Demand for electricity is soaring: how can our power systems keep up?

Divya Reddy, Manager, Energy Initiatives, World Economic Forum: There really is currently this moment for electricity.

Robin Pomeroy, host, Radio Davos: Welcome to Radio Davos, the podcast from the World Economic Forum that looks at the biggest challenges and how we might solve them. This week the global demand for electricity is rising fast, with wars in Ukraine and the Gulf squeezing fossil fuels.

Divya Reddy: And that then led to a pivot in the focus on electrification as an alternative to using fossil fuels in the energy system. And now the opportunity to use clean energy for electrification is also more available than it was any time before.

Robin Pomeroy: Add to that the rising demand from artificial intelligence.

Varun Sivaram, Founder and CEO, Emerald AI: AI data centre globally grew by 50% in their energy consumption in just over a year.

Divya Reddy: By 2030, they expect demand from data centres to double. That will still be just around 3% of total electricity consumption, but the pace and scale of it is, is quite remarkable.

Robin Pomeroy: We look at the pressure on electricity systems and speak to the head of a startup who says there is a solution to the data centre challenge.

Varun Sivaram: Think of the grid as a big, massive super highway. It's got eight lanes, right? And most of the year, it's quite empty. You could add a ninth lane to that highway. It'll take you 10 years, and it's very expensive, and everybody pays for it. Or that data centre can just reduce its consumption during that peak moment so you can use your eight lane super highway, which is available almost all of the year.

Robin Pomeroy: Follow Radio Davos wherever you get podcasts, or visit wef.ch/podcasts.

I’m Robin Pomeroy at the World Economic Forum, and asking how we can meet surging demand for electricity...

Varun Sivaram: You've avoided building that expensive ninth lane.

Robin Pomeroy: This is Radio Davos

Welcome to Radio Davos, where we're talking about energy, specifically the massive increase in demand for electricity and how we might meet that demand. Later in the episode, we'll hear from a tech startup that says it has a solution to the challenge posed by the rise of AI data centres.

But first, to set out the scale of the issue and to talk about the whole range of issues, I'm joined by my colleague Divya Reddy, Manager of Energy Initiatives at the World Economic Forum. Hi Divya, how are you?

Divya Reddy: Good, thanks. How are you, Robin?

Robin Pomeroy: Very well. Thank you. What do you do at the World Economic Forum?

Divya Reddy: I'm the manager specifically on the future of power systems initiative and we bring together stakeholders from across the electricity value chain to really decide what are the key priorities facing the electricity industry and arrive at some joint solutions for addressing them.

Robin Pomeroy: Electricity, the demand for electricity, whenever we do a radio Davos about energy, at the moment we can talk about oil and gas, we can talk about all kinds of things, but this massive increase in demand for electricity, when did that really start and what's driving it?

Divya Reddy: So it's been over the last 10 years, and it's a combination of factors, but it really is currently this moment for electricity. And part of the more recent driver for this is the number of energy crises that we faced.

First, it was the Russia-Ukraine situation that resulted in gas supply shortages. And then more recently, the Straits of Hormuz being closed off, again, resulting in oil and gas shortages.

And that then led to a pivot in the focus on electrification as an alternative to using fossil fuels in the energy system. And now the opportunity to use clean energy for electrification is also more available than it was any time before.

Robin Pomeroy: So it's partly that transition to electrification for things like cars and in industry as well.

Divya Reddy: Exactly. So it's across the board. So the three sort of end uses would be transportation, it would be industry, and it would buildings as well.

Robin Pomeroy: Also though, add to that, in the last what, three years, the rise of artificial intelligence as a thing we're all using now, which probably we didn't realise three years ago, the massive energy demand that would create. Tell us something about that.

Divya Reddy: Yeah, exactly. So in addition to just general electrification of end uses, there is this massive new entry of data centres, specifically AI data centres that is coming into the system.

Now, it's still a relatively small part of the overall electricity picture. And it probably will remain so, but it's the pace and scale of the change that's really having a dramatic impact over the last few years.

So just to put some numbers on this, the IEA estimates that last year alone, there was a 17% increase in data centre electricity demand. And that's compared to just 3% for overall electricity demand growth. And by 2030, they expect demand from data centres to, to double that will still be just around 3% of total electricity consumption, but the pace and scale of it is, is quite remarkable.

And AI data centres in particular draw around 10% more electricity than even standard data centres. And they also have a very kind of volatile and more unpredictable demand profile than traditional loads do.

So it's sort of a combination of factors. It's the pace and scale of growth. It's also the nature of the demand and it's also location because. What happens is that when data centres come into the electricity system, they typically connect on what we call firm connections, which means that they basically are guaranteed service at any time. And so that requires that this network needs to accommodate them at all times and that it can exacerbate peaks in supply and exacerbate tightness that then leads to higher costs that then need to be paid by all consumers. So there's also a locational issue if they're entering into the system at congested areas.

Robin Pomeroy: Later in this episode, we're going to hear an interview I did recently with the head of a startup company who says he has a solution, at least in part, to some of this, and it does go to that thing you're saying about peak demand, which is a problem with electricity in general, right? Maybe we can talk a little bit about that. We'll get back to data centres in a minute, but in general, you know, we've got wind and solar energy, which over the last, what, 10 years, a bit more maybe, have suddenly become this very cheap way of generating electricity. But what they don't have is that kind of constant, you can't, if the sun's not shining and the wind's not blowing, you can just switch it on and you need that baseline power. That's a real challenge, isn't it, for this massive increase in demand.

Divya Reddy: Yeah, exactly. So essentially, the way that electricity systems have traditionally operated and been planned for is that you have these steady sources of demand and steady sources of supply. And it's sort of been this traditional model of balancing the electricity system at all times.

With the advent of more variable sources of generation like wind and solar, you have more volatility in the supply profile. And now with the advent of things like data centres, you also have this more volatile demand coming into the picture.

Now there are a lot of tools that can address these. So there are now increasingly options for flexibility solutions that can help address these volatile supply and demand profiles, but they need to actually be enabled to participate in the system the way that they are able to.

Robin Pomeroy: And where's the low-hanging fruit? What are the things we can do quickly to improve our electricity systems?

Divya Reddy: So given the pace and scale of what the electricity system needs and that it does need to grow, we also do need to enable the building of new infrastructure. But more immediately, there is a lot of infrastructure currently that is underutilised. And so the low hanging fruit really is to enable the optimisation of existing infrastructure and assets that we already have. And there are a lot of tools, particularly flexibility tools, but there are also digital tools, there are AI tools that can support this. So those are the kinds of solutions that really are the low-hanging fruit.

Robin Pomeroy: Where can people get more information? You publish useful reports. Where can people go? What resources have you got?

Divya Reddy: So we currently have a tool on our website, which is called the Innovation Playbook for Future Power Systems. And it includes, it's a sort of database of over a hundred solutions backed by case studies. And it really is aggregating solutions that are already in the market that then can be deployed and scaled to address these challenges of modern power systems. And it covers both technology type of solutions, but also things like financial models, business solutions, policy and regulatory settings that are all part of what we call solutions.

We also have an upcoming white paper that will be released in early November and that is really looking at three focus areas that are kind of most pressing for today's power systems.

The first is optimising existing infrastructure. The second is how to quickly enable and finance new infrastructure. And the third is how to connect large loads such as data centres.

Robin Pomeroy: Let's get back to data centres then. Just remind us again what the problem is here and of the work you're doing on that and then we'll introduce this interview.

Divya Reddy: Yeah, so the problem, again, is that the pace and scale of demand is greater than the electricity system in many locations can accommodate. And so it's creating this tension. It's requiring network expansions. It's increasing costs.

There are some solutions that can help alleviate this challenge. One is from the demand side that data centres are actually able to operate more flexibly.

So they can compute at different periods, they can shift jobs to different hours, they can cut their computing in certain peak moments, and they can also redirect processing to a different location.

And all these have significant benefits to the system. They can free up more network capacity. They can lower costs or unlock savings because it avoids peak charges. It avoids new infrastructure builds and it sort of optimises the existing grid more efficiently. And they can also connect new sources of demand faster. Currently we hear a lot about connection queues where large demand is just not able to connect to the existing grids. So it alleviates all these types of problems.

And then on the supply side, they can also offer their own onsite storage as well as onsite generation that also limits the amount of electricity they draw from the grid.

Robin Pomeroy: So that brings us to this interview that I did at an event that I believe you were involved in here at the World Economic Forum. His name is Varun Sivaram. What was he doing in Geneva?

Divya Reddy: The World Economic Forum hosted an energy forum at our headquarters in Geneva in September. Varun Sivaram served as one of the co-chairs for this event.

And what it did is it brought together stakeholders from across the energy spectrum to discuss challenges with current energy systems, to discuss joint solutions, and to kind of facilitate partnerships across the energy space that can take forward these types of solutions.

Robin Pomeroy: He's a very interesting person. He's the chief executive officer of Emerald AI, and that's what he's going to talk about. But if you look at his CV, he's had all kinds of interesting jobs in industry and at the policy level in the United States, in India. Let's get on, let's hear this interview with Varun Sivaram.

Varun Sivaram: I'm Varun Sivaram, CEO and founder of Emerald AI.

Robin Pomeroy: What is Emerald AI?

Varun Sivaram: Emerald AI transforms AI data centres into flexible grid allies, good citizens to communities and power grids alike. With Emerald AI's technology, we can get far more data centres connected to today's existing power grits. We can keep rates low, energy bills low for communities, and we can keep the grid stable and reliable. It's a win-win-win, but the data centres have to become good grid citizens, and that's what our technology enables them to do.

Robin Pomeroy: Let's remind everyone of the challenge here. AI, generative AI, four years ago most people weren't even aware it was a thing. Now most people who have a computer are using it. We weren't initially aware, some of us, of the energy requirements of AI. Give us some idea of how big a problem that is.

Varun Sivaram: The energy consumption of AI data centres is massive. Now, it hasn't always been massive. The International Energy Agency reports that historically data centres represented 1% or less of global electricity consumption. Now, that figure is set to hit 3% by 2030. Still doesn't sound like a big number, but it's important to note the growth rate.

AI data centre globally grew by 50% in their energy consumption just over a year. Nowhere is this more pronounced than in the United States, my country, where the US accounts for 45% or so of global data centre power demand, and data centres are growing rapidly.

And the big problem that we're running into in the US is that we don't have enough capacity on the electricity grid to plug in all the data centres we want.

Look, by 2030, the United States is gonna want to plug-in some 70 gigawatts of data centres. But it's only really going to be able to plug in half of that number, and that leaves a large unmet demand in order to power innovation, life-saving cures for diseases, and all manner of other economic productivity enhancements.

Data centres have hit the energy barrier in the U.S. It's not true in China, where in China there's plenty of energy, 400 gigawatts of available spare peak capacity in 2030. Their limit is chips. But in the United States and in most of the rest of the world, it's energy that's gonna constrain the growth of AI.

Robin Pomeroy: So, what's the answer? How do you solve the problem?

Varun Sivaram: That's why we founded Emerald AI. Emerald today is a unicorn technology company. We're proud to be a WEF technology pioneer. And we're supported by 12 Fortune 500 global companies, such as Nvidia. So Nvidia is not only our partner and investor, they've set the reference architecture for the power flexible good grid citizen data centre of the future.

And Emerald software powers this. Our solution is, to make the data centre capable of reducing its power use every so often during moments when the grid is stressed out. If you can do that, you can actually unlock a hundred gigawatts of power on today's electricity grid in the US and much more around the world. That power is just not used today. It sounds counterintuitive. Let me explain why.

Think of the grid as a big, massive super highway. It's got eight lanes, right? And most of the year, it's quite empty. There aren't very many cars on that road. But two times a month, just two times month, that highway faces peak rush hour, where the cars are clogging up that road, that's like a hot summer day when everybody's using their air conditioners and the grid is really stressed out. But most of the year, the grid runs at 50% capacity utilisation.

Now imagine you want to connect a new data centre. That data centre during those peak moments can't be served by that grid that's facing peak rush hours. And so what you could do is you could add a ninth lane to that highway. It'll take you 10 years, and it's very expensive, and everybody pays for it. Or that data centre can just reduce its consumption during that peak moment. So you can use your eight lane super highway, which is available almost all of the year, to power that flexible, friendly data centre.

And in so doing, the revenue that it generates helps the local communities pay their energy bills. You've avoided building that expensive ninth lane. You've broadened revenue for the communities. And the best thing is you also help them prevent blackouts because when the grid is really stressed out, hot summer day in Arizona, we showcased with our partners NVIDIA and Oracle and others that that data centre is actually able to reduce its consumption.

And that's why we're really excited that we've announced this new leading coalition and alliance of the world's leading companies committed to flexible AI data centres that are good grid citizens. We call it the AI Energy Management Alliance.

Robin Pomeroy: Walk us through how it works then. It sounds fairly simple. If you're a big consumer of electricity, somehow you work out now there's a peak demand from everyone else. I reduce my consumption. Is it as simple as that? That's my first question. I'll ask you a second one. Does that mean then those data centres will be performing at less, doing things less quickly, will users of those AIs realise, oh okay, it's a hot day in Arizona, that's why chat GPT's taking so long to reply.

Varun Sivaram: How is it that a data centre can moderate its electricity use in the moment during those peak rush hour hours, you know, three hours during the afternoon in Phoenix on a hot summer day? Or here's another example, Robin, that you might appreciate. Turns out that at halftime of the soccer game, the football game in England, everybody goes into their kitchen and turns on their tea kettle. And the tea kettle spike puts a gigawatt of load onto the grid. That's a moment of peak demand on the UK grid.

So in both of these cases, in Phoenix on a hot day. In the UK, at halftime of the soccer, Emerald has worked with real commercial data centres to reduce their power consumption in those moments.

How does it work? There's a few ways you can do this. One, you can slow down some of the computing that's happening on site. There may be many AI jobs that are running. Some of them can be slowed down without the customer really noticing or minding.

These could be a fine-tuning run, for example, of an AI model where that run doesn't need to complete right this moment, and you can delay it by half an hour or two hours.

Another example is batch inference process or a range of other batch processes.

And finally, there are things that you can move from one location to another location. Let's say you're talking to a chat bot, chat GPT and you want your answer right now. Well, I can run it equally well in Virginia or Chicago. So on the coldest day of the year, we showcased that when Virginia's grid had a peak demand moment, everybody turned on their heaters, we moved the load over to Chicago, which as it turned out was better equipped to handle those cold winter days.

Now, using this technology, Emerald's conductor software platform, we can decide how best to protect the AI workloads that need to keep running right now, those critical workloads, how to slow down other ones, and how to move the final workloads to other locations, all in a way that the AI customers often don't even notice while the grid gets the precise response it needs.

Hot summer day, the grid says, I need help, and the data centre responds precisely based on what the grid needed, let's say a 25% reduction in power for those three hours. So that's how it works, Robin.

Robin Pomeroy: There's been a lot, particularly in the US, a lot of local opposition to data centres. If you've solved the problem, no one should be worrying about this anymore. I wonder, are there blockages still in the way of making this happen? Where's the future?

Varun Sivaram: Yeah, absolutely. We need three things to happen in order for us to solve this problem.

First, we need the proof. Second, we the coalition. And third, we we need compact, or the bargain. Let me walk through each of these points.

First, we need the Proof. Emerald with NVIDIA and our other partners like EPRI have already done these six demonstrations around the world. London. Iceland just happened. Phoenix, Virginia, Chicago, Portland. So we've showcased at real data centres that you can do this.

And at Davos, we're very excited that we will announce the world's first 100 megawatt, multi-billion dollar power-flexible AI data centre. NVIDIA and Digital Realty are building it, and Emerald is their partner in making it a flexible data centre that can listen to the grid and react in a way that the grid needs.

So the proof is seeing real data centre's do this, and we are well on our way. That's point number one.

Point number two is the coalition, or the alliance. I'm so proud that Google, NVIDIA, and Emerald AI founded the AI Energy Management Alliance. We have 20 launch partners, Anthropic, for example, Analogue Devices, another Fortune 500 semiconductor company. Or on the energy side, Constellation, NRG, AES, Fluence, a range of large publicly traded Fortune 500 companies. And with this cohort, this coalition, we're going to policymakers at the American federal level, at the state level, and eventually we'll make this global with the collaboration of the World Economic Forum, to say there is a technology that if you bring it onto your grid faster will bring great benefits.

And that brings me to my third point, the compact or the regulatory bargain. What's the grand bargain here? The grand bargain is if you have a data centre willing to be flexible, you should reward it. The reward is faster and larger connections to the power grid, because that's a friendly data centre. That's a good grid citizen. It's going to lower your rates. It's going to bring on more clean energy and stabilise the grid. It's not a data centre that's going to cause the problems that communities are correctly worrying about. They're worried that their electricity rates are increasing. They're worry that these data centres could destabilise their grid.

If you want communities to support a data centre, that data centre needs to demonstrate that it's being a good citizen. In a way that they may not traditionally have wanted to do, right? Traditionally, data centres have wanted to say, give me 24-7 access to power. I never want to think about the power grid. Being a good grid citizen means you have to think about power grid, you have to be responsive. When your neighbours need that power, if you're willing to step forward and say, okay, I'll take a sacrifice in this moment, that's the kind of good citizenship that should earn you a fast lane onto the grid.

When we talk to our partners, the Utility National Grid, our investor and partner in the United Kingdom. Their chief strategy officer, when he heard about Emerald, he said, my goodness, if data centres could do this, I wouldn't be worried about connecting them and I wouldn't be making them wait years on end to connect them. I'd want to bend over backwards to connect to them as quickly as possible because they could make my system work better. I would compete with other utilities to connect with them faster.

That's what I want, a virtuous race to the top where communities and utilities compete with each other to bring the friendly, good grid citizen data centres. And that virtuous cycle means that communities benefit truly when the data centre enters and enables them to bring on all kinds of other loads, like hospitals and advanced manufacturing. That's the virtuous cycle.

Robin Pomeroy: So when you're talking to someone like the National Grid, and they're very excited about it, is that also because not only are you able to say we can put on a data centre and it won't just drag everyone else's electricity away from them and put up prices, but there's also some kind of stabilising mechanism because all power grids, they'll have inputs. There's suddenly all the wind turbines are going faster than we can use that electricity. Is there that balancing thing happening?

Varun Sivaram: Yeah, that's such a good question, Robin, and there are three important reasons, I'm sorry, I used to be a consultant, I talk in threes. There are three reasons why utilities absolutely love these flexible data centres and why their regulators and their political leaders, they love these data centres.

The first is the one you said, Robin. If a flexible data centre comes onto the grid and it can respond to whatever the grid needs or when it's in stress, it can stabilise the system and avoid a blackout. That's important whether the tea kettles are going on at halftime in the UK. Or it's a hot summer day or lightning has struck and taken out of substation, the data centre can respond and help to bring the grid back into balance. So reliability and energy security is one.

The second reason is because data centres that are flexible help you avoid building out very expensive infrastructure to meet that rare peak. They help you avoiding building that ninth lane on the eight lane superhighway that you basically will never use except for two times a month on rush hour, right?

Now, to be clear, utilities do like to build stuff. They get paid when they build stuff, but the good news now is that the data centres come in and say, I'm willing to be flexible. I'll use your grid when no one else is using it during the rest of the year when the eight lane superhighway is pretty empty. And the money that I bring in, you can now use to go build out infrastructure to serve even more load. You can bring the manufacturing plant, the factory, you can bring that hospital, you can serve your neighbourhoods and everybody can put air conditioners in their homes because there are heat waves in the UK, et cetera, et cetera. It's the data centres that are helping you fund this instead of the local community's energy bills. So your grandma's bill doesn't go up, but the data centre comes in and helps you to fund all of these upgrades. So that's the second reason.

And the third reason is that it avoids the utility being disintermediated. That's a long and painful word. What does that mean? The utility likes to have a relationship with its customer, right? It likes its customer to be part of the public power grid. The public power grid works because so many different diverse customers are all on it and we can socialise the costs of that infrastructure.

What's about to happen though, Robin? Is data centres are going to say, you know what, I'm fed up with this. I don't wanna wait 12 years to connect to this grid. I'm going off grid. I'm building my own little utility by myself in the middle of Texas, or I'm actually going into space.

Did you hear Elon Musk has a $2 trillion company called SpaceX that can put me in space? Let me do that. If that happens, utilities in the public power grid will shrivel, because American, European, and global GDP will move toward AI and AI productivity. And if the data centres are not on the public power grid and their revenues aren't supporting local communities, those local communities will be left with larger and larger bills.

Utilities don't want somebody else to power those data centres. They'd like to power them themselves. And they can do it if the datacenters are flexible, friendly, community positive data centres, that's why we wanna make these good grid citizen data centres a reality.

Robin Pomeroy: Let's be clear, then. A data centre company, a hyperscaler, could go to a town somewhere and say, right, we're about to put spades in the ground. We're going to build this massive thing here. And I know what you're going say, everyone. You're worried about the things we're talking about. You no longer need to worry. You'll still get the electricity you need. Your prices won't go up. Can they already say that today?

Varun Sivaram: They can't say it unless they do this. Unless they are flexible, they can't in good conscience go to a community and say I'm not going to raise your power bills.

Even if they say I am going to put spades down, I'm going to build this data centre, and I am not even going to connect to your grid, I am just going to a big gas plant next to my data centre and power it, there are various knock-on effects.

Suddenly you've raised the cost of gas turbine components for the public power grid or the cost to electrical switch gear implicitly. Suddenly, there's a chance that the emissions, the noxious fumes from your gas plant near this community are going to cause air pollution.

Unless you are willing to connect to that public-powered grid and be flexible and respond to the community's needs, you can't in good conscience say, I'm going to make sure that your rates aren't increased by my data centre coming into your community.

!!!1227

And to your point, Robin, this is just a conversation that's going to happen not just in the United States. It's going happen all over the world. And here's what I want to happen. What I want to happen is the data centre says, look, I want to be a good grid citizen. I want be a citizen. I'm going to come to town. I'll pay my property taxes. I'm even going to fund some improvements to parks and schools. I'm gonna be a flexible power consumer that will help your bills and it will help your reliability.

And the community says, well, we would love this because it brings lots of benefits. We can also fund the manufacturing, the hospitals, et cetera. But there's still some of us who are worried about AI generally.

They're worried AI might take your job. They're worried that. AI might cause safety risks, they're worried a bioweapon might come out of an AI lab one day. And then they'll weigh their risks, their pros and cons, and they'll say, well, this particular asset could be really positive to my community, and I have to weigh that against what my general feelings are toward AI more broadly.

And I hope communities around the world will even-handedly assess this and determine whether they want to invite the data centre. But where I don't want the argument to go is, Well, I should just reject this data centre out of hand because it's going to raise my power bills and it's not going to be a good citizen.

I think there's so much room now for the data centres to change that narrative and to prove, in fact, quantitatively, I will be an excellent citizen and you will have no economic opportunity cost. In fact, you will be strictly better off. If I come to your town, everything about your economy will be better if I come town than if I don't.

Robin Pomeroy: Tell us about this alliance you mentioned earlier with these big companies. Why have you formed an alliance? What's the difference between just serving them as customers?

Varun Sivaram: Exactly. Well, so if you remember, the three components that I'm passionate about, first is proving that this is real, that data centres can be flexible. Second is bringing the Alliance together. And third is getting the regulatory grand bargain.

The Alliance, that second component, I think is critical for that bargain to happen. The alliance brings together leading players from across the AI and energy value chain. As I mentioned, you have leaders. We are the founding members, Emerald, Google, and NVIDIA, but also leaders like Anthropic, one of the world's leading AI companies, analogue devices, but also on the energy side, leading energy producers and utilities, National Grid I mentioned earlier, NRG, Constellation, etc. And they've all come together and they've said, you know what, we're going to present a unified voice.

Because governors, regulators, utilities, they may not know this is possible. They may not know about this grand bargain. So we'll tell them, here's the grand bargain, if the data centre's willing to do these things, modulate your power use during rush hour. And we don't care how you do it. Right? You might decide you've got a battery on site and that's how you're going to run the battery during peak grid rush hour. You'll run the battery or you'll slow down your compute, which is what Emerald helps you do.

However you do this, we don't mind. We will just give you faster and larger power access if you can enforceably and verifiably be a flexible power user. That grand bargain sounds like a very wonky technical thing, Robin. You know, we're talking about ramping up and down your megawatts. It's very wonky, but it couldn't be more central to the social compact of AI and communities.

If data centres can do this one party trick, control their power, data centres can be net positive for communities, and the growth of AI no longer needs to see a trade-off, Robin, between AI innovation, energy affordability, and energy reliability. It's that rare case, and I'm a former policymaker. It's the rare case. Where you've got a win, win, win with basically no trade-off.

Robin Pomeroy: What about the timing of this argument? Because maybe a year ago, it wasn't such a hot issue, data centres. And now, I've certainly seen in US media, it's a very big issue. And you'll know very well, once people have made up their mind about something, it's going to be very hard to change. So if you go into a town and say, yeah, we're going to build this huge data centre. Don't worry. It's fine. That problem you thought was a problem, we fixed it now. Either your timing is amazing and it's perfect, because this is a problem so many people would like to solve. Or you've kind of missed it, and already the public is against these awful data centres. Which do you think it is?

Varun Sivaram: Robin, the point you're making here is that we're trying to close the barn door after the hors has bolted, right? And let me tell you, I've got a lot of faith in the American people that present them with new facts and new data, in this case, the opportunity to have economic development and prosperity and AI innovation alongside protecting their bills and protecting their energy reliability and security. I think a lot communities will take you up on that.

And so you're right. It is a polarised political topic today. But just, just you wait. We're going to get to January 2027 to Davos, and you'll see the announcement that Jensen Huang at NVIDIA has already previewed that NVIDIA, Emerald AI, and Digital Realty are launching this world's first power-flexible AI data centre, or AI factory. You know, it's important, of course, because it's 100 megawatts, that's meaningful. But it's really important because that's the blueprint. That's the first time you're going to see a purpose-built AI data centre that can perfectly respond to the local grids. Dominion Energy is the utility. PJM is the grid operator. They're all our partners with EPRI. That's the first time you'll see one built to the specification, and that specification is then public.

And the next thousand AI factories can be built to that DSX reference architecture. That's what I'm really excited about, because we have real proof. And yes, we will now need to take this door to door and convince communities that this is reality and that they're priors. Which are reasonable, right? The American people have had reasonable reactions. Their priors are that data centres might not be responsive, flexible entities. We've gotta change those priors with proof and with data. I have a lot of faith in the American people.

Robin Pomeroy: That was Varun Sivaram, the CEO of Emerald AI, talking to me a few weeks ago at this Energy Forum, which Divya, you were involved in organising here. If people want to find out more about your work, remind us where they can go and what was that report that's coming out in a couple of weeks' time from now?

Divya Reddy: So everything about our initiative can be found on the World Economic Forum's website. If you look under the Future Power Systems Initiative, you can find the Innovation Playbook for Future Power System there. The upcoming white paper on enabling the electrification era will also be published there in the first week of November.

Robin Pomeroy: Have you got a busy Davos ahead? We're already thinking about the Annual Meeting in January, 2027.

Divya Reddy: Indeed. And given the attention that electricity is facing right now, they're discussing at COP to set a global electrification target. So I think the topic of electricity will be very front and centre. The climate conference that's happening in Turkey this year. So heading into Davos, I think the topic of electricity will definitely be front and centre and we will be having a number of sessions that really look at different aspects of it.

Robin Pomeroy: So there's loads on that, on your part of the world Economic Forum website. Radio Davos has done many episodes about energy and electricity. I'll be putting links to those in the show notes. Divya Reddy, you're manager of energy initiatives. Thanks very much for joining us on Radio Davos.

Divya Reddy: Thank you, it was very nice to be here.

Robin Pomeroy: Radio Davos is available every week wherever you get podcasts, or you can find it at wef.ch slash podcast, where you can our other two weekly podcasts, Meet the Leader, Conversations with some of the most interesting people in the world about how they lead their organisations, and Agenda Dialogues, which is the full audio from sessions recorded here at the World Economic Forum.

Radio Davos will be back next week. Thanks very much to you for listening, and goodbye.