Video: Leadership Lounge: From Concept to Confidence | Duration: 2108s | Summary: Leadership Lounge: From Concept to Confidence | Chapters: Welcome and Introduction (29.46s), AI Adoption Journey (256.32s), Hybrid AI Partnership (410.725s), Building Enterprise Trust (585.055s), Global Local Manufacturing (761.94s), Reducing Deployment Friction (981.435s), Token Optimization (1104.11s), Reducing AI Friction (1208.39s), Data Center Planning (1360.6599s), AI Productivity Gains (1439.42s), Final Advice (1577.24s)
Transcript for "Leadership Lounge: From Concept to Confidence":
Hi. Welcome to Digital Realty's leadership lounge. I'm Jules Johnston. I'm the senior vice president of Global Channels and Alliances. And here with me today, our guest and my friend, Flynn Malloy, the chief marketing officer for Lenovo. Delighted to have you here. We we're so glad to have you, and we're gonna have a great conversation. But before we even get into the AI questions that I had planned for you, I have to just start with there's so many exciting things going on at Lenovo. And top of mind for me is learning a little bit more about Lenovo and FIFA because with the World Cup, we just saw Lenovo everywhere. So Give us a little bit of an update there. Yeah. 2026 has been a big year for Lenovo. You know, we started the year off with our annual event, Tech World, at the Sphere in Las Vegas, which was absolutely amazing. 40,000,000 people streaming. It was a fantastic show. And we did that really primarily to announce, to the world our very exciting, sponsorship for the World Cup, which was this is the first year we've done that. It's actually the first year FIFA's ever had a technology sponsor. So we did that. That's the way the year started. And then as we closed out our fiscal year in the spring, we had amazing results, and our stock tripled. And so we're all feeling really good, and then starting in June, kicks off the World Cup. And so we are the end to end global technology sponsor. So that's from cell phones. So Motorola is a brand that we own, all the way to, of course, cloud infrastructure, AI solutions, and everything in between. So you are really underpinning the you you were underpinning the FIFA communications and technology that was allowing them to broadcast all of that. Yeah. That's why I mean, it's, you know, different than perhaps like Visa or Coca Cola or other global sponsors. When you're a technology sponsor, you bring all the technology, and you work with them to build all the technology. And and certainly the from the broadcast to the stadium, you know, this was all, you know, Lenovo technology. But what we were excited about was to bring, really, for the first time in an international sports competition like this, AI solutions Mhmm. To the games, you know, to improve the customer experience, the fan experience, the broadcast experience. And we were able to do that in lots of really exciting ways. You know, the ref cam. So I'll talk a little bit about this. It's so cool. The ref cam so the refs run around with, you know, like, a camera and a microphone, and, there's a camera on there, and it if you try to put it into the broadcast, it's jumping around because the guys are they're they're running. They're running. And stabilization technology has been around here for a while, but this high fidelity feed that comes off with this is super low latency. So we put an AI solution in that stabilizes, right, the their feed, you know, with a local GPU rich solution to, you know, to process it and get it back to the broadcast center. And, you know, we thought we'd get a few shots. I don't know what we were thinking, but I didn't think that it would be you know, when do they show such a thing? Like, when does the the RefCam come on? Well, it's right after the big goal and Messi celebrating or whatever it is. And, like so brought to you by Lenovo AI all over, all over the cameras, all over the feed, and it was really great to see improving the way that fans enjoy the game. We had, we built an AI pro, like, a LLM for all the coaches and teams so that they can search and and, you know, level the playing field on analytics so the big teams and the small teams have access to the same Oh, nice. Kind of analytics, on and on and on. Many different solutions. And and we did that as a showcase for, you know, our technology, but also how we can bring AI and technology out all the way to the very edge. You know, in the most high pressure, 6,000,000,000 people, they say, watch watch a game. So that is as high pressure as it gets. That is. Well, congratulations. Yeah. It was it was really fun. It was crazy time. I I Super busy, but it was it was really fun to see. No doubt. But it's a good setup. You know, listening to you talk about all the different things that you were helping people do for AI, you know, we we've teed up this conversation to talk a little bit about what we're observing that customers are trying to do As they struggle with going from AI idea to AI impact, from AI concept to AI confidence, they're getting from pilot to production. What are you seeing customers doing? What's the Lenovo observing is happening? What customers are trying to achieve? Yes. And that that is one of the biggest questions that everybody's asked asking us and asking the, you know, the industry and looking around on, you know, how goes the progress on enterprise and mid market, on business b two b onboarding of AI. We see it all. We read it all. Everybody's talking about the bills are going up. And and what we are observing, we do, survey in partnership with IDC, our own. There there's lots of data out there, but we do our own with, you know, current customers, partners, prospects, 3,000 plus 10 countries, and we ask them, what are you doing? How far is it going? What are the obstacles? What's what's in front of you? How far along are you? And what we have learned sort of agrees that with with the point that you just made is that, you know, pilots are coming to an end. So as the pilots and the projects are starting to, you know, come to an end, you know, functions inside of all of our all of our our customers and our clients are starting to ramp to get into, okay. Okay. This needs to be part of how we do marketing. This needs to be a part of how we do supply chain. This needs to be a part of how we do finance. Comma, they're still, you know, 65% or more, of the companies that we surveyed say that they're they're still coming out of the pilot stage into the ramping stage. So Mhmm. Most what what we believe is that the enterprise, the big build out where you're taking AI into the core processes of, you know, of supply chain and marketing and legal and all of these different functions. That's not happened. Not just yet. Yeah. So if you if you feel like you're a little behind, you're not. Everybody is on the same curve. We're seeing the same thing. And and you'll see you'll see industry leaders, you know, really grab it and run for the the front of the bell curve. Yeah. There's some amazing things they're doing, but the rest of the industry is still ramping. And there's a number of reasons for that. The AI tech is not quite there yet. The app stack is not quite there yet. Even the technology and the data center facilities environment, you know, a lot of learnings that the industry is going through on what's what's the what what is gonna be the right solution for us. And so what we're seeing is a lot of build out, a lot of getting ready, budgets are are moving in the right direction, but it's not it's not done. It's not peaked at all. That's right. We're seeing the same things that you're talking about and feeling confident that we're actually better together and that the combination of Lenovo and Digital Realty can help customers to get that confidence and to move into production. Talk a little bit about why you think this partnership works and we're better together. Well, you know, I'll go back to, kind of one of the core principles. One of the reasons why, you know, I think Lenovo and Digital Realtor are so aligned and have been for a while is we both, you know, we both often talk about, what we call sort of, of course, hybrid cloud, but hybrid AI Mhmm. Which is you don't have to take your data to the AI in the clouds. You can bring the AI out to your data, out to where you are. And in that way, building you know, aligning our strategies, aligning our portfolio, aligning our experts, aligning our go to market with the opportunity that is balancing. You know, we we all do a lot of business in the cloud. Everybody has business in the cloud. But what has happened in terms of the enterprise and and mid market use of AI so far is they're coming out of their pilot stages for their, you know, their private experiments. Everybody's got these cloud subscriptions, and there's so many of them. And those bills are are really high. And so what we often together talk about is you need to find the right balance. Mhmm. Some things should be should be coming from, you know, from the cloud and from the partners and the the subscriptions. Others, you know, how can you bring it into an on prem environment, into a private environment? How can you build those stacks that that bring AI to the data that you need, closest to to your business and to your customers? And in this way, I think better together, neither of us can do that solution for our our clients by themselves. That's right. And I think that's where when we think about hybrid AI and we have been talking together about hybrid AI for for two or three years now, going to each other's sales, you know, kickoffs and doing joint marketing together, and that there's a reason for that because it is a better together solution. When you combine, you know, best in class technology with best in class modern facilities connectivity, you know, you need those two components. And that's what we to answer the question, we have aligned strategy, and we've been working together for a long time to put those two pieces together for our clients. No. I I I agree with you, and I especially think we are poised to help the customers that are grappling with that question you just raised about what should live in the cloud and what should not. And it's been our experience that what they find when they think about what should not is they can't bring that back to where it used to live. They let their former data center operators go. They let that facility go. And even if they haven't let that facility go, that facility may not be able to handle Lenovo's AI optimized systems of today. And so they've got it to your point, choose a modern data center that can bridge them from air to liquid and that, you know, can handle the power needs that they're going to have as their AI ambition scale. So I do think that that's a really powerful combination. Let me let me pivot to a question about trust because so one of trust has always figured into enterprise customer decisions. But right now, I mean, these are nontrivial bets that enterprises are placing. This is not a short term tech stack decision. It's about the company decision for many, many of the folks making those AI infrastructure decisions. And so as they are looking left and right, wanting to keep pace with their competition, they are making really career betting decisions. And, you know, so what what to your mind are things that we do together or individually to help them get that comfort level, get to that trust? Well, we we hear that all the time as well. You know, as we're as we're moving forward and we need to move forward at the speed that the market is moving forward, you know, many CIOs and IT teams are looking down those same questions, and they're asking, you know, where where we can place our bets. And, you know, what I would say is as you know, it starts with what do you need to do. You know, there are some companies to our previous, you know, to our previous topic. There are some companies that, I'm I'm fine running everything out of the public cloud for for, you know, what I'm doing, the brilliant AIs that are in the cloud, the innovation that's there, that's got me. For others, finance, manufacturing, health care. There's so many industries that that's not gonna be the complete solution. And so as they start to think about, okay. I'm I'm gonna need to build out an app stack and a solution that's you know, that goes together with what I'm doing in the cloud. Lots of questions because you do have to make a level of investment. You know, you wanna ask the questions. Can it be backwards compatible? Do I have flexibility of capacity in the future? You know, is what I need to do, the tokens that I need to produce, the inferencing that I need to do, is that going to be covered by, you know, the solution that I'm looking at? And the best way to build trust is, of course, to find partners that you trust who've been working together for a while, but more importantly is to go test it. Right. So we have put together, you know, the Digital Realty Innovation Lab Yeah. That you guys have. So we have our technology in Ash Virginia Mhmm. In, I think, The UK, in Japan, where you can go and see the technology in action in the Innovation Lab. You can run it through its paces. You can come in through the firewall and put your stuff on there and, you know, see it and be able to do some testing of this is what I think I need to use. This is the capacity that I think I need. This is the token throughput that I think I'm gonna have to, forecast and build for, put that technology through its paces in the innovation labs. It's a great way to derisk, you know, what you're gonna go do. And then when you do that, you also meet the brilliant experts of Digital Realty who are in these centers as well as the some of the technology experts that can come and help you assemble that solution. You don't have to do it alone. And you don't you're not, you know, buying click click click okay. You know? Hope it works. You know? This is run it, derisk it, test it, and then you can run from there. No. I I think that's it's important to give them a chance to try that proof concept in this real world, a data center environment as possible. It's huge for for customer trust. And, so we're glad to have you there. And I know you mentioned Ashburn, and you mentioned, Tokyo and London and soon Singapore and Africa Absolutely. South Africa and Johannesburg. So so we're excited to have Lenovo partnered with us there. And is now thinking about how we're going around the globe together, it prompts me to ask about, you know, us striving to meet the customer where their data lives. And and so, that for sovereign reasons, some of the industries you just mentioned have particular regulatory concerns. But, you know, does Lenovo have have, a point of view or observations around meeting customers where the data is? Very much so. Yeah. And let me let me do a little, maybe promo on on Lenovo and and how we approach the answer to this question, just in inside of our company is unlike a lot of our competitors out in the industry, Lenovo, it's like Jensen Huang's favorite thing about us. He always says that. Is that you're both an ODM as well as an OEM, which means we have design teams, engineers that design our tech from the ground up, and it helps inform our original OEM side, the the the technology that we manufacture and sell. And unlike our competitors, we are vertically integrated. So we bend our own sheet metal and burn our own motherboards in our factories around the world. There's, today 12 factories around the world. And for example, in in, EMEA in in Europe, just at a factory that a lot of you guys have been to, by the way, in Ulo, Hungary, just outside of Budapest. 98% of all the technology we sell and ship into Europe is manufactured there in Hungary. So if, you know, you you're looking at a cage, for example, in Frankfurt or in The UK, the technology you're gonna put in there, if you choose to partner with Lenovo and Digital Realty together, it's all manufactured in Europe locally. It's a global local approach. The all the firmware checks are local. All the transportation is local. It's very much a, you know, a global local approach. And for us, we we do this all around the world. So whether it's from Saudi Arabia, whether it's, you know, in North Carolina, in The US, Monterey, Mexico, We've got India, factories all around the world that takes this global local approach, that helps answer the question of I need I need to locally build, with Sovereign AI. I've got all kinds of issues, that compliance issues that I need to address. And as part of that, if you're able to deliver a global local solution as we are, you know, that's when you start to look at, the all the rest of the ecosystem partners that exists there locally, like in Europe, in and around Frankfurt, all of the the IT partners that you might be working with, all of the, you know, the the different kinds of software companies, the application companies that might be stronger in Europe than elsewhere. And so putting together that local ecosystem that's responsive to the local market, to the channel partners that are successful there, it's a lot of times in in our belief, this really answers a lot of the biggest questions, that our customers have on, do you do you work with my ecosystem? Can you build locally? Can you ship locally? Can you get you know, can you help me address those those compliance issues? Whether, again, it's finance, health care, any of these industries. I know. I I those are those are key to helping our shared customers as I feel is simplifying the experience for them because, this is nontrivial complexity. And so some of the things that we're doing together in integrating and integrating some offers are simplifying things for customers, and we're engaging channel partners to help us do that on behalf of our shared customers. A little bit about your thoughts there. Yeah. So, you know, taking the friction out, I love that expression of, you know, how do you because we hear a lot. We we touched on it earlier. I hear this we, our customers, our sales leaders, hear this from our clients around the world. I love your stuff. I got nowhere to put it. It there's too much power. I I it's not plumbed for liquid. You know? What am I gonna do? Can help with that. Right. Right. And so, you know, okay. So how do we do that? You know? You've gotta contact the different companies. You've gotta bring your partner in. You know? There's contracts. There's it it it's a it's a complex set of things that you need to pull off. You know, when your end goal is, you know, tokens, you know, time to value by the end of this quarter, I need to be up and running. So everything that we, our ecosystem, and as partners, we can do to take the friction out of that system, which means, you know, for example, sales tools that our that our teams can have that specs out the tech or specs out the, you know, sort of indicative pricing for we got some room in a cage in Dallas or Austin or whatever that is to to help this, you know, the the the conversation go quicker Mhmm. Is is really important. We've invested a lot of time and energy, a lot more coming on that subject, by the way. As well, we've been working on the offerings themselves. Can we package together T shirt size, combinations of our AI optimized tech, you know, in in, you know, with the facilities, connectivity, all the things that you need, small, medium, and large, make it sellable by our joint channel partners. Mhmm. So lots more. We've been doing some great work. I know we've got some announcements coming this fall at supercomputing, here in the fall, but lots more to come on that, which, again, what we hear from all of our clients again and again is as we start to step in to really building out the, you know, the on prem or the enterprise, the private side of of this hybrid equation, the more you can give us an easy button for what we want, test it, you know, in a in a drill facility, and then give me an easy button for actually onboarding it and getting time to token, that's all gonna help make their lives easier, faster, and more responsive. It it absolutely will. Okay. So now I'm going to actually come to Flint with some questions that we ask all of our guest leaders at the leadership lounge to share your personal thoughts and observations as we're all grappling with AI. So I'm gonna turn to you few of those questions. And the first one is, what is the question that you wish more business leaders who are prepping for AI or asking themselves right now? Yeah. I'm gonna go with I mean, if you're you know, most business leaders today and certainly IT leaders, are are reading you know, there's been this this, pendulum of, you know, I need token maxing. I want my team choosing as many tokens as possible, and then the the budget runs out in February for the whole year. Oh oh, okay. We can't token max. You can blow entire budget on that. So we're gonna have to think of tokens differently, and we're gonna have to token mins. And, you know, what I would say is that one token does not fit all tokens. There's not just one size fits all for tokens. And the token optimization is what a lot of, you know, AI experts as the enterprise and the b to b onboarding ramps, how do you optimize around tokens? So, you know, small models for cheaper tokens, the big tough ones for that you know, like, one does not fit all. And so think about where you run your tokens, across a hybrid environment and, you know, figure out or, you know, work with your partners and our services experts to help you optimize those token packs so that you're getting the best ROI for the type of token and for the type of AI that you need to deliver. Yeah. I think that Token optimization. Token. It's not really a lightning round answer. My bad. So I No. No. No. No. I I explanation to get there. It's a good one. Okay. K. New question. Okay. So, Gwen, you and Lenovo are in a position to observe the AI evolution. And, as as we are all watching this, what observations do you have about the gaps you might be observing between partner readiness and customer demand? And where might you encourage partners in addition maybe to token optimization, but what else might you encourage partners to focus on helping customers on their journey? Friction, friction, friction. So I'm gonna go right back to that. I I know this is a lightning round that you said at the end, but the friction is what everybody is experiencing now. You know, you've got a goal. You've got a north star. Here's what I wanna do. How do you take the friction out? How and look for partners upstream that are focusing on this. Let me give you another example of friction. One of the things, that that we focus on in Lenovo is as we build out our new technology, it's super important that we build it so that it is backwards compatible, so that it's 19 inches deep, so it fits in your current environment. So much of that is just friction. If you if you design a solution that doesn't fit into a 19 inch rack, because the footprint, like, in your data center or anywhere else, And then all the way down the line, you're putting a solution together, and now you can't get the damn thing or the darn thing into an elevator, for example. So that's friction. You know? There's friction on, well, who do I call, you know, to assemble? I need a couple of changes in the contract stuff. Well, who do I call? How do I do that? How do I build these solutions? And so what I would say to a channel partner is prioritize partners who have been working on friction reduction Mhmm. For a long time. It AI is new. And if we go back, for example, to the first, like, five years, of of cloud, so twenty ten, twenty twelve. Back in the early days of cloud, it was the same thing where everybody's running. We're we're all, you know, putting up, you know, contracts in place and running like crazy, and there was friction everywhere. And it took, I think, IT managers a long time to realize that IT ops is not cloud ops. How you run, procure, manage, and run your IT that you're used to, you have to kinda rethink the entire thing when you're running in cloud. You gotta turn the lights off at night. You've gotta manage it from a central, you know, broker. It's a it's a different process. And I think we're in the same phase of AI right now. Everybody's running to it. They're trying to build. There's a ton of friction in the system, so it's incumbent on the providers, the builders, you know, the facilities, teams as well as the technology teams and our channel partners to be looking for how can you take friction out of that system. I I mean, that is so right. And I would also put a plug for partners who are trying to, add more value to customers that they do well to school themselves on, the consideration of the data center and the downstream destination earlier in the cycle. Oh, man. Because we are in an era where demand is outpacing supply. There are power constraints. There are fewer facilities that can handle liquid than the technology suggests are necessary. And so if partners wanna be in a position to derisk the downstream data center destination impediment and derisk the project being delayed or worse, not having to settle for less, I think schooling themselves out, about how to have that conversation holistically upfront and pull the data center discussion forward so that it can handle, the Lenovo's AI optimized systems and the like. I think that's a really important thing for customer for partners. Absolutely right. Think about the friction of getting that solution. You run it. You're so excited. You go all the way there, and you're like, uh-oh. I I I don't have the power for it, or I can't do x y z, or I won't fit in the elevator. Like, that that derails the whole thing. Back to square one. That's friction. It slows everything down. Superm-mm. So you gotta look all the way end to end as you're building those solutions. Absolutely right. That's absolutely right. Okay. So, that one, which is really more about sort of the rules and skills that organizations and leaders in particular are are thinking about as they scale their AI operations and what's becoming needed. I mean, I'll I'll say from my own perspective at Digital Realty is we try to optimize AI internally as well. We've accepted that being able to write a good prompt as a table stake. And, and as a leader myself, I'm challenging myself to do things differently. So if I know how to do it, and I used to do it myself, but I can I can have AI do it more efficiently or faster for me, and I can create think time, that can create strategic time or coaching time? I owe it to my team and and to to try to do that Mhmm. And to try to do it differently. So so both personally and as an organization, we're thinking about that. Thoughts thoughts on how you and your team are doing anything differently. Well, I can say that everybody is saying the same thing. And and the same thing is, you know, we need to go faster. We need to go better. And, you know, oftentimes, in the first like, go back to the cloud example. The first phase, the ramping part of a tech of a new technology, Internet back in the olden days, it starts with productivity. It used to take me four hours. Now I can do it in fifteen minutes. And, yeah, that's really important. I mean, it for for marketing teams, oftentimes, that translates for okay. So we can reduce your budget. Right? Well, okay. You can, but that is a choice. We don't like that choice. What we can do though is talk to how if I'm driving, you know, 20 to 30% more productivity from the various pieces of of my team, the different functions in my team, we measure that. And then how do you get the increased output out of that? So that's that's where we've approached. So whether it's branding, whether it's building content, whether it's, you know, reaching customers, whether it's analyzing all of these things, how can you get 20 or 30% more productivity out of that and then to be able to do more with the same amount? So we we spend a lot of time on that, and then the team will say and to your point, well, how how do I you gotta get smart. You gotta read your medical journals or your AI journals, so to speak. You've got to take the training courses. Yeah. You gotta get certified. You know, in the early days of cloud, we all got cloud certified because you can't really understand how to use the technology until you understand the technology. Yeah. Totally, totally true. So, okay. Let me, let me ask you, what excites you most about AI over the next twelve to eighteen months? Well, I'm gonna I'm gonna go for I'm gonna go for I'll I'll bring it back to FIFA because, you know, what we did at the World Cup for Lenovo was truly taking AI out to the world, all the way to the end. So where is your data coming from? Where are your customers? It's all the way out, you know, in the manufacturing site, in the stadium, you know, in the broadcast center, in in the hospital, in the, you know, in the in the bank, all of those places. That's where customers are. That's where the data is. That's where it lives. And, I think what's exciting, that we got to showcase in the FIFA, you know, all of the games and everything that we did there is that, you know, when AI is done in the right way, you can improve how people enjoy the game. You can improve, you know, the experience. If something you thought, everybody knows global football. Like, that is the most popular sport in the whole world. How can you possibly help improve its fan experience That you with a game experience? And that's where that challenge is. So as you think about, know, what's going on in your hospital or in your manufacturing site or in, you know, your financial institution, there are unknown ways that AI is gonna help you improve your customer experience, improve how you do business. And I think we've only just started scratching the surface. As enterprises really start building AI into the core workflows of all functions, supply chain, legal, marketing, finance, everything, you are gonna see massive change that's going to be better for the world, better for the customers. So I'm very excited about that element. Okay. Me too. The possibilities are immense. Alright. Now let me ask you. Do you have one bit of advice for the CIOs who are watching us now as they prep to action their company's AI ambitions? What's that bit of advice? Look around. I think, too often, especially in this high paced environment, you know, we'll we'll we'll stand up and look around a little bit, look outside our company a bit, and then go right back to what we were doing. And I think, I I say this to my teams all the time. I say this in turn to Lenovo, you know, watch what everybody else is doing. Not necessarily to copy it, but to learn, to see the pace. Who are the leaders in your industry? What are they doing? How are they doing it? I think that that looking around, we all don't do enough of that. That's right. And I as I talk to customer after customer, I oftentimes come in and I'll have you seen the way they're doing it? And look outside your core competitor set. So you might be in manufacturing. What are they doing in health care? What are they doing in finance? Many of the same problems we're all facing. And so that would be my advice, which is keep your head on a swivel. Keep looking around, especially in as things like AI are moving so quickly, you know, seeing what what others are doing, seeing what is working, see what the industry leaders are doing, not just in your own, you know, a a function or vertical, but across the industry, what are others doing. I think that's how we learn, all of us, in this fast paced environment. And I think partners like us are in a position to help you by sharing some of what we're seeing across across verticals across your, you know, your similar company sets. So take us up on that. So by way of recap, I hope some of these things we were able to talk about are were interesting to you watching. I purse I love that we're better together. I love that we're helping customers move from concept to confidence. I think what they're grappling with with hybrid AI and repatriation is nontrivial, and they do need, to look around at at other others and learn from those. And, and I think you you've certainly sent me to doing some reading on token optimization, perhaps more for all of you too. So thank you again for being here with us. This has been a pleasure. Great conversation.