 Hello, and welcome back to SuperCloud 4, live here from Palo Alto on our studios. We're really excited to be talking Gen AI and AI transforming every industry. And for that, I'm probably bringing to you one of my favorite industries and one of my favorite companies. I am a season ticket holder for the Patriots, so I will put that out there and they happen to have beat the bills this weekend, which I'm going to bring on board here are my one of my favorite, you know, colleagues and former customers of mine. But Aaron, I'm going to screw your name up again. Amanda Leah. Yeah. Amanda Leah. There I go. Get it right the last time, but welcome on board. You're just, you know, deputy CIO for the NFL, National Football League. Really happy to have you on here. What's really exciting is fan engagement and the fact that you have back end operations, you have a franchise model. There's all kinds of exciting things going on within your industry and the media and entertainment industry. And AI is touching so many different pieces of that. Why don't you kind of give people, you know, an understanding of what kind of you do for the NFL. Sure. How that really works. Yeah. Sure, Rob. So as you said, I am the deputy CIO and I really focus the core of our technology. So my team also responsible for our infrastructure and cloud operations, our data centers and things like that. We have a lot of physical assets at these stadiums as well as what we do for our event Super Bowl draft international games. Those are Lee owned and operated, including our fan facing app one pass and responsible for our PMO and a lot of the IT services that go across the whole organization. So another part of my job is leading our innovation framework, which jointly developed the CIO here, Gary. And that is how we're trying to transform. So I think a lot of what we're doing the AI, we've been doing AI for many years at the NFL, but now we're on this new era of AI with Gen AI and machine learning starting to emerge. Some of those things really come to maturity on the game, on game day on the field, as well as for what we do as fans. So yeah, we can talk through some of those things. Yeah. And I mean, none of that helped your bills this weekend, but that's okay. Well, it's running it in. Well, I have to because I'm not going to have many opportunities this season. But you know, I do love the fact that you guys focus on game day, you focus on the fan experience. How is AI really coming to the field, coming to that fan experience? Why don't we start there? Yeah, as you start with the fan experience, you know, really, we've started to look with a lot of our partners. We have some of the most advanced technology partners, you know, in the world. We have Amazon, we have Verizon, we have Microsoft, amazing partners to work with. And you start to think about, you know, what happens around some of these things we've delivered that already used machine learning, already used automation such as next gen stats, such as what we do for our broadcast today. How do you start to combine these things to create a more immersive experience? So around the fan engagement side, that's really where we're aiming. Now in the past, latency has been a challenge there of synchronizing the live game to the data inputs from sensors and potentially from the future as we look at optical tracking from cameras and computer vision. But these things, the processing time is starting to come down and the ability to overlay live stats, live feedback from the field with the live broadcast, not the delayed broadcast, but actual live game time. That's where we're heading toward with the fan experience. And I'm really excited about those. And I think that one thing that Gen AI and AI that it's emerging can do that we could not do in the past of machine learning is really looking at multiple data sources. So you have sensors in the pads on the players, but you might have cameras doing optical tracking. And how do you take two different data sources and combine them to make some sort of either a decision-aiding experience or a fan engagement experience from two disparate data sets near real time? That's really the challenge and the opportunity we have right now. Yeah, I think that, you know, just as a fan and as sitting in the stands and sitting there and being able to have more information at my fingertips and being able to understand what's going on, the game within the game seems that seems to be where you guys are going. Not only that, but player safety and other things as well, right? And it's a fine line, meaning we don't want people holding up their phones and staring at those all game. We want them engaged. We want you to feel the energy. And we know that's what the fans want, too. You're there to experience the game, not to just be absorbed by your screen. So how do we build experiences that perhaps are AR or their mixed reality, but they don't interfere with the game to experience? They're also not clunky, right? We're not asking you to do something you wouldn't normally do when you sit in that seat or if you're in a suite. You know, we don't really envision 70,000 people in the stadium putting on a headset, but that might be an experience for some. So how do we tune these opportunities? We have to actually be applicable and enhance and add to your game to experience, first to track or distract from it. That's the opportunity challenge in front of us. I think the other thing that we really consider with all these things is what is meaningful to the day and what is accurate? We don't want anything that potentially misleads or has inaccurate information or is not consistent in the experience. So, you know, Gen AI is famous right now for our times being inconsistent. Our hallucinating, we can't have that. There's a objective reality of what's going on, and that has to always match and be accurate for us, whether it's a fan experience or it's a decision making experience for the game. And especially when, I mean, the fans are already hallucinating from the parking lot before the games anyway. So, you know, it's you got to you got to you got to bring more enjoyment. And I got to say, you know, obviously I'm a big fan, you know, season ticket holder, all of that fun stuff. I love the product. And I think that you guys do a good job of balancing the enhancement of the product from, you know, again, trying to get in the way of what's going on on the field. And I think that to me is really key. How does that really impact kind of the back office aspects of what's going on? Well, let's start with game day. So you think of the workflows that could benefit from AI. And for us, we've been looking at things within the officiating of the game and with the game operations. Again, accuracy, latency are two really important things. If you're going to impact the ability to have real time decision making, and here I'm talking about aiding decision making, right? Our humans and our officials, our skilled people in the field, that's where the decision is going to happen. But how can we take some factors of information supplied to them faster, supplied to them more accurately? Do we have wearables or other types of devices that we can pipe information to officials, you know, potentially to other people who have aspects operating the game? That's what we're looking at. But now those things have to work. And they have to work across indoor, outdoor, high temperature, low temperatures, all different types of environmental conditions. So whatever solutions we have, not only have to be operating on these environmental conditions, but then they have to do with the low latency, the accuracy. On top of that, we also consider what's the business side of this. We don't play as many other games as other leagues. Our games are very intense and we set them up, but our stadiums often have other events in them. So the infrastructure that's applied for these purposes has to make sense from a business perspective. It also has to operate across a very large. So if we're providing wireless coverage or you're even providing other types of endpoint devices here, they have to be in a place that a lean team can maintain. They can pop out on a game day, on a Sunday, on a Monday, on a Thursday, Saturday during playoffs and, you know, retract, store somewhere in the stadium, come back out safely. So our sort of diligence around these areas has to be really strong. And the AI component is really just one piece because delivering whatever innovation or help in decision making AI have, all has to get down to the field, down to a person there in real time. Yeah. And I think that's the exciting thing is that you see it from the infrastructure perspective where it has to go through very, you know, intensely built stadiums with lots of concrete and metal and all kinds of stuff that typically, you know, stops the signal. And as a person who's gone to multiple different stadiums and, you know, use multiple different sets of Wi-Fi, because that's really up to some of those franchises, which Wi-Fi they use and some of that stuff. Where does the line drawn between you and the franchise is about how this technology gets implemented? Yeah, I'm going to make it even harder for you now. So I'm going to expand the scope and say we also have to recreate the same exact game experience that we deliver at these domestic stadiums internationally. We'll be in Frankfurt in the upcoming weeks, playing two games. We just came back from the UK. Yana fell. It's no mystery. People have seen on the internet. We're looking at Spain. We're looking at Brazil. We're looking at other markets to expand to. We can't decrease the amount of technology or limit what we do in those other markets has to work the same. So now the challenge is even harder. How do we create portable kits that can do these things? And it's really about using the best technology and leaning on these really fantastic partners. I mean, you have to look at what's public versus private services. And when we talk about wireless, if we are working with extreme networks on our Wi-Fi, Verizon on our private LTE and CBRS implementations we use for communications today or the public side, what is public on the 5G side that we might use? Now, all these stadiums, as as many know, in this industry, it's not inexpensive to place all of the radios and equipment that you need to cover the places that we cover. So we actually look into, I would say, a patchwork of solutions that are unified together, right? We might have to lean on one for the primary and fall back to another, depending on what market we're in. But it's very important to be agile, right? And to have multiple options across these. And then when we do build on a standard, like we built into CBRS with our coach communications with Verizon, that's an area we have to protect, monitor and manage with our partner and then try to expand the use cases of it. So we get the most out of one investment for us trying to go to another standard. Now you have to make sure that investment works. In London, Brazil, potentially Spain, potentially, it's quite a challenge, Rob, it's not easy. Yeah, and I'll give you feedback on Frankfurt when I get out there to see the patch, hopefully take down the cults out there and I'm excited to see that smaller stadium than we have here. But I think that doesn't change the challenges that you have or what the experience will be like for those fans, and including myself, I think I'm really excited about that. But I think also some of the things, it's about the data, right? And the data being consistent and being accurate. Because the last thing you want to do is put something out there that is inaccurate that changes somebody's perception of what's happening in the game. That has to be a heavy lift and go into the considerations about how you use the AI that you do have deployed. Yeah, it has to be deliberate and we have to be very diligent in the validation of the data coming in from whether it's machine learning or it's other forms of AI. So we're going to put these through our tests and we're going to put them through many different conditions of validation. On top of that, we're going to make sure it's the right fit. We don't do technology just for the sake of doing technology. We do technology could improve either the game experience for the fans or improves the operational experience. Does it reduce things that delay from the game, things that take more time away that are more of the official decision-making things that pause the game? We want to be accurate. We don't want to rush anything. But at the same time, we're not going to do something just to put a technology out there. It has to be right. It has to be impactful. And it seems like that's already, I mean, even this year with reviews and things like that, it seems like that technology is having impacts and enabling the refs to make quicker decisions on something in or out or what have you or where is the line to be and things of that nature. Is that the kind of, when you're talking about making the game better, is that what you're really aiming at? That's what we're looking toward. These things are all in development right now. So the other thing about the league is everything is reviewed and we have a lot of committees that look at these things. We have the competition committee. We have the future football committee. We involve coaches, clubs, owners, the league officials. Everyone is looking at us to make sure it is the right decision for our sport. Again, we don't put the technology out there just to put technology out there. We have to be sure. It also has to fit the workflow and the tempo of our game and make sense for the game itself. So there's a lot of diligence there. But I would also say the opportunity is very high. And a lot of the solutions we're looking at today not only serve potentially operating game but serve the fans. And some of the systems we've been testing, I know you could see the Toy Story Allcast we did. We did that out of Wembley, a stadium that's not one of our normal stadiums. And that Allcast was derived off of an optical tracking system that we also would use potentially for officiating enhancements. So there are a lot of opportunities here not just in one area, but to take the data and create it as a source to be used for our broadcast partners, for our over the top partners, for our Alps and for our fan engagement experiences. So it's a multi-pronged value proposition to us. When you look at it from and take a step back and to your point about having to go to Frankfurt, potentially Spain, in London, multiple, actually multiple stadiums in the London area, how do you really think and consider your infrastructure? You said you have to have it for the right size. It has to cost the right amount and get that ROI out of it. It has to be flexible. What is it really that goes into that testing and really looking into those for the future? I think the first thing we consider is the experience is consistent under the most amount of conditions or failures, right? So when you have that decision of, do you put something to the edge? Do you put something compute directly in that facility or do you rely on the cloud? And we might have multiple tiers and layers, but the first kind of challenge we do with infrastructure when we think about it in our thought model is, this has to work under all of these conditions. The stadium could become isolated from the outside world and if there is a game operational task that comes out of the system, it still has to work. And if it's a fan engagement experience, does it work inside the stadium with the same latency that it can work outside of the stadium? So the experience is consistent. And if the fan is a customer of all different carriers, if they have that experience, it has to be equal. So equal access. So we kind of start with reliable functionality and performance are there and then equal access. And then we start to look at how resilient, how robust this experience is and then how do you make it portable? So we're gonna make decisions about what we place, are replacing compute inside of the stadium, are replacing edge compute in there that's a little lighter but keeps the latency faster. Or are we looking at cloud because cloud allows us to be pretty nimble, right? We can operate the cloud essentially, move it potentially between hosting regions and then have a lighter footprint to bring things up into the stadium. So we are always playing with these factors to have the best outcome. But really, if you say our North stars are an agile infrastructure, an effective and powerful infrastructure and one that is resilient, those are things we really build around. Yeah, I think that makes total sense and something that we're really actually talking about quite a lot during this entire day today is the fact that it's more, how do you have a cloud operating model versus cloud as a place per se? And that cloud is where you bring it to your edge, to your data centers, to somebody else's data center for that matter. And is that really how you've kind of built your infrastructure out is to really be able to operate the same way no matter where you're at? Yeah, it really is. That's exactly that. Whether we're in Frankfurt, we're in Spain, we're in Brazil, we're in, you know, in the future potentially in other locations, we've also played games in Mexico City. So all of these locations or across the US, I mean, games could be relocated within the US for different reasons. These have to work, these solutions. So therefore, we can't say there's going to be a ramp up to build out this site when we might have to move the game to that site. Everything has to be able to be plugged in within, you know, essentially, I would say 24 hours notice to be able to work and operate. Now, there are obviously some things that, you know, take a longer term plan to build out, but that's the real theory that we use is that these are technologies that can are portable and come with us no matter where we go. Yeah, and you work with the teams, which is, I think, an added interesting, I guess, twist to everything that you're doing because some of the things you're considering, you have to take into account that these teams are competing against each other. And how are they going to perceive this as an advantage one way or another? How do you take that into consideration? I would always start with equity, right? We start with designing these experiences with one, equity and integrity. So the operation of them is transparent. The experience of them is equal. We are able to manage and monitor that. They're secure, right? That everybody feels that this is a secure and reliable solution. And then we're going to make sure that, again, it's something usable. You know, the technology, if it can't be adopted by every team equally, if it can't be used by every team equally, then why are we doing it? It really has to be something that all can use and enhances the game for all. And it creates a better experience. Again, we won't put something out there that we have to cover gaps in. We won't put something out there that it's just too far of a reach for our sport. You know, that it doesn't make sense if we're doing it just to have that technology said to be associated with us. We're doing these things because they have an impact that's positive to the game. Yeah, no, I think that's key to everything that you guys do and looking, being, you know, working backwards from the customers and the customers you have are not only the fans in the stadium, but they are the franchises that, you know, these multi-billion dollar operations that are businesses unto themselves as well. How involved do you guys get to work with those teams about some of the back office stuff and being like a service provider to them? It's a big partnership. It's an ecosystem. So very often it's obvious that we would invest together. And what we often find is we learn from each other. So the league may be pursuing some back office technology and that's at a different function level than a club, but there are some similarities that are lessons we learn. And then you have clubs who may accomplish some amazing things in different areas. Maybe one has a really great mobile app. Another has a really great, you know, broadcast experience. Someone has really awesome in-stadium technology, food and beverage ordering in-seat, but the main strength we have is not competing with each other in these areas, but sharing what's successful, what are strengths, what work, measuring the feedback of the fans across multiple stadiums. And then comparing those notes and improving everyone. The league, for example, issues wireless standards for Wi-Fi and cell. And we partner with our carriers and Verizon. We partner with Xtreme and Cisco and these organizations to improve those standards to make sure all the OEMs and all the providers are able to build to them. And then we consider the clubs because as you said, before each stadiums, architecture is different. So we have to go across all these different sites and make sure these standards work. But having a standard that's the derived raises the bar for everyone and gives everyone this sort of drum to pace to. And that's a key function of the league, is looking across and bringing the best ideas and the best practices into one circle for all of us to, you know, share from and learn from. Yeah, no, I think that's one of the keys is that you have so many different customers of what you do in some of the standards you're trying to drive. And I think it gives everybody that, you know, doesn't understand your business that, hey, it's the NFL, it's one big family. And I think it is a big family, but there's a lot of different people in that family that have different views and different viewpoints and different ways of, you know, producing revenue for their piece of that family. So it's... It's always a co-operative. Yeah, of course it is. We are, you know, the teams are competing on the field but we are cooperating as an organization and as a brand and as a fan experience. We want that to be great no matter which stadium we go to, which event you go to, that all has to be aligned. And technology is a massive part of that. Like I've never known, I've been here for 20 years now and there is more technology in the NFL than ever before and there doesn't seem to be a slowdown to that. Yeah. It's on the field, it's on the experience. I think that's a great lead into where do you go next? Where, what are you looking at? What are some of the things? Because I think a lot of times, other organizations, you guys kind of push the edge on how you actually use things. You're looking at various different ways. You know, I was on with one of the guys who's the head of AI for the Texas Rangers for instance and Doughton actually they won last night and ousted the Astros. I know not your league, but again, looking at how they use computer vision and very understanding and providing things and they're obviously a franchise within that and how they use that data. How do you look at the future of gathering all of this information and where it may lead to for the league? Yeah, I think one of the most important practices we have is seeding technology in advance, right? So again, not rushing to implement it, but definitely going strong into learning about it. So as Geno AI emerged, as computer vision emerged, we were looking at these things and comparing them to some of the workflows or solutions that we have and then we're mapping the opportunities. Where are areas that we either have pain points or areas we know we have for improvement? And then what are other leagues doing? What are other industries doing? It's not always just about sports because a lot of things we've learned or we implemented have come from totally different industries. And so we're looking, we always have our eyes open and we're very curious. And then we're gonna seed it and we're gonna test it. We've talked publicly before that, we've been doing line of game testing with computer vision. So they have the chains today, they come out and they measure for a fourth down. Well, that's one way to do it. And some people may say, all right, those chains have been around for a long time. We enjoy the drama and the excitement of them coming out and that measure, but is it the most accurate? Perhaps computer vision, it might be more accurate and faster way to do it. Don't pause the game for all of that time. Can we engage fans in a different way through that? Tennis has the three-dimensional renderings where the ball is hitting and you're seeing where it is. Can we provide the fan with more information about why that decision was made rather than trying to eye up what's going on there from the side? So that's something we've been testing this year and we've been learning a lot. A lot about what configurations of cameras work across all these different stadiums and environments and which are the most accurate. And then potentially getting that information down to a wearable and official on the field. And what connectivity is the best for it? Is it a private LTE 5G network? Is it a wireless network? Another short that goes across the field is a Wi-Fi. We have to learn all these pieces not just to get the AI right, but as we talked before, having the infrastructure actually deliver that is a whole nother part of it. Otherwise, the AI exists in the bubble out on a computer screen somewhere. It's got to get into the game. Absolutely. Now that I think makes so much sense and I think that one of the things that I love is that it is about the game and how you bring the game forward. And again, as a customer, fervent customer, those that are actually standing around me helping produce this today are wondering why the San Francisco lost last night. And I had Brock Purdy in my fantasy football league and he did not help me last night. But you start to look at all of that. I'm still waiting for all the Gen AI to help me do a better job with my lineup and help me understand that piece of it as well. And I would assume that that goes also beyond the field for what you guys must be looking at is how to keep the fans in between the games engaged with the game. Yeah, yeah. Yeah, and not only that, just at our events, you know, you have the games, tremendously exciting games. We have draft, we have what we do during the times between with training camps and other things. How do we create this ecosystem year-round where the fan can engage, get informed, get information about what's going on, but also play on their next attendance. So, you know, we've even looked at Gen AI in those areas of help me attend the game in Frankfurt, help me attend draft in Detroit. So help me build a schedule on itinerary for these things and know a little more about me as a fan. Know that, you know, I have these preferences. I'm gonna bring my kids and my kids are of these ages. I'm a Patriots fan, but I really wanna be a Bill's fan. Know all these things about me, right, Rob? So, you know, that's where there's opportunity here to personalize the engagement. We have a data team that's been working on it doing a tremendous job, but AI is gonna unlock some of those opportunities. And we're pretty excited about how that increases, you know, that experience, right, that the fan wants. Yeah, and I think what's interesting is people probably don't think of the NFL as a data company, but a lot of your intellectual property is in that data and how that data gets displayed. I mean, obviously there's the broadcast aspect of it, but there's the data that goes along with that as well and how that plays out. Yeah. How to bring that. And there's gonna be new category. I mean, Next Gen Stats has been groundbreaking in this area of you have the traditional stats, but then you had these insights of a combination of multiple factors happening on the field, right? And that came from the sensors and the positions and saying new insights were unlocked, right? And Amazon helped us unlock those insights given these positions, given this situation in the game, here's something that you should know, you should know about the offense, you should know about the defense. I think that AI unlocks more opportunity on the future. And as we start to implement optical tracking and other sources of data, combining these data sets, you're gonna get even deeper insights about what's going on in the field, probably new things to enhance your fantasy game and certainly new experiences as a fan. I mean, I really look forward to what we can do with AR, VR in the future by capturing the entire game from all angles. Yeah, no, I think that to me, that also makes a lot of sense. And I think that this is probably a good place for us to end it here, end our conversation because you and I can talk forever about this stuff. We find it, we find it. So, Aaron, I wanna thank you for coming on board today. I think again, helping people understand a little bit more about their favorite game in the National Football League and what you're doing. I think again, it's just super interesting to see how you guys really push the edge with AI to really help fan engagement, which I love. Thanks, Rob, appreciate it. And thank you all for being with us with SuperCloud 4 here, live from Palo Alto. We're gonna be back next. Dave and John are gonna be back on here with you and we're gonna be talking to some of the generative AI startups that we're gonna be in there making new stuff and helping you deliver value to your customers. We'll be back right soon. Stay tuned.