Inside FedEx’s AI Strategy: Why the CEO and CIO Must Move in Lockstep

0:00Vishal Talwar: I obviously have somebody as capable as Vishal with me to make sure that all the crazy things I dream about, that somebody can actually do it.

0:07Raj Subramaniam: You cool with that, Vishal? I'm 100 % cool with anything Raj says here.

0:11Vishal Talwar: For 50 years, FedEx was run by Fred Smith, the founder. And then four years ago, it changed hands. And we have the man who is now taking it to the next level, Raj Subramaniam, with us. And with him is his chief data and information officer, Vishal Talwar. We're going to talk a lot about AI, but I want to ask you a question that's very important to this crowd. You're the CEO. You are the CIO. Who runs AI? Well, at this point, you know, I think... I think we just got the answer, folks. I don't think any company today can succeed in AI without the CEO taking active leadership role in AI. Because?

1:02Vishal Talwar: Because I think the gap between business and technology is fast moving away. You can't have a business strategy without technology, and the reverse is also the case. And AI is the defining technology of the era. So the CEO has to drive it and obviously have somebody as capable as Vishal with me to make sure that all the crazy things I dream about, that somebody can actually do it. But the strategy is going to have to, I think, has got to be top down.

1:32Raj Subramaniam: You cool with that, Faisal? I'm 100 % cool with anything Rod says here.

1:37Vishal Talwar: Well done. Brave of you to say so. You know, the conversation we've been having, we have a lot to talk about, so I won't spend too long on this, but the conversation we've been having in this room over the last few hours and in the breakout groups is exactly about that, because the challenge of AI transformation isn't just technology, and I think in the view of many of you, isn't primarily technology. It's a work processes, operations challenge. It's a culture challenge. It's a people challenge. And ultimately, it's a leadership challenge. Yeah. Well, when we think about AI and the way, at least, we have mapped it out, obviously the first and foremost, the fuel for AI is data.

2:23Vishal Talwar: And you just mentioned it, two petabytes of data every single day. That represents the movement of the world's most premium supply chain. So$2 trillion of commerce passes through FedEx every year. And so the information and the insights about global supply chain is really in that data. That's the fuel for AI. And prior to the pandemic, supply chains were not that exciting. But then, you know, after that, it became, I don't know, I think even talk show. Yeah, what we did became cool. And so it became boardroom conversation. So we evolved our vision to make supply chains smarter for everyone. And that's where this comes in.

3:05Vishal Talwar: So, of course, we are going to use our AI and data to make our network more efficient, our back office more productive, and so on and so forth. But that's where most of the conversations in most companies stop. We actually think that's just one base layer. But the next layer on top, where we use AI to differentiate our services for our customers, to drive more business, that really is really important. So, for example, the healthcare customers, we improve their predictability of what's arriving so they can schedule their workflow significantly better. and we have grown our healthcare transportation business by 30 % in the last couple.

3:44Vishal Talwar: Because it's more predictable. Because of the, yes, value adds that we have given. That's one example. Yeah. But the value adds that we've given our customers because of the use of AI. I'm going to let Vishal talk here for a second, if you're okay with that. Please. Now the conversation will get smarter. At least you gave me an opening here. But give us some other specific examples. Because, you know, we think of FedEx as, you know, the truck drives up and it puts a box on our doorstep and then it drives away. And, you know, what does AI have to do with any of that? So give us a couple of examples where you see transformational capability at FedEx out of this data.

4:26Raj Subramaniam: Sure. And Raj spoke about two things, Alan. So he spoke about two pillars where we are operating. He spoke about efficiency, which everyone's going after, differentiation and how we serve our customers. The true transformative is the third pillar, which is how do we start to create new business models for FedEx through this process? And there's enough points on the board that we've put in the efficiency layer. So as an example, our most valuable asset is our air fleet. And we've cut down air fleet maintenance research time from 30 minutes to three minutes. And that's a 90 % improvement. That just allows me to put an aircraft back in service that much faster.

5:10Raj Subramaniam: We've taken the parcel volume and package, which if you don't ship a standard package, and there's a surcharge associated with that. We were leaking surcharge through our business every year because we couldn't keep up with tracking all the packages that were non-standard size if the information was not corrected. we've recovered about tens of millions of dollars of annual revenue as a result of that because now we're using AI to identify what those abnormal packages are that should ideally have had a surcharge. Raj started to speak about this differentiation that we're giving to our customers.

5:52Raj Subramaniam: The one thing our customers care most about is don't surprise us. They want certainty. And when you think in terms of what that means for operating a network our size, when something leaves 8 ,000 miles away and needs to be, you know, in another destination 48 hours or 36 hours from that, for us to be able to give you certainty in saying within two hours of window, we can put anything that you can pick up from China or Singapore or Asia or from U.S. to put there, we can give you the certainty of within two-hour window of when that thing will be.

6:27Vishal Talwar: Those are all great examples, Vishal. You can provide more certainty. You can make more efficient use of your airplanes. But at the end of the day, what FedEx did for us all was to increase the speed by which something could get from point A to point B. It's just an incredible model for rapidly increasing that. You didn't mention increasing the speed of getting from point A to point B. What's it going to take for us to see significant increases, or have we reached a limit on the speed of delivery? Well, we are leaving China and delivering to every zip code in America overnight, so I don't know how much faster that's going to be.

7:08Vishal Talwar: You're doing pretty well, yeah. Now we are at the speed of aviation technology, and that's kind of it at this point. So it's going to have to be something kind of, what's it? Teleportation. Teleportation, yeah, teleportation. That's what's... Well, great. Are you guys working on teleportation? You don't want to be disrupted.

7:28Raj Subramaniam: We are working on something different, though, because I think the focus that we have is not about the speed while an object is in motion. You can optimize the speed. We are trying to figure out wherever the friction points are in supply chains end-to-end. So an example of that would be I can put anything from China to U.S. in 13 hours, 14 hour flight time. Then it sits in customs for clearance for another four days. So the number of amount of time that's spent in something waiting when it's not in motion, that's where the frictions in supply chain is.

8:06Vishal Talwar: Is that your problem or their problem or are you trying to help them solve their problem?

8:10Raj Subramaniam: This is a problem in industry at large. It's not a problem that's just for FedEx, but it is eventually a problem for society where we are leaking about$1.9 trillion of annual value just because of these inefficiencies and friction points within supply chain. That is a problem we are out to solve. That's, when we solve it, will make supply chain smarter for everyone.

8:36Vishal Talwar: Raj, a year ago, when you knew you had to find a new CIO, what were you looking for? What was the job description? Well, I think I had a pretty clear vision from where this company had to go. Because, as I said, a couple of things happened. One is that we recognize the value of our data that we are sitting on, on the supply chain insights that are 700 planes, are 200 ,000 vehicles, and 5 ,000 facilities, and most importantly, are 500 ,000 team members every single day. We were on top of that. And then along comes the AI revolution, which then really superpowers that. So that's one set of things that happened.

9:22Vishal Talwar: The other thing is the importance of supply chain, and that now becomes a boardroom conversation. So I knew from a point of where we need to go, So, you know, of course we need to continue to deliver on our transformation and our business. But at the same time, that's not enough. We had to take it to another level. So I was looking for somebody, first of all, who would or understood and aligned with that vision. That was, you know, so that when we have this conversation, at least at the high level, that somebody would say, yeah, that makes sense, that doesn't make sense. At least, you know, we are generally aligned.

9:58Vishal Talwar: And second was ability to take that and execute. and get points on the board. So those would be the top two things, but importantly also a very important cultural fit to the organization because without having a teamwork, this won't work. So those would be the top two, three things. And who was your first choice? It did not take, I mean, if you notice the time difference between, the selection process was like that. It was pretty quick. And Vishal, you were at Accenture. You had a pretty good gig at Accenture. You were running their global technology program. What attracted you to this kind of stuffy old moving boxes around job?

10:43Raj Subramaniam: Exactly what Roger said, which is... Well done again, too.

10:53Raj Subramaniam: When I first got associated with FedEx, my assignment was... It was January of 2024. My assignment was, can you spend one day a month in trying to figure out what we are trying to do here and help the team? I stayed back in Memphis about 186 days that year. And the reason I stayed back is exactly the reason I joined. The vision and the opportunity is so extensive. and it's rarely that you get as a CIO an opportunity where the CEO firmly organically believes that technology is a feature and not a bug. And second, a CEO that's also strategic is transformative and is looking to blur the lines, if not completely erase the lines between business and technology.

11:41Raj Subramaniam: So when you get a CEO that embraces technology as extensively as Raj does, I knew that if I wanted to spend my career putting technology to good use and also look at the mission that FedEx was on to solve for the world and society at large, which is making supply chain smarter for everyone, this was an opportunity not to let go of it.

12:03Vishal Talwar: So you two have a pretty good common vision of where you're going. 100%. I think that's very important for this job description. I would say that and then we talk almost on a daily basis. Every day? Pretty much. I would say. Do you ever disagree? 100%. Yes. Perhaps on cricket, yeah. On cricket, yeah. But anything else besides cricket?

12:24Raj Subramaniam: We debate all the time. I mean, we have so many debates. Every part of our strategy, we have so many debates. We were looking at making... Who wins the debates? Sometimes he does, sometimes I do. That's right. But I would say always FedEx wins the debate.

12:38Vishal Talwar: Yeah. Can you talk, Vishal, a little bit about DataWorks? Explain what DataWorks is. Well, let me ask you before you do that. If you think about the transformation, the vision that you both share, and put it on a scale of 1 to 10, where are we right now?

12:59Raj Subramaniam: In execution?

13:00Vishal Talwar: Yeah. I would say we are early innings, very early innings. I would say we are two. And I think we are just getting started. it. I think we're building the infrastructure. We've got the first, you know, obviously we are seeing growth in our transportation business because of the technology that we have and the solutions we are putting in place. And that's starting to grow our business, especially in the industrial sector. That gives us proof points that this is working. But there is a long way to go. With the$1.9 trillion of supply chain, inefficiency that's in the global system, we want to be able to have a crack at that.

13:38Vishal Talwar: So you're going after all the inefficiency in the global supply chain. You're not just talking about moving stuff from A to B. No. I mean, I was talking to, I want to name one of the very top customers, you know, between the wafer that they produce in Taiwan to the revenue. But we're not going to say who it is. Revenue recognition. there is billions of dollars that's in that cloud of dust. And so if you were to speed up that supply chain or make that faster, there's a significant amount of value you can generate. DataWorks, I assume, is an important part of the equation. DataWorks is certainly a big...

14:17Vishal Talwar: Explain what DataWorks is.

14:18Raj Subramaniam: So DataWorks is... So if you think of FedEx as the transportation business, which is the association most folks in the room, I would assume, have, DataWorks is bringing digital solutions in the area of supply chain to our customers. And it's a separate independent operating unit within FedEx with just that purpose. When we go beyond transportation to look at solving for supply chain more holistically, DataWorks is the unit through which we do that.

14:45Vishal Talwar: Yeah. And what about, I mean, we're just really at the beginning of seeing the potential for AI in the physical world. A lot of talk now about world models and, you know, Robotics are still in the early days. How important are those things to your vision of what FedEx will be? It's a great question and very important. If you go to our large FedEx surface hubs, we are ultimately an inventory in motion system. And so these packages come in through the facility. There's massive facilities. And they leave the facilities in a matter of minutes. and there's nobody touching it. It's automated, except in two places, truck unloading and truck loading.

15:33Vishal Talwar: Now, these are extremely complex physical robotics challenges to have because imagine you're loading a truck, packages of different sizes, shapes, weights coming through. You don't know what's coming exactly. And then it has to figure out how to actually play Tetris on the truck. And incredibly, over the last four or five years, this particular problem is getting solved. Actually, we're deploying, you know, 60 robots before this peak in one facility just to test this out. So we may only be a couple of years away from that. It could be. So this truck loading and unloading problem could be solved.

16:13Vishal Talwar: So that's one aspect of physical AI, the middle mile. We have autonomous trucks. We're working with Aurora to move. because we have all these facilities on the highway, so you can go facility to facility. You don't really want a big truck with an unmanned truck on the interior roads, perhaps, but definitely on the highway. So we are working on that, and so there are a lot of physical AI applications that are coming. And I don't know, Michel.

16:43Raj Subramaniam: On top of that, when you think about the two petabytes of data we just spoke about, and when you think about real-world model, a real world model for supply chain or logistics does not exist today. And who's going to create it? It has to be one of the physical operators of that network, which is us. And when you think about the two petabytes of data, this does not include the amount of intelligence our physical network on a daily basis will start to generate when every aspect of that network becomes intelligent. Our labels are still passive labels today. That's how the industry operates. It's not a smart device because the cost for technology is not there.

17:15Vishal Talwar: And I think I'm right that to date, even with this aggressive adoption of technology, you're still adding employees. Yeah. So last year in the United States, we added 25 ,000 new employees. But at some point, does that turn? At some point, do you start shedding employees if you've got robots that can load trucks and drive trucks? I think you're not going to see AI delivering packages anytime soon. and the last mile, the last 10 feet, those are even more complex problems to solve. So we have a, and these robots are going to work hand in hand with our team members. So I don't expect that to happen.

17:58Vishal Talwar: I think we are thinking of AI as a driver of growth. Growth drives packages, packages drives people. You know, the last time I talked to Fred Smith, who I think you all know died a year ago, but the last time I talked to him, He was still CEO. And Russia had just invaded Ukraine like two days after it happened. And he was in his element. I mean, he was talking about how FedEx planes were having to be redeployed because you couldn't fly over. You couldn't fly over Russia and you couldn't fly over Ukraine. And you how is geopolitics disrupting your business? We've been talking about technology, but you've got this, you know, you've got wars, you've got gasoline problems.

18:40Vishal Talwar: How do you think about that? So we operate in 220 countries and territories in the world. And every twitch and turn that happens anywhere in the world has an impact on FedEx. So I think the first thing I do before I open my email every day is to say a little prayer.

18:58Vishal Talwar: But as you can imagine, the Russian war, we had to take a southern route to avoid that space. Now we are taking even more southern route for the Middle East. I mean, these are all real. Are you using Dubai? Well, so Dubai hub is not operational right now. We are picking up and delivering in Dubai, and we are moving other ways to get it out of Dubai. But you're no longer using it as a hub. Our hub is, you know, we want to get that open as soon as we possibly can, but not right now. We have started flying back into Riyadh. but these are the kind of things that we have to make decisions on every single day.

19:42Vishal Talwar: It's a very dynamic environment, but what I'm really excited about is how the team has delivered in this kind of circumstances. What I like to say is that the world has got more complex. Before, basic math would have done. Now we need advanced calculus, and there are very few people who do advanced calculus, and we're one of them. Vishal, have you figured out how AI can help you deal with geopolitical problems?

20:06Raj Subramaniam: Not yet. But it is a part of a lot of the discussions that Raj and I are having with CEOs. Because the question that you asked, that's a question on every CEO's mind. So when they're thinking about supply chain, they're thinking in terms of what should my location strategy be? What should my supply chain strategy be? Where should I start sourcing from now? Where is the next geopolitical shock that's going to come? How should I do scenario modeling for tariffs? Where should my manufacturing capacity be? And which lanes should I be in and not be in? And that's where FedEx DataWorks is starting to provide a lot of help to those CEOs in helping them get just a little bit more edge than what they have through reactive information.

20:50Vishal Talwar: Can the two of you envision a day where DataWorks is a bigger part of the business than the movement of packages? Well, I think the way I see it is that DataWorks exists only because of the physical. The physical work that we do every single day is the moat. And that is what the data created from that physical work is why DataWorks exists. So it cannot be in... They can't be separated. It is the data that feeds it, which is actually a good thing. Otherwise, some tech company can't figure this out. We're running out of time quickly, but I want to talk a little bit about leadership because that's what we do at the Leadership Institute.

21:31Vishal Talwar: Raj, you were a chemical engineer. Where did you learn to be a CEO? Well, I often say, and you just mentioned Fred Smith, our founder. So taking over from a founder is obviously an incredible honor. Some would say an incredible challenge. Well, in this particular case, I say that I can see far because I'm standing on the shoulder of a giant. And it has just been an incredible experience working with him. It's not only just business school, but geography school, history school, perspective. And I had the good fortune of spending hours and hours and days and days with him and years. So he was it.

22:18Vishal Talwar: So it was very helpful, especially in this journey. Vishal, where did you learn leadership?

22:25Raj Subramaniam: I'm still learning. And for me, I think I was fortunate enough to have a long stint at IBM, at Accenture, work with clients globally, just have a different view into problems that most organizations across the globe had been solving. So a wide exposure from all of those. And the ability to extract out and zoom in and do that rapidly is what I guess is helping me now.

22:53Vishal Talwar: Do either of you watch Ted Lasso? Yes. I've not watched it, but I have one of our team members or senior leaders in Europe who I send over from our unit here. I call him Ted Lasso. So you know enough. I know about Ted Lasso. You know enough. Last question. You watch it? Oh, I love it. Yeah, I've watched it four times.

23:17Raj Subramaniam: Leadership lessons there? So many of them. I just keep it simple. is the foremost one from there.

23:25Vishal Talwar: Last question for the two of you. We're in the middle of this great debate. I mean, the last week it's been very hot, right? Is AI going to transform human discovery and prosperity or is it going to lead to human extinction? When the two of you discuss that issue, who takes the optimistic side and who takes the pessimistic side? Where are you on the scale of great opportunity or great risk?

23:53Raj Subramaniam: This one I can say safely for both of us. There is no debate. We both think that it's in our hands and it's up to us to apply it for good. And I think it can do a lot of good for the society.

24:08Vishal Talwar: We're an optimistic seller. Let's call it that. Solidly. I like, Vishal, you have an AI for good pin on your lapel. We should talk about that. Yes.

24:17Raj Subramaniam: And it's one of the things that, you know, when you speak about what we are doing in AI, AI for Good is a newly formed commission that happened with public figures, government figures, industry, tech, to say how do we leverage AI for the better of society? How do we learn from other technology industrial revolutions that have happened in the past and not make the same mistakes? How do we not disadvantage societies that don't have access? How do we make sure we don't continue to have cognitive decline in the next generation that's coming. So it's a topic that's pretty close to our heart at FedEx.

24:51Raj Subramaniam: We are part of the founding commission of AI for Good. I've been wearing this pin because of the narrative that's been leading in the media for the last couple of years.

24:56Vishal Talwar: And you feel pretty good about the path we're on. 100%. Thank you so much. That's encouraging. Great to have the two of you here. You're clearly a fantastic team, but we know who's in charge. Thanks so much. Today's episode was produced by Kevin Burke, Michael Claudio, and Lulu Chang. It was edited by Nicole Zirilla, and my executive producer is Lulu Chang.