speaker 5: And that is unacceptable for in my opinion, for a
[30:16] Speaker 5: challenge this big. And so yes, we have looked Tapestry
[30:20] Speaker 5: is looked outside the US to start, and we're super
[30:22] Speaker 5: excited about bringing those lessons as quickly as we can,
[30:26] Speaker 5: you know, sort of globally. And now we're working in
[30:28] Speaker 5: the US in PJAM which which for those of you
[30:31] Speaker 5: who spend a lot of time in the US will
[30:33] Speaker 5: know that's sort of the northeastern part of the grid.
[30:36] Speaker 5: Seventy percent of the world's Internet runs through that network,
[30:41] Speaker 5: and you know, sixty seven million people interconnection queue you know,
[30:45] Speaker 5: years long and so long that PJM had to sort
[30:49] Speaker 5: of say, we can't look at these these applications as
[30:51] Speaker 5: quickly as we'd like to. And so that's what Tapestry
[30:55] Speaker 5: is working on now, is we're applying some of the
[30:57] Speaker 5: best technology that we can possibly sort of discover. And
[31:01] Speaker 5: by the way, that is changing every two weeks. I mean,
[31:05] Speaker 5: you really need to understand how quickly technology is moving
[31:10] Speaker 5: right now.
[31:10] Speaker 2: It is.
[31:11] Speaker 5: It is breakneck and it is amazing if we use
[31:13] Speaker 5: it properly, and we're we're moving as quickly as we
[31:16] Speaker 5: can to apply that technology to this incredible challenge of
[31:19] Speaker 5: looking at interconnection cues and transmission planning in pj M
[31:23] Speaker 5: and for us on the on the on the kind
[31:25] Speaker 5: of who who pays for this? We think about we
[31:28] Speaker 5: have got to build more, of course, you know, we
[31:30] Speaker 5: have load growth.
[31:31] Speaker 6: We have a lot to do and we need to
[31:32] Speaker 6: do it quickly, and we've got to use more of
[31:35] Speaker 6: what we have.
[31:35] Speaker 5: I mean, it's just really sort of basic first principles,
[31:38] Speaker 5: use more of what we have.
[31:39] Speaker 2: Uh.
[31:40] Speaker 6: And I always think, can I turn more nose into
[31:42] Speaker 6: a yes?
[31:42] Speaker 5: If someone says I want to connect to the grid
[31:44] Speaker 5: and the answer comes back, hopefully it's not years later.
[31:47] Speaker 5: Let's say we can do it in a couple of days,
[31:49] Speaker 5: and the answer comes back, No, it shouldn't be.
[31:51] Speaker 6: I don't know why it should be. Here are ten
[31:53] Speaker 6: things you can do to make it a yes. You
[31:55] Speaker 6: could use you know, these other technologies. And so that's
[31:57] Speaker 6: how we think about applying lessons.
[32:00] Speaker 2: Great At the risk of being the skunk at the picnic.
[32:04] Speaker 4: Here for a second, We've heard all morning about this
[32:07] Speaker 4: incredible demand growth for AI, and I'm struck by just
[32:10] Speaker 4: how quickly this conversation has.
[32:12] Speaker 2: Changed about sort of load growth, load.
[32:14] Speaker 4: Growth in the last eighteen months or so. What if
[32:17] Speaker 4: we wake up tomorrow and it turns out that, let's say,
[32:20] Speaker 4: the next Nvidia chip is unbelievably more efficient, or it
[32:24] Speaker 4: turns out there isn't as much demand for all these
[32:26] Speaker 4: AI services as we think, where does that leave all
[32:29] Speaker 4: of you in the transmission and systems game?
[32:32] Speaker 2: I mean, really quick, Michael, if you yes, so, first racket.
[32:36] Speaker 3: So we study that, we run scenarios on it, and
[32:40] Speaker 3: our lines are valuable to the system with or without
[32:45] Speaker 3: this big load growth. That's sort of the quick answer.
[32:49] Speaker 3: And then I think the national answer to that question
[32:52] Speaker 3: is it's the devil is in the details of who
[32:56] Speaker 3: is paying these incremental costs to the system. And if
[32:59] Speaker 3: it an AI provider and they're paying for the upgrades
[33:05] Speaker 3: and they don't show up, then thank you very much,
[33:07] Speaker 3: y'all we get some no free grid stuff.
[33:12] Speaker 2: Frank Send.
[33:13] Speaker 7: Yeah, So I would say that.
[33:17] Speaker 8: AI maybe there's fifty gigawatts of declared interconnection capacity that's
[33:23] Speaker 8: in the queue. There's another maybe sixty behind the meter.
[33:26] Speaker 8: It's a debatable. It's hard to know what's behind the
[33:28] Speaker 8: meta right now. Self supply, I just want to remind
[33:31] Speaker 8: folks that we have two point three terra watts of
[33:34] Speaker 8: generation in the queue right now, so twenty three hundred gigawats.
[33:38] Speaker 8: So we're gonna have plenty to do on the generation side,
[33:42] Speaker 8: even if AI doesn't connect at the levels that people
[33:46] Speaker 8: are thinking about. And then you've got the whole manufacturing sector,
[33:49] Speaker 8: you've got electrification more broadly, like there's a lot of
[33:51] Speaker 8: activity that's placing stress on the system, and people have
[33:54] Speaker 8: been talking about those stresses well before AI showed up.
[33:57] Speaker 8: We're really happy to be working with AI partners bring
[34:00] Speaker 8: an urgency to the equation that wasn't there before and
[34:04] Speaker 8: it's been great and we've helped a number of parties
[34:06] Speaker 8: in pgen elsewhere solve those solutions.
[34:09] Speaker 7: But it's not the only thing that's happening in this sector.
[34:13] Speaker 2: Lead the same for you.
[34:14] Speaker 4: Poles and wires are going to be needed, like just
[34:17] Speaker 4: the demand is going to rise, regardless.
[34:20] Speaker 9: I didn't hear that last part, but I think the
[34:22] Speaker 9: question was what happens if we become more efficient, do
[34:25] Speaker 9: we still need all this poles, wires, transmission? And I
[34:29] Speaker 9: think the short answer is absolutely. You know, just as
[34:32] Speaker 9: an example again, I'll lean on Texas Houston. While we've
[34:36] Speaker 9: seen such enormous growth in certain technology sectors advanced manufacturing,
[34:42] Speaker 9: I mean, we've seen growth here in a lot of
[34:44] Speaker 9: other segments that we're so diversified, whether it's gas and oil,
[34:48] Speaker 9: whether it's medical logistics, electrification, whether it's on the industrial
[34:54] Speaker 9: side or the ev side, transportational electrification. There's just so
[34:58] Speaker 9: much happening all the time. I mean, there's certainly enough
[35:03] Speaker 9: to go around.
[35:05] Speaker 2: Okage anything to add on that, No, I would just.
[35:07] Speaker 6: Say, let's think about what the point?
[35:10] Speaker 2: What is the point?
[35:10] Speaker 5: I mean if you were at this conference three years
[35:13] Speaker 5: ago we had been talking about bitcoin mining and blockchain.
[35:15] Speaker 6: Oh my gosh, is what's gonna happen with crypto? How's
[35:17] Speaker 6: the grid going to handle it?
[35:18] Speaker 7: Is it too big?
[35:19] Speaker 6: Then it was electric vehicles and now it's AI.
[35:21] Speaker 5: And yes, AI is a big low and growth, but
[35:24] Speaker 5: we need more electricity.
[35:26] Speaker 2: Period.
[35:26] Speaker 6: Let's get to solving it and figure out as we're
[35:29] Speaker 6: in it.
[35:30] Speaker 2: Please join me and thanking. This's excellent panel.
[35:33] Speaker 4: I think we troubled a lot of ground.
[35:44] Speaker 1: Today's episode of Switched On was produced by Cam Gray
[35:47] Speaker 1: with production assistance from Kamalas Shelling. Bloomberg n EF is
[35:51] Speaker 1: a service provided by Bloomberg Finance LP and its affiliates.
[35:54] Speaker 1: This recording does not constitute, nor should it be construed
[35:57] Speaker 1: as investment and vice investment recommendations, or a recommendation as
[36:01] Speaker 1: to an investment or other strategy. Bloomberg ANNIAFF should not
[36:03] Speaker 1: be considered as information sufficient upon which to base an
[36:07] Speaker 1: investment decision. Neither Bloomberg Finance LP, nor any of its
[36:10] Speaker 1: affiliates makes any representation or warranty as to the accuracy
[36:14] Speaker 1: or completeness of the information contained in this recording, and
[36:16] Speaker 1: any liability as a result of this recording is expressly
[36:19] Speaker 1: disclaimed

Shows are listed from Apple Podcasts' charts and directory, and Podcast Index where it is used. Every show and episode belongs to its publisher and plays from the publisher's own feed; this site hosts none of them.