In this episode, Duncan Angove, CEO at Blue Yonder, shares with us what he calls the age of uncertainty when it comes to supply chain management software, what it looks like to utilize AI to resolve risk, and helps us to understand everything that goes into building the world’s largest supply chain on the data cloud.
In this episode, Duncan Angove, CEO at Blue Yonder, shares with us what he calls the age of uncertainty when it comes to supply chain management software, what it looks like to utilize AI to resolve risk, and helps us to understand everything that goes into building the world’s largest supply chain on the data cloud.
---
BUILD 2024
Dive deep into the world of apps and generative AI from Snowflake BUILD. Catch up on the latest announcements focused on building apps, data pipelines, and machine learning workflows in the age of LLMs.
Learn more and register for a BUILD event in your region.
[00:00:00] Producer: Hello and welcome to the Data Cloud Podcast. This episode features an interview with Duncan Angove, CEO at Blue Yonder. In this episode, Duncan shares with us what he calls the age of uncertainty when it comes to supply chain management and software, utilizing AI to resolve risk and breaks down all of the work that goes into building the world's largest supply chain on the data cloud.
So please enjoy this interview between Duncan Angove and your host, Steve Hamm.
[00:00:32] Steve Hamm: Duncan, it's great to have you on the podcast today.
[00:00:35] Duncan Angove: I see. It's great to see you. Thank you for having me.
[00:00:38] Steve Hamm: Now, Blue Yonder is a leader in supply chain management software, but some of our listeners are probably not familiar with the company, especially because of the name change a couple of years ago.
So please describe Blue Yonder's business and a bit about its history. And also why did it change its name?
[00:00:56] Duncan Angove: Okay. Yeah. So we're the largest supply chain software company in the world, and we deliver digital supply chain transformation for like largest retailers, manufacturers and logistics providers.
We've got 3, 000 customers and they're literally the largest brands in the world. So whenever you see a truck or you have a product in a grocery store or a car is coming off the manufacturing line. It's in general, it's been orchestrated and it's there because Blue Yonder software put it there. So that's, that's basically what we do.
And, you know, we deliver business outcomes around customer data and multiple growth and more resilient supply chains, which is very important in today's world and ultimately sustainability. So that's what we do. The company changed its name because we felt like our old name, JDA, didn't really reflect where the business was going.
We had acquired a company called Luanda, ironically in Germany, it was sort of a, it was, it was a cloud-native, authentic AI company, and they'd been doing AI for 10 years. And, you know, when, when they decided to, to basically look and they had some weird German acronym, and when they decided to change the name of it, they hired a marketing firm.
And I remember the professor that founded the firm, he was up on an island just north of Hamburg, and he was stood on the beach and he got this email from his, from this marketing firm, and it gave him the choice of sort of three names. And he looked out and it was like this beautiful blue sky, beautiful dark blue ocean and white sand.
And the first name said Blue Yonder and he looked out at that and he, he sent the picture to his executive team and he put Blue Yonder on it and sent it to them. And he said to me, what it evoked for him is that if you're a pilot and you're flying really, really high, you can see the curvature of the earth, right?
And what you can actually see is you can see beyond the horizon. You can see what's coming. And that's fundamentally what they did. They used AI to predict and see what people couldn't. And that's why the sort of name you have sort of Blue Yonder. Yeah.
[00:02:49] Steve Hamm: You've been CEO at Blue Yonder for a couple of years now.
So tell us a little bit about your career path and what your business goals are for Blue Yonder.
[00:02:58] Duncan Angove: I've always been in the supply chain software and come from an engineering background, and that's very much the philosophy that I brought to Blue Yonder. We're very much an engineering centered company. And what that means, we have a belief that if you get product right, a lot of other things will take care of itself.
And that's reflected in that, you know, we're spending in the next three years a billion dollars in R& D. So that's very much our core philosophy. And, you know, sort of like the five whys. When you see customer problems or you see problems that Blue Yonder might be having, if you ask the five whys, you ultimately end up with something you can solve in a factory in engineering.
And that's very much a core tenant of ours. And the second thing would be innovation. Innovation is the engine that drives the company. You know, last year we filed over 130 patents, mostly around AI. So it's really something we do. It, we, we challenge conventional thinking in everything we do. We look for step changes in, in values for customers, you know.
The dirty secret of enterprise software is that the R and R and D doesn't exist. Oh, really? And it's actually research that actually drives fundamental breakthroughs. And that's very much we do. So that's sort of core to the company.
[00:04:06] Steve Hamm: Yeah. So you, you're in the supply chain space. Do you actually occupy every major position in your portfolio in that space?
Or there are, are there some gaps?
[00:04:18] Duncan Angove: We're arguably the broadest, most complete end to end solution in supply chain management. You know, it's everything from, you know, the assortment you put in front of the consumer to how you forecast demand, where, when, what, where it's going to take place, how you fulfill it, all the inventory management, you know, the production line in a manufacturing plant, you know, how you orchestrate thousands of trucks for, you know, a CPG manufacturer or a large logistic company.
All the way to a warehouse operation. So we run the gamut of it. You know, that's one of the reasons why a lot of large companies come to us because they want an integrated, cohesive set of applications that they're applying to work with. Called the supply chain, right? You actually, we want the chain to be linked.
So I've been in supply chain software my whole life, right? You know, I'm from England, but the U. S. has butchered my accent. So I sound Australian or South African. But My first job at uni in London was actually implementing a warehouse management system, which is basically, you know, we have one of Blue Yonder.
And then from there, I went on into engineering and wrote, literally wrote forecasting and replenishment fulfillment systems from the ground up. So I've always been on sort of the customer value side, which gives you a lot of empathy. I've applied thinking around how you do software and then on the engineering side.
And that's, you know, it's very rare to have an engineering CEO and software. Normally come, they only come finance or, or IT or, you know, or sales. And so we're, we're an engineering led company.
[00:05:47] Steve Hamm: Duncan, in the wake of COVID global supply chains. And supply chain management software have been tested like never before. You call it an age of uncertainty. What kinds of uncertainty are you referring to, and what can we do about it?
[00:06:04] Duncan Angove: It's a great question, Steve. It represents a generational and structural shift in how supply chains Are being tested today and it's driven by both the demand side and the supply side in terms of what's creating this uncertainty on the, on the demand side over the last 20 years, we've seen a massive shift in consumer behavior, right?
That being driven by the rapid growth and adoption of digital technology. So it's e commerce, personalization, you know, on demand services, all of those things, right? Which are fundamentally bad e commerce supply chains weren't designed to efficiently ship. Like single parcels in the last mile to someone's house.
Right, right, right. So that demand side has caused a lot of structural shift in it. But the bigger, the bigger disruptions have occurred on the supply side, you know, and if you think about it over the last 20, 30 years, world supply chains were powered by, you know, using China as the factory for the world and a labor arbitrage play and this notion of globalization and just in time, right?
And what the pandemic exposed and all of the regional. Friction we have with China is that, that, that, that's not going to work in, in, you know, going forward. So it's had supply chain executives step back and say, how do I, how do I solve for this? And, you know, when you, JP Morgan did a study of the S& P 1500 and what they found over the last two years.
Is the working capital gone up by 40%? Right, so that's over 500 billion dollars of fundamentally inventory people have added to their balance sheet to try and be more resilient in response to a supply shock or a demand surprise. And our perspective is that's not the answer to resiliency. The answer to resiliency is software, right?
But a new kind breed of supply chain software in our world is fundamentally enabled by Snowflake.
[00:07:51] Steve Hamm: I understand that you're building your suite of cloud applications on top of the Snowflake AI data cloud. Why did you pick Snowflake as your strategic partner?
[00:08:02] Duncan Angove: So, you know, if you think about all the challenges I talked about earlier around the age of uncertainty and the need for a new paradigm, You know, we're very, very fortunate that there's been the emergence of sort of revolutionary technology over the last few years, right?
You know, the rise of artificial intelligence and not just gen AI, but predictive AI and machine learning and all of that. The rise of cheap and abundant robots, right? And then lastly, the rise of the data cloud with Snowflake. And when you take those three things together, It's basically a new canvas on top of which to reimagine how supply chain software should work to enable this new world.
How do you deliver resiliency and sustainability and profit without trading them off? And, you know, I talked earlier about the fact that supply chain applications are very, very siloed, right? You have different apologies across the landscape and they're all sitting on data silos. They end up moving data around and doing transformations.
There's a lot of latency. There's a lot of inefficiency and because the volumes are so enormous here, you end up with a lot of batch, right? And companies are not as agile as they should be. And that's the fundamental thing that Snowflake allowed us to shatter that paradigm by almost creating a single database for the world supply chain that solved that mass scale.
[00:09:19] Steve Hamm: Right. Because you talked about the importance of integrating all of your applications. Well, well, perhaps the most important thing. It's having them all draw from the same database and that's the Snowflake cloud, correct? Yeah, correct. And you
[00:09:32] Duncan Angove: know, yeah, yeah, exactly right. And you think about the premise of Snowflake around following this challenge of data gravity, you know, where data gets so large and so diverse, it becomes inefficient to move it around.
Right. So what you should be doing is moving the apps to the data and that's exactly the architecture that we built. Very good.
[00:09:52] Steve Hamm: So at Blue Yonder, you're mashing up a lot of different powerful software, your cloud applications, the Snowflake AI data cloud, large language models, generative AI, and as I understand it, You're using relational AI as knowledge graph management system, which is something I just learned about a few months ago.
So, how do these technology pieces work together to provide supply chain management capabilities that were not previously available?
[00:10:20] Duncan Angove: Yeah, and they're not previously available, the key here, right? You know, It truly is revolutionary technology that allows you to shatter constraints that have held supply chain management back.
You know, the idea that you have to have different data silos and different types of applications, facts to do planning or warehousing or what have you. And that's fundamentally the big, one of the biggest things that Snowflake did. If you think about their mandate, it was let's revolutionize analytics, right?
Let's revolutionize collaboration and let's revolutionize enterprise applications. And, you know, we were the first enterprise software company To write our application natively on top of Snowflake. And we do a ton of work with them. You know, you, you just look at our machine learning workloads. We run 10 billion predictions a day on the data cloud.
I mean, it's staggering 10 billion a day, and that wouldn't be possible in a normal architecture, right? You need the. The way Snowflake is built with this sort of the in infinitely elastic compute, separating storage from compute, you know, things like zero complete clonings, which allows us to run mass simulations.
All of that, when you combine it together, is what is allows us to do what we do, which is let's kill batch. You know, let's not have an hourglass, you know, when it's running a compute in the user interface, right? Let's get rid of all the latent that you have by moving data around. And if you think about supply chains from a business perspective.
All of the metrics in this industry are oriented around time and speed, right? It's just in time. It's, you know, order, you know, on time in full, how fast are your turns? It's all, they're all about speed. So what we do is we leverage Snowflake in that application technology stack you talked about. And we look to eradicate latency everywhere, right?
How can we build faster software rails in the world for supply chain management and allow our customers to get inside the competitive sort of, you know, decide, you know, sense, decide, and absolute. That's fundamentally what it allows us to do.
[00:12:16] Steve Hamm: So your, your customers can be very responsive. And it also kind of reduces their use of resources and, and capital for those resources.
[00:12:27] Duncan Angove: So yeah, very interesting. It's working capital efficiency, right? Yeah. The faster you're learning your inventory. I mean, all of that. So speed is, is, is everything in the supply chain.
[00:12:37] Steve Hamm: Yeah. No, I I've watched a number of videos of you talking about how technology can help transform business. And you're really good at this.
You have a knack for it. I wonder if you can give us and give the podcast listeners today, Yeah. Kind of an example, kind of blow by blow of how your customer and their end customers benefit from this kind of responsiveness.
[00:12:59] Duncan Angove: Yeah. So that maybe this, maybe there's two examples, Steve, right? The first would be around really at the root of age of uncertainty.
And the second one I'll give us more on the demand side and consumer behavior. So if you think about disruption. What happens today, because supply chains aren't joined up on a common data cloud and let alone inside the enterprise, and certainly don't do it across multiple organizations, a supplier, a carrier, a retailer, if you like, a manufacturer, you know, so what happens today, on average, it takes four and a half days to learn of a supply disruption, right?
A ship could be leaving late or a supplier runs out of a raw material, takes four and a half days to learn of that disruption, right? And normally it takes. Email, or who knows, FAS, right? And then it takes on average over two weeks to actually resolve it. Right, that's what happens today because it's opaque.
You don't know what's going on. So, by having all your applications on sort of a real time, single data cloud, all the applications are designed to work together. And then having a network, we just were, we announced the intent to acquire a company OneNetwork. Which takes that visibility outside the enterprise across 150, 000.
Companies do not have visibility, right? Not only do you instantly see when you have a supply disruption, because everyone's joined up and you have a single version of the truth, you can very quickly Orchestrate an optimal outcome and execute, right? So that's basically fundamental. You don't just need visibility.
You need the ability to be joined up, make the best decision. That's where AI comes in and then orchestrate multiple stakeholders to actually solve it. That's, that would be an example of how we've gone about solving uncertainty.
[00:14:35] Steve Hamm: Well, in that scenario. So does that mean that all of your customers, suppliers, all the way back to, to all the sources are on Snowflake too, or is that, you know, Or what's the, what's the source of transparency and visibility?
[00:14:51] Duncan Angove: Yeah, they, they don't have to be, but obviously if you're a retailer, you're, you're communicating to your suppliers, you're communicating to your, you know, your logistics provider or your tracking fleet, all of that. So all of that data gets pulled into Snowflake. And then we all have to use AI to optimize.
How do we, how do we resolve this risk or this disruption? How do we capitalize on this opportunity? That's basically how it works.
[00:15:15] Steve Hamm: Good example. What was the second one?
[00:15:17] Duncan Angove: Yeah. So the second one was more on the demand side. So, you know, one of the things happened, you know, and digital technologies certainly accelerated.
This is the consumers want to personalize the products that they're buying, right? And probably way of all configure them, certainly. And where that's probably most prevalent and obvious to the audience is in the automotive space, right? Every single car you go, you know, you want to buy now you configure.
I want heated seats. I don't want heated seat. I want adaptive control. You know, I don't. And yeah. What that has is, and by the way, it's part of this process. You're seeing dealerships get disintermediated, all that as well. Right? So what that has is a profound, profound impact on how the automotive manufacturing chain, you know, production line works.
So you think about it, every car that's coming down that production line could be completely unique. This is the furthest thing from like, you know, Ford and the Model T where everything is black and it's exactly the same, right? Every car that's coming down, it is completely different. So what that means is you have to swap out tooling.
Okay. This thing, this tool here, this robot installs a heated seat. Wait, it's not a heated seat. Take it out, put another one in. So you have this change over time. And the second thing is the raw materials or the components that you're staging along that line are also different. You need heated seats. You don't need heated seats.
So it drives a huge amount of complexity and inefficiency in the production line and make it less productive. Right? So we have a solution that basically optimizes how you slot and sequence that whole line to minimize change over. And we ripple it all the way down the supply chain in terms of when components need to arrive in time.
Right? That's an example of how changing the behavior on the consumer side has a profound impact on the supply side and how you can kind of optimize the two things. So you have happy consumers and you have a, you know, a happy automotive manufacturer. Yeah.
[00:17:02] Steve Hamm: So it seems like the way that would play out would be that the dealer would have a bunch of the most popular trims sitting on the, on the lot, but not having, not having to have a huge inventory. And, and many of the cars they sell would be essentially custom built for a particular, particular customer. Correct?
[00:17:24] Duncan Angove: Yeah, that's right. So the whole world is what they call, make the stock to make the order or they get a order and it comes back to speed, right? How quickly can you get, like, I want to wait a year, but you know, the faster you can make, the more efficient you can make it.
It just, It enables that whole consumer behavior.
[00:17:52] Steve Hamm: So you’re a business leader also with an engineering background and you take a deep interest in technology. So I'm going to ask you to put on your kind of visionary cap for a couple of minutes. Looking out five years or more, how do you see technology transforming business and even society?
[00:18:10] Duncan Angove: I could spend a long time talking about this one.
So, let it rip. Yeah, yeah. Obviously, you know, come back to generative AI in a moment, the obvious one, right? But when you, when I step back and I look at what we're, what we're doing, Blue Yonder, right? You know, part of our purpose is to basically deliver something we call sustainable abundance. And what that is, and I'll, I'll unpack it, like the abundance part is, well, we call it our good quest, and it comes from a famous quote by a Facebook engineer who said, sort of, the tragedy of their generation is the brightest minds are, are working on getting people to click on ads, right?
And they're at the root of all these algorithms that are basically, you know, reprogramming humanity and it's, it's, you know, and undermining trust in society, which is a really important currency. You know, so good quests, that's a bad quest. Good quests are things that are operationally hard and consequential and make the world better, right?
So curing Ken, colonizing Marth, things like that. So our good quest is all around sustainable abundance. And if you think about supply chains, they're at the heart of human endeavor. They underpin commerce and trade everywhere it happens. And they've lifted over time, all the way back to the Silk Roads.
You know, they've lifted billions of people out of poverty into the middle class and they'll continue to do that. But if you look at the role of technology and invention, it's always been an amplifier, that, right? Though the role of technology fundamentally is you turn scarcity into abundance, right? We now have cheap clothing, we have abundant medicine for most people.
And that's sort of the role of technology. So think about the invention of money, the invention of the steam engine, the invention of the internet, you know, the shipping container, all of those things were an amplifier to supply chain. So the first part of our good quest is how do we usher in the next wave of abundance, right?
And lift even more people out of, out of poverty, middle class and all of that. That's sort of the first part of abundant, but the other part of it is do it in a more sustainable way. So supply chains produce 60 percent of all the carbon emissions in the world, right? So that's, we can certainly do better than that.
And supply chain, the, the, the dark side of abundance is it also creates waste, right? So you take food waste. Globally, a billion meals a day are wasted, but there's almost 800 million people that go hungry every day. In the U. S. fashion industry, brands overproduce 500 billions of goods annually, right? So, that's another part of sustainability.
It's not just carbon emissions, it's also waste. So, that's sort of our missions. How do we deliver sustainable abundance with You know, the software and the technology that we provide, the largest brands in the world. That's, that's sort of first part of what I'll talk about. Generative AI, obviously, though, is a super important moment in the arc of human history.
And it has, Some interesting impacts on enterprise software and even how enterprise software works, right? So I always talk about the, the notion in, in, in software we've always had a user experience. We call it a ui, right? And fundamentally, the way a user, a human interacts with a computer to instruct it, to automate certain things.
And when software first came out, it was like a, it was like a dos prompt. It was like a text box, kind of Ironically, we've gone back to that. And then you had the RASC interface, and it was a GUI, and I had a mouse, and then mobile came along, and I was dealing with that. So every computing paradigm has brought a different user experience to it.
The difference this time is it's not the eye, it's not the interface being updated. It's the you, the user is being, and it's like, we've, we're downloading sort of a cognitive upgrade for all the humanity, right? Like a patch we're being augmented. Yeah, that's nice. That's very nice. We know that. So we'll be, we'll be replaced.
Like, how does it work? And I think it's going to happen an awful lot faster than we appreciate, you know? So that's something we're wrestling with is it's, how do we. How do we help our users in this sort of change management journey? How do we amplify what they do? You know, in some cases we view it as our end users are all getting a promotion.
If you were an individual contributor, you were a planner, an inventory manager, a warehouse operator. Now you might be the manager of 20 digital agents, right? What does that user experience look like? Right. So in business, it's going to be very, very profound, you know, and it will have a huge impact on, on society as well.
So it's, it's actually a super exciting time.
[00:22:27] Steve Hamm: Do you think the managers of futures will force their, their digital agents to have annual reviews?
[00:22:34] Duncan Angove: I don't know. It'll, it'll be like that 360 loop will be very, very interesting. You know, we just hired a couple of people from Tesla, basically to help us figure out how do you build.
Trust with humans when things become more autonomous, right? And if you think about the evolution of the automobile, we've introduced autonomy, little bits at a time, right? You don't have to sit around with a fan to get the temperature you want. It just does it automatically. The wipers come on when it's raining, the lights come on when it's dark.
And obviously then we introduced adaptive cruise control. It will slow, it sees it like, when do you ultimately trust and take the steering wheel completely? And we're going through that same journey with software, right? At what point do you trust it to do more and more and more and more? You know, the complexity will be, we've always talked about explainability in AI, right?
Why did you recommend to do that? And it will come to a point where the AI is so smart, it will become very, very hard to actually explain it to a human. So there's a law of change management here. The moment critical that we have humans in the loop. For sure.
[00:23:35] Steve Hamm: Yeah. So are you using large language models and generative AI now, or is that something for the future?
For sure.
[00:23:42] Duncan Angove: We've been doing this for 18 months. I mean, we're, you know, one of the frustration, you see, right, is that everyone's now an AI company. They just act on AI. The main name, they use OpenAI, and they put a wrapper on it, and they do text summarization. They're an AI company. We've authentically been doing this for over a decade, but it was in machine learning and predictive AI, which apparently, Anymore.
Right. You know, and like I said, we do 10 billion predictions a day and it's growing at a billion a month. Right. So it was very natural for us to, to leverage generative AI and we've been doing it, you know, at, at scale for, for a while.
[00:24:17] Steve Hamm: For your information, there's a lot more to Ogerson people think.
[00:24:21] Producer: Really need to dig deep and get to know the real you. In the real up close and personal.
[00:24:26] Steve Hamm: I understand you had a close working relationship with Frank Slitman. Snowflake's former CEO, and that you're developing one with the new Snowflake CEO, Shridhar Ramaswamy. And in fact, I, I heard that you, the two of you had breakfast together recently.
So without giving away any state secrets, can you tell us what you talked about?
[00:24:48] Duncan Angove: Yeah. So, you know, Frank and me were on stage last year at our user conference and, you know, we've got a very, very deep partnership and, you know, we share the same vision around, you know, building the world's largest supply chain data cloud.
And our relationship is, is, is being very, very close on the engineering side. Like I said, you know, we were the first company to write applications natively on Snowflake and we worked very closely around, you know, Snowflake's roadmap and Shreed obviously comes from kind of the product and engineering side, much like Meet.
So even before Frank, the part of Shreed was actually our executive sponsor and we worked very, very closely together. So this was, you know, this was a very clean and great transition for us. You know, we had a great, great breakfast. You know, we always start with talking about how do we, how do we make our customers even more successful?
How do we deliver more value? But then, you know, the neat thing about us is that our software is, is bought by the business. It's not bought by IT. So we bring a very applied business oriented mindset to where, you know, Snowflake can, can take its product and kind of what are the workloads and problems that we're, that we're trying to solve.
Like I said, if you console. The problems, the technical problems I talked about in supply chain with our scale and the mission criticality of it, you can solve it in any industry. And those are the things we'll always talk about, right? How can we go faster? How can we innovate all on behalf of end customer success and value?
So that's, that's always what we talk about.
[00:26:15] Steve Hamm: Very cool. Well, this has been a great conversation. I mean, I think you're very, in addition to being a smart businessman and, and, uh, you know, deep on technology, you're also a good talker. So very nice speaking to you today. I really loved when you were talking about the age of uncertainty, you know, we're in it and not just in supply chains, but in, in, in lots of ways.
And, you know, sometimes it takes a bit of a crisis or real urgency for organizations to change. And I think the opportunity and the need is there. So hopefully we'll see a lot of transformation. The other thing I thought you said that was really, it's very encouraging to me because I'm really onto this, but the idea of you believe in sustainable abundance, that's your good quest.
And, you know, whenever you look around the world at, you know, GDP growth and, and more consumption and things like that, that's kind of great for the economy. But on the other hand, it's not so great for the environment. So if somebody can come up with really powerful tools. That allow us to bring up the standard of living of, of, you know, millions and millions and millions of people without kind of burning, you know, the earth to a cinder.
That is a great deed. And I'm glad that you're very consciously pursuing that. So congratulations.
[00:27:39] Duncan Angove: One thing on that Steve that's sort of interesting is that, you know, sometimes it's combinatorial innovation, right? And that's why we, even you look at robotics, right? The sort of rights law, you know, the more people buy them, the volume drive cheapest, and you get this kind of like Amazon flywheel, right?
But when you, we're starting to see physical robots, particularly in a warehouse that are getting combined with Gen AI. So now this dumb robot can actually see and observe the world with human level intelligence, just like you have with Gen AI. So there was a prototype we saw where someone asked a robot that was basically picking toys out of a tray in a warehouse, and it said, pick up the extinct toy, and it picked up the dinosaur.
Oh my god! Brilliant. On the robot. Yes. So when you just think about like, so that's going to power a robot to be even more and more and more. And the challenge you have a lot of times when you want to train autonomous systems and it's why we struggled to get the self driving cars is it's the edge cases, right?
It's the edge cases that cause accidents, but the things we've been able to generate images and video now that we're generating edge cases. Billions of hours of them in training robots on simulated educators, right? So it's the combination of all of this, but the place I was going to was energy. Energy is the other thing.
You look at fusion, fission, you look at battery tech, all of that. So talking about unintended consequences. So AI is about data, the algorithm and computing. Okay. And compute is about energy. That's where the energy comes from. So if you look at the utility stocks in the U. S. Are up 20 percent this year.
And the reason why is AI. So AI right now consumes 4. 4 percent of all the energy in the United States. At the end of this year, it's going to be 10. 8. Yeah. Mind blowing. Think about it. Lights, traffic, all the stuff we use, like, over 10 percent of it's going to go to AI. So we have to solve the energy thing.
Right? And I think all of these things together, like, we will figure it out.
[00:29:40] Steve Hamm: Yeah. Well, we have to make AI more efficient. We can't, we can't, you know, between crypto and AI, they talk about, you know, like whole countries were using the energy of whole countries. And so we need some really efficient algorithms, you know, and training has to be efficient.
So hopefully. It won't be your company that's solving those problems, but hopefully somebody will.
[00:30:04] Duncan Angove: Yep. Well, we're owned by Panasonic and they're the lot, they're the battery company behind. Oh, okay. Oh, very good. They've got two bands, us and batteries.
[00:30:14] Steve Hamm: Wonderful. Wonderful. Well, Duncan, fantastic talking to you today. I think it's just a great conversation. I think the podcast listeners will love it.
[00:30:23] Duncan Angove: Awesome. Well, thank you for inviting me, taking the time. Great questions, good dialogue.
[00:30:27] Producer: Dive deep into the world of apps and generative AI from Snowflake Build. Catch up on the latest announcements focused on building apps, data pipelines, and machine learning workflows in the age of LLMs.
Watch now at snowflake. com slash build.