Episode #14: AI in Professional Sports

Welcome back to another episode of the Big Cheese AI Podcast hosted by Sean Hise, filling in for Andre Harakas who is attending a leadership conference. Joined by co-hosts Brandon Corbin and Jacob Wise, the episode takes a dive into AI and sports with their first guest, Carl Ceresoli from Pacer Sports & Entertainment.

Show Notes

Big Cheese AI Podcast Show Notes:

Topics Covered:

- The ethical implications of using low-wage workers in training AI models, with a discussion on the dark side of AI and how it can resemble modern sweatshops.

- The new features being added to Google Chrome, such as automatic tab grouping and generative AI features for content writing assistance.

- How AI is changing the landscape of gymnastics judging by adding data-driven evaluations that could reduce subjectivity in scoring.

Relevant Quotes:

- "AI is this beautiful shiny thing and then you kind of look and you go, but wait a second. That doesn't seem about right."

- "The people at Pornhub have to actually watch video. They have to verify that the stuff that's coming in is within their respectable range. But with this, there's not video."

- "The net is that everything ultimately points towards some element of paid, right?"

- "AI is not just this black magic box, right? You have to have training data. But you have to reinforce that training data with real-world examples."

- "You're not flipping burgers anymore. You might be training AI algorithms."

- "You're literally just sitting there and just be like, yeah, that's good. That's bad. That's good."

- "The NBA summit was predominantly AI... focusing on things like injury prediction, player performance."

- "We're seeing a lot of AI that I don't think the industry is really ready to talk about yet."

- "Pro Sports is a very high-profile industry. We're a local company, but we have a global brand."

In conclusion, the episode explores both the potential and the challenges of AI in professional sports, discussing how AI can transform the industry both through back-end services and as a tool used in the public-facing aspects of sports. The conversation highlights the need for considering issues like privacy, training, and how AI may impact contracts and player evaluations. Carl Ceresoli, with his extensive experience in both technology and the sports industry, provides insights into the future of AI in sports, as well as the innovative developments happening at Pacer Sports & Entertainment.

And all of the variables that nobody had any idea were there. We're like, wait, what? AI is this beautiful, shiny thing. And then you kind of look and you go, but wait a second. That doesn't seem about right. That doesn't seem right. [MUSIC PLAYING] Welcome back to the Big Cheese AI podcast. My name is Sean Heiss. I am hosting again today as Andre is out at a leadership conference. I'm excited. I'm kind of getting used to this role. You should take it. I like it. Sorry, Andre. Dre is usually sitting here. But Brandon's sitting next to me, so I'm going to have to-- I can watch out. This dude's always moving around and slapping his mic around. So I'm kind of nervous about that. We also have Jacob. Or how you doing, buddy? Hey, what's up? And he's also wearing a Pacer shirt today because we're excited because Carl, Sara Soley-- Our first official guest. Yeah, right out of the gate. Cheers. Nailed it. Our first guest, and he got his name right. I've been trying to-- yeah, I assumed it was-- I tried to put my Italian thing on here. I've never said it correctly until just now, so I'm excited. I've known Carl. Jacob and I have known Carl since-- It's been a couple of years, actually. And then COVID, so it's even longer. Yeah, exactly. COVID, the world froze. We worked a little bit together when Carl moved. And we'll get into that. But we're really excited to have you on, Carl. And we're going to have a really exciting conversation about AI and sports today. But before we get going on that, we've got some news updates in the AI space, starting off with an interesting article that we found while doing some research for the show. And it's something that you might never even think about in AI. So Google's got their model, BARD. Yeah. I don't know. Maybe the worst name ever. I know it is. I don't know. Do you use it, by the way? No, I don't use it. Jacob loves BARD. I'm not the only BARD user here. OK, yeah, yeah. But this made the news. And it's one of those articles, you're like, OK, great. And then you start scrolling, and you're like, oh, I see the story here. So Google reported-- or actually, CNBC reported that this company, Appen, A-P-P-E-N, lost their contract unexpectedly. They're an Australian data company with Google. And basically, this company was providing services to help train BARD. And so they provide a tremendous amount of service. But they're doing it with, I guess, really low-paid wage workers. And they're basically the people that are testing the model before-- the large language model before people like us get access to it. And so it opened up this whole conversation about Accenture, who's providing 8 an hour. [LAUGHTER] So maybe the next-- you're not flipping burgers anymore. You might be training AI algorithms. But man, I don't know. Out of all the shitty jobs that you can have, that one seems like it might not be that bad. You're literally just sitting there and just be like, yeah, that's good. That's bad. That's good. I've actually built a few different tools for an insurance client of mine that lets basically just people sit there and just a photo comes up. And you've got to go highlight where the damage is. And you just click, click. And you say what the damage is. Click, click. And that's all your job is. I think we're going to get to a point where you're just going to have entire armies of people. That's all they do is they just click, identify. Click, identify. And they become kind of like these human intelligence tasks. Their only purpose is to ultimately feed the artificial intelligence. Which will ultimately replace them. Yeah. Yeah. But I think that that's also-- it's an interesting thing. Because most individuals' assumption with AI, at least how it's sold, is that that is an automated process. Right? That that's this. But the reality-- again, that's another layer down. It's beyond the shine of AI. And then you realize, wait a second. These models, these things that everybody assumes are happening like this. And it's, oh, the AI is wrecking-- no, no, no. Kind of. Not exactly. Maybe. Maybe in a couple of weeks once, or maybe in a couple of years. But there's still a human element in there. And that human element, depending on who it is-- and especially with this article-- the thought is that that element is adding a variable into the equation. Totally. And it's not controlled. It's maybe not as accurate as thought. Well, one thing that kind of blew my mind was-- so when we talk about a lot of these open source models that are being released, there's like releasing the model. And they're like, well, it's not comparable to ChadGBT. And what people don't realize is there is-- ChadGBT, yes, is a model. But there is layers and layers and layers of process and rules and guardrails that make it function the way that it's supposed to function. And only those came up because some human was sitting there basically doing that kind of work. So yeah, so check it out. There is multiple sides to AI. For our next article, we had an interesting development, which is basically the next version of Chrome is going to ship with generative AI features. So you've had the concept in Chrome of grouping your tabs for a while. I don't know if anybody-- some of my employees and some of the people I see, I see, oh, you're grouping your tabs. Good for you. I'm still like, pshew, all across the screen. Do you guys group them? I do profiles. Yeah. Because I work on-- Oh, you do profiles. I do profiles. But no, I don't group them. No, I don't group them. And that's just-- I don't do manual stuff. So now Chrome's coming in, and they're saying, OK, we're going to group your tabs for you, which I think is a good example of, hey, let's release this feature. It's been out for a year or two, maybe, probably longer. I just never used it. But now they're saying, hey, we're going to do that with AI. Let's do it for you. Nice move. So that, and they've got basically help me write, so generative content. The writing-- And some text boxes. OK, so that one absolutely makes all the sense to me in the world. That's an actual practical feature that makes sense that people will actually leverage. The tab thing, whatever. Do you think that's the gateway drug where Bard gets a lot more market share because everyone uses Chrome? And what model are they going to use to do the generative AI for your input, your text box? Do people care? Right. The question is that Jacob's posing, though, I think is, who's paying for this? Yeah, it's expensive. So is this a gateway drug to Bard's paid model? And like, or is the future of AI, like generally this free, please use Google, please use Microsoft? I guess that's a question for Carl. Like with Copilot and all the stuff that's going on, like with trying to build these tools into like office products like browsers, and is there-- The net is that everything ultimately points towards some element of paid, right? I mean, free is great. Free is the drug. Free is the thing that they put out there. But the net is that the alignment is to ultimately generate revenue somehow from this. And that's the component. Whether it be-- and I hate to use it, but because I'm so fatigued by the word subscription. Everybody wants you to subscribe. Another $9 h ere, an o t h er$ 20 here, another-- but that's the goal, right? How else does this survive? How else do you fund these types of components efficiently unless you can push people? And ChatGPT is a great example of that. There's the free version, which does some neat things. But hey, you really want to do the cool stuff? You really want it to do this? ParseError: KaTeX parse error: Expected 'EOF', got '&' at position 14131: …hat was with AT&̲T. I've always…0.50 that he's going to make that one shot? So micro-betting is particularly interesting because it requires a lot of data to function. So the concept in general, just to review it real quickly, is you walk into any arena or a pro sporting event, and you have whatever app it is that that league requires or arena requires. And when you walk in, the app knows you're there. And it says, hi, so-and-so, welcome to Bob Law. You already have your profile all set up. And it said, would you like to participate in micro-betting? And yeah, OK. So then you're sitting in the stands. And AI knows you and knows what you like to bet on. It says, do you want to right now place $0.50 so that when they inbound the ball, it's going to bounce once, as opposed to twice? OK. Or that they're going to pass from this person to that person. Or this very specific, unique thing is going to take place. And if-- it's obviously more complex than that. And there are a number of different variables involved for it to happen successfully, specifically when it comes down to infrastructure, not only just the AI, but the infrastructure in the arena to allow that type of low-latency type of event to take place. It adds another element of fun. And it's very unique. So you're talking about really almost like an in-person retail aspect of betting, not just your typical fan duel, in-game, the spread-change stuff. You're talking about pattern recognition betting. It's very, very specific, very unique to you. And you might be sitting here in three seats over, somebody might be on the same application, and they may get a different micro-bet than you. How are these-- whoever's facilitating the bet, getting their odds? I imagine that would be very difficult to understand the risk of, OK, if I ask how many people can fit in the stadium, 16,000, 17,000? So-- But that's the magic of AI. That's all just magic. And it just happens in the real world. [LAUGHTER] And nobody really knows. Why didn't you tell us the magic? We were looking for it the whole podcast. There it was, right? It's just, ah, whatever. Well, I know if it's Andrew Nembhard, he's going to bounce it. It's going to bounce twice. Yeah. Every single time he brings it up. Every time. I'm like, just go with it. You would get that bet. Yeah, that's that. And that's one of the more interesting concepts that's generally out there. I think that they've tried it. And it was fairly successful, I think, last year at MOB All-Star, one of these events. They've tried it in different arenas. But as I said, here's, again, one of the pro sports AI. When you look at just the general concept, part of my job is when these entities come in and it's a fan-facing concept, it's not only is that tech sound or can we actually get there? It's to take that and say, OK, this thing, this concept requires infrastructure to run on. And when you get into these stadiums, when you get into these event centers, that varies wildly. So to really have a shot at pulling this off, it comes down to, at the end of the day, latency. Everything comes down, especially with real time-- with AI in this fashion. It's all about latency. That reminds me, sorry. But I coached wrestling for a long time. There's this app called Track Wrestling. And the number one problem they had-- it's real time score update or tracking the actual match and putting it out there to anyone who has the app of what's going on, what's the score, all that stuff. And they're doing it at a computer. But most of these events happen at schools. And schools, while they have fairly good Wi-Fi, they're not-- it's not great. And it's different between that school and that school. So that was the number one feedback I always heard was-- Latencies. Yeah. I mean, Wi-Fi, channels, because it's incredibly complex to do that at scale, to provide that type of environment at scale. What we're talking about is much different than all the conversations we've had. You're talking about a model that's being trained on these huge GPUs, on these unbelievable data sets. It's not real time. Now, sure, ChatGPT can go to Bing for me, which I do use that. But you're talking about real time analysis. And that's what I think is an interesting distinction. So a lot of the tech that's in the NBA stadiums is based off cameras right now. Yes. Right? And so they're taking cameras. And they were doing that. They've been doing that for much longer than even other sports. Wouldn't that-- hasn't some of that basic infrastructure been around for quite a while? It's been around for quite a while. It started a number of years ago with Intel doing the experiment, and coming in, and putting all of the cameras in. And they did NBA, and I think NFL, and some NHL arenas. And that was more-- and they did that on their expense to kind of see what was possible. So a number of years ago, if you were watching an event, a basketball on television, and it was nationally televised, they did the-- and it would come after they came back from a commercial break. They would do the, here's a shot. And they would swing around. So that was done by the system. There wasn't necessarily AI. It was just video processing. And the images were being built with AI. But that was all just video processing. But that was just to see what was possible. It's advanced several generations, and very, very quickly from that. Those systems are no longer in operation, but there are other systems that have been deployed that are fully operational. So who's making that investment? Is that the NBA that's making that investment? Sure. It's a combination of the leagues that are paying for that investment, as well as the tech industry and a couple of key companies that are bringing that element into the equation. So from an injury prevention perspective, what are they looking for? What types of patterns? Is it just they just figured out a way to analyze it? Or is it more like stats? Is it more-- is it the camera stuff? Or is it just raw data stuff? So it's not necessarily what happens in the arena. So they now have sensors in the gear that they're working out with, and the exercise equipment that are measuring all of these variables. And as well as-- Sorry, real quick. They've got cameras in the workout rooms as well? No, not in the workout rooms. Oh, so biometrics? Yeah, so it's biometrics, as well as embedded sensors in the gear that they're using. Oh, so the actual machines that they're using. Yes. And this was, just to be clear, this was at-- this is not necessarily tech that's deployed. This was at being highlighted at the event. So it was a number of different companies that were there talking about, well, if I look at all of these variables, and I know the human body, and then you input all of this information on player XYZ, and there are various conditions, and I run you through this MRI scan that looks at these things, I can then predict certain things that you're doing. And what they found is that they can determine the difference between muscle groups, and even left leg versus right leg, and players that choose and favor certain muscle groups. And if you were to change your training to eliminate this bias, this unknown element, you can improve your performance by looking at all of these various things. That's really, really interesting. It's very, very complex. But I think we're still a couple of years from it really translating from, again, a controlled presentation to something that your trainers could operate. I mean, there's that element, right? When you implement it in the real world, you don't have a whole tech organization there to facilitate that every time. This reminds me of a couple of our conversations we've had before of, we would look at this tech, and we'd start the conversation with, who's going to lose their job or whatever? But it's interesting, because it sounds like a trainer could do their job better in this case. That's the goal. Yeah. And definitely, LeBron's going to sit on the second night of a back-to-back, but that's just probably-- that's built into the system, probably, right? Well, what level of accountability do you really have? You're a professional athlete. Your job is to take care of your body, right? And the training staff's job is to help you do that. The devil's advocate here. Let me throw-- and this is not necessarily-- I'll put this out there, but-- so you're a professional athlete, and there's all of this information out there on you now. Your contract's coming up for renewal. Yeah. I was just thinking about the social credit score for the NBA. Is DeAndre Jordan on the Nuggets really going in, getting in the low-weight room at 4 AM? Is he Kobe? No. We probably not. Maybe he's so mad at me right now. Sorry, dude. But yeah. If I was the NBA PA, I would be very concerned. So yeah, and this is true for most pro sports. And I'm speaking not just for the NBA. In no way do I represent anything. Yeah, nothing-- Carl does not represent any thoughts of the Pacers, Sports and Entertainment, or the NBA. But when you get to know and understand how these industries work, there's two sides of the equation. There's the players' associations, generally, with most of these leagues. And the players' associations are there to look out for the best interests of the athletes. And you have all of this information that's about to be dumped into that process, and these variables that teams want to look at to assess value that could be potentially brought to this side of the organization. Well, that value will now-- it's a variable, right? And if you're looking at this, and AI says, well, we're seeing all of these things. But the reality is it's not part of the equation today. And it could be tomorrow. And that is something that seriously has to be looked at, not only from a privacy perspective, but the downside, the negative consequences of that. Yeah, so you're talking about health data, about a person being shared with the person that's doing payroll and trying to figure out the next contract. So is there a wall up between the training staff, potentially, that they might write into the agreement? Right, and those walls are only as thick and as-- right? That's what the players' associations would say. Well, there's a wall up, but the reality is-- And just like the gymnastic situation, an athlete might have had a dip in their performance from a workout perspective or whatever. But maybe his brother was in the hospital for a few weeks. Or he tweaked his heat, or just like-- There could be penalties associated with that. If I'm a gymnastics and I miss several training sessions, or I'm not, AI would be able to determine, hey, are they giving 100% in practice? Well, according to what it's looking at, the answer is no. But it doesn't take into account why. It just says no. And the repercussions of that, well, that immediately triggers penalties for this, this, and this. I think you said it earlier, we're getting past this phase now. But everybody thinks AI is magic and also flawless in its assessment. And it's only as good as the data you put in. This is just another real world example of that, where it's like, it can give you an idea of what's going on. But it's still up to you. Whatever application you're using it in, still up to you. It's your responsibility to understand what does that actually translate to? What were the other variables it couldn't possibly have understood? I was reading this article from-- it was from 2020. And it was about an analytics employee from the Atlanta Hawks. And they were talking about AI in the NBA. And it was really the machine learning stuff they were doing with the pattern recognition. So they were doing new stats that never existed. So they used to have points-assist rebounds or PER, just basic stats. But they were doing more like, what's your pick and roll rate? And what's your-- the stuff that was derived from cameras and pattern recognition. And they were talking about-- he was talking about how their models predicted that LeBron and D. Wade going to Miami would have no impact on the success of the team because their roles were too similar. And they took bad shots. And I'm like, well, that's the AI element. But the human element shows that they won-- what? I guess they only won two championships. You said they were going to win not six, not seven. Wait, you're going to be 50, LeBron. You can't keep counting up. Two is more than zero, though. Yeah. But more than zero, right? And they didn't understand the human element of that. It's like-- It's LeBron James and Dwayne Wade. They're going to be able to adjust their game. Or Eric Spolstro. Yeah. Or just like the vibe that that team-- the other players, they had. Right? And so bad take, bro. But you just looked at the data. You didn't look at what it could be like if a human says, well, I need to take less shots because LeBron James is the best player in the world. Right. That comes down to, again, with the humans in the equation, the scouting staffs will tell you. That human element-- and if you look at-- this was a while ago. This was before AA. But the movie Moneyball, right? Yeah. We were just talking about this. You had the scouting staffs sitting around the table going, you're never going to win. The reality is it's more of a balance. It's another data factor. But it's not the only one. And a lot of times, it's by far-- there's a number of elements that far outweigh that one factor to determine success. Yeah, the Moneyball stat that's most impactful you have to learn from is the ugly girlfriend stat. [LAUGHTER] What? Yes. Oh. He goes, great hitter, ugly girlfriend. That reminds me. I was reading this book. And they were talking about predictions later in the season, specifically for basketball. But it was teams that aren't very good in the beginning of the season. And then they give a lot of-- if there's a lot of butt slaps and lots of camaraderie, is generally an indicator for future success. And I'm like, they're doing this manually. They were just literally watching videos and saying, that team is giving more high fives to each other than that team. And sure enough, at one data point, but I'm sure there was more. There is no doubt, pro sports, sports in general, it's a very emotional task. It's a very emotional event. So those elements, that AI is pretty binary, ones and zeros, they tend not to translate and capture components like that. Superstitions, all of those things are real things. They're real elements of the game that I'm not sure-- well, AI, I know today, there's no component of AV in the future. But right now, when a batter walks up to the plate, there's the tap, tap, tap, tap, tap. There's those elements that derive confidence that help them feel more at home. That it's not a one or zero thing. Or an AI would start telling you, hey, you need to tap your shoulders, do, do, do, do, do, do, to make sure you get that hit. You missed one. Watch, he's just straight out. So what's the conclusion? It's 2024. AI is taking over. But what is the future of AI in sports from your perspective? Is it a backseat? Is it in the forefront? Is it transformational? Or is it just something that's just there? So I mean, it certainly has the potential to be transforming. Depending on who's selling it, it has that potential. But I think right now, today, it's something that's present. And everybody's just searching to go, OK, well, this is here. And the obvious pieces are gambling. It has the potential to generate a lot of real time stats to feed the casinos, to feed the online apps, and things of that nature. There's those elements that are relatively obvious. I think the other components are still-- we're still a little ways away. I mean, they've still got to play out well. There's the piece that I always say, you can have the best bit of tech in the world, but there's the training element that comes for the staff. Because you still have the physical therapists. You've got all these people that-- they're not going to lose their jobs. They just need to be up leveled on the training now to incorporate this into their process for evaluation. And that's not going to happen overnight. So I'm not-- again, I'm not very good at the sports, as we would say. But so at a-- You do have some chicken legs. Hey, now. From a betting side, you've mentioned it multiple times now, that the betting play. Is there-- does the NBA have a-- is betting going to be a thing? Or do they have to keep a separation of it? It seemed like you'd need to, right? Yeah, we-- that's being driven by the number of different partners that we have in that space. But you've seen-- even ESPN has now. You've seen mainstream huge media companies are now-- They're leveraging-- so we have the ability to leverage stats. I mean, that was-- we did a project. I did a project on that where we did projection mapping in the real time. We took the stats feed and put that in the arena. Yeah, we saw the real time NBA stats. I mean, every single second, we're getting these boom, boom, boom data feeds with just any single thing. And those are getting consumed. And they're different properties are getting added to those objects all the time. So I mean, that's a real thing. But we're in the business of generating those stats and then putting them out to partners and then what the NBA wants to do with them from there. I can't necessarily speak to that. But it's all about-- it comes down to-- that's kind of the end. There's all of this work that goes into insuring. Because once you say that you want to use these for-- and then there's money involved, you've got to be accurate. You've got to have all of these other things lined up. And they have to be bulletproof, right? Oh, gosh. So because-- and I think the classic example of that, and I think this within the public domain, was this started in arena. Gambling started a while ago, right? A number of season-- a number of different companies are trying something. And they were running an app. They were running the app in an arena during a live event. And they were looking at the stats. And they were allowing actual wagers to take place. And I don't know if you remember, that was the bottle being thrown in during the follow shot. So-- Oops. All of a sudden, they saw this huge spike. Nobody knew what was happening. And then they went to go take the phone some way through the bottle. And it was all set up, right? So at that second, a second after that happened, everything stopped. Because you can't control those elements. And that's the difference between that and micro betting. You don't have that component, because it's-- Right, instant. Yeah, so there are a lot of challenges. It all sounds great. And like I said, you can put together a hell of a PowerPoint presentation on, oh, it's going to be awesome. And these things are going to happen. And you can do this, and this, and this. It's all going to work. My biggest takeaway from today is that Carl's seen a lot of AI PowerPoints. Yeah, and they don't work. So I have to ask, just because-- so I worked at a company called X-Ray Glass. And the whole premise here is for the deaf and hard of hearing to basically be able to have subtitles of whatever is going on. So this conversation on having subtitles, right? They're starting to see that there's potentially a draw into events, into games, and like this. Are you guys doing anything currently? Yeah, that's great that you brought that up. We just actually finished an app. And we worked with Microsoft on it. So Microsoft brought in their team. And they actually did it just to assist with hearing impaired and people with visual challenges. So the app is during a Pacers game. You can go into the arena, launch our Pacers app. And then there's a button during the game that says if you want to press this. Yeah, and the traditional closed captioning is a system, and it goes off. And it's very-- there's been very little innovation in that space for quite some time. I need to watch that system at some point. But we worked it. And hats off to-- yeah, hats off to Microsoft. They actually use-- it's the chat GPT engine behind it. Oh, really? So in real time during an event, the broadcast-- not the broadcast, the MC in arena, everything is closed captioned. And it's real time. And that model is being trained as we speak. But it's getting more and more and more accurate. And we hope to, in the next little bit, have a button on there so you can change to different languages in real time as well. So if you want, X-Ray can already do that. Yeah, so we can do it as well. Props to X-Ray. It's available, I think. We just haven't enabled it. But it would be available, French, Mandarin. Is that all through Microsoft then? It was a project that they helped us do. Oh, but it's your product. Yes. It's running in our environment. We asked them to help collaborate. So you're the only-- is there other NBA teams doing the same one? It is available. So we will make it available. It wasn't done-- we're not going to drive any revenue from it. It was done because we saw the need for individuals coming into arenas that are asking for these types of services. And there was a gap in what existed today. So we developed it with them with the thought that, hey, we get this, and we get it right. Go out to everyone that wants to enable us. It's very low touch. It's very lightweight. It just requires-- I mean, everything's running in the cloud, and it comes back down in the app. Awesome. Well, props to you. And then for all the young engineers out there that are listening, accessibility is very important. And you should take it seriously because-- Yeah. Huge market. It is a huge market. There's lots of opportunity there. And it was a space that kind of lagged for a while. But now you're seeing a lot of different elements happening within the arena space to extend that experience that historically may not have been as fulfilling. And accessibility is just good usability. Yeah. That's what we found. I mean, that's why Gen Z watches Netflix with no volume, right? It's just-- With subtitles. Yeah, with subtitles. I just sit there with them muted. When you make something accessible, the applications are endless. It's not just-- I mean, sure, it's really great for people that absolutely need it. But it's something that is-- it just becomes more useful than you ever thought it would be. Right. And that translates to-- this is a great example. All-Star is an international event. And there will be people in the arena speaking all types of languages. And this app's going to make that experience even better. Yeah. That's awesome. We talked a little bit about this a few weeks ago. But just the intersection of good and good for business. And accessibility hits that mark very well. It does. And AI especially is very good at achieving some of these things that weren't possible before. We use a text-to-speech tool a little bit. And if you've ever used the auto speech readers on a website, they suck. I'm super excited for anything involved in that space, because it's just easier to consume the information. One of the challenges, though, the funny thing is, is that, well, the tech is there. The ability to do this is fairly-- it comes down to the infrastructure then. So your MC's mic and how they're speaking into the mic, and when they yell, and the distortion-- [INAUDIBLE] All of that. So we tried to use it-- [INAUDIBLE] We tried to use it during a concert as well, because it's the same thing. We can take any feed in the arena. We've got control over that. So we got the artist's permission to be able to take a feed from the stage. What happened was-- I won't say who it was, but they were an older artist. And the people that were there knew all the words of the song. But what he was saying may or may not have been anything close to what it actually was. Sounds like Molly Crew. Yeah. [INAUDIBLE] Yes, it was that. It was that. It was that. And I was just like, I have no idea what this guy's saying. See, you've got to have that human element. That's why there was a closed captioned guy on the Commodore 64 of-- For now. [INAUDIBLE] Yeah. So it was pretty fun. Well, this has been, I think, great. It's been awesome. It's a great big change. It goes quick. It goes so quick. It's been an hour. Has it been an hour? It's been an hour. And I really, really thank you. That was awesome. I think it was-- we covered everything and more. And I think that this foray into having some unique perspective is going to be good for us. So for all those who don't know anything about Indianapolis, that's where we were based out of. The Indiana Pacers are playing basketball in the best arena in the world. It's basketball, right? It's sports ball in Indianapolis. We get to sports ball. And one of the greatest events is the NBA All-Star game. And that's going to be held here. And people wonder why events like that get held in Indianapolis. But it's because Indianapolis puts on events like no one. I grew up-- my dad worked at a hotel downtown when I was a kid. And so I grew up in that. It's going to be a hell of a show. It's going to be a great show. I can't wait to see what you guys are putting on. I love the NBA All-Star jerseys with the pinstripes. I saw this morning. And there's going to be a great Hoosier feel to it. Now, let's hope we get the same weather that we got for the Super Bowl when the Super Bowl was in India. Yeah, right. And yeah, if not, I'm sure they'll have a plan. So check us out. And a backup plan to that plan. Yeah, exactly. Nobody does events like Indy. So thanks, Carl, for joining us. We look forward to great success for you and the Pacers. And we'll see you soon. Thank you. [MUSIC PLAYING] ( special thanks to everyone who helped make this all possible! ) ( special thanks to everyone who helped make this all possible! ) (upbeat music)

Social

Show Notes

Subscribe

Let's Connect