Accordion

• Transcript

  Melissa Harrison (00:07): 

  Welcome back to Inside Innovation, a special edition of CES Tech Talk. I'm Melissa Harrison, and today we're looking at one of the most dramatic shifts happening in tech right now, how AI is transforming travel from a world of clicks, searches and forms into a world of conversations. 

  (00:23): 

  And considering that every January, the entire tech ecosystem packs up and heads to Las Vegas, tens of thousands of attendees, exhibitors and partners all moving through airports at once. Anything that makes travel smoother or smarter is something this community cares about deeply. Because for decades, booking a trip has looked basically the same. Type in some dates, scroll endlessly, hope you didn't miss something and cross your fingers when a delay hits. But a new wave of AI is rewriting that entire experience, and Sabre is emerging as the first mover. 

  (00:54): 

  To help us understand what's changing and why now, we're joined by Joe DiFonzo, Chief Information Officer at Sabre. Joe brings nearly four decades of experience leading large-scale technology transformation at companies like Syniverse and Convergys. Today, he oversees Sabre's global engineering and IT operations, driving the modernization of the company's technology foundation to deliver greater reliability, agility and performance for customers around the world. 

  (01:20): 

  And Sabre is doing something big. They recently launched the industry's first agentic AI APIs powered by their model context protocol, a kind of universal translator that lets AI shop, book and service travel in real time. It can rebook disrupted flights, confirm hotel check-ins, handle visas, file expenses and collaborate on complex itineraries automatically. All of it sits on top of Sabre IQ and the company's massive travel data cloud, over 50 petabytes of intelligence built with Google's Vertex AI and Gemini. It's the foundation for a future where the next travel influencer might not be a person at all, but a conversation. 

  (02:01): 

  So today we're diving into how travel is shifting from static search boxes to dynamic dialogue and what conversational commerce means for airlines, agencies and travelers, and how Sabre is balancing speed, trust, transparency and scale as AI becomes more autonomous. 

  (02:18): 

  Joe DiFonzo, welcome to the podcast. 

  Joe DiFonzo (02:21): 

  Thanks, Melissa. 

  Melissa Harrison (02:22): 

  Joe, you've spent nearly 40 years leading tech transformations at several companies, including the better part of the last decade at Sabre. While it's not a household name, Sabre touches almost every part of consumer travel. Can you tell us a little bit more about what Sabre does? 

  Joe DiFonzo (02:37): 

  Sure. Sabre is, if you will, the transaction engine behind the travel ecosystem. Think about this. If you book a ticket on an airline, if you book a hotel room, if you shop for airlines or hotel rooms or rental cars, we provide all the systems that run behind those websites that you use. So we're not directly interfacing with the customer, but basically all of our customers are, and we provide the glue that binds that travel ecosystem together. 

  Melissa Harrison (03:07): 

  I love it. So we would love to talk a little bit about why today is such a pivotal moment for the travel industry and just how technology is leading that. Tell us a little bit more about how Sabre's uniquely positioned to lead the change that we're in right now. 

  Joe DiFonzo (03:22): 

  Sure. I can do that. Sabre's been in this industry for a long time. You might say Sabre really invented this industry because we are really the first SaaS platform for the travel industry. We're going back into the 1960s and early 1970s now. Much of this technology was invented by Sabre and we've basically been keeping it up-to-date and modernized. So the last few years though, we've really gone on a tear, upgrading our technology, migrating everything to cloud, building the partnership with Google and basically accelerating our adoption of all kinds of new technologies, including AI technologies, but also various other cloud technologies, big data technologies and other things that are very relevant to the business. 

  Melissa Harrison (04:09): 

  Sabre has really been on the forefront of agentic AI conversational commerce, maybe terms that our listening audience isn't as familiar with. Can you define what those mean in your context and how they differ from the traditional online booking and digital retail that we're all used to? 

  Joe DiFonzo (04:26): 

  Sure. Let's start with the AI part. This is really moving fast. So I'm going to give a little bit of background. If you want to think about it, original AI technologies were really statistical exercises. We would examine large amounts of data and we would do that in order to build programs that could make more accurate predictions about what to do in the future based on that data from the past, so how to optimize the price of a ticket or how to optimize the route of an AI line, for example. Now we have what's called generative AI. 

  (04:58): 

  The difference between the traditional or classical AI and generative AI is generative AI actually creates information. So this really came into prevalence probably late 2022 when you started seeing the first generative AI launches with tools like OpenAI's ChatGPT, where people actually could ask it questions and it could generate output. And then they started using it to generate images and generate videos and all kinds of interesting content. 

  (05:24): 

  Agentic takes this to the next level because we now have gone from what were called large language models, which would actually create this output from a given input to things called large reasoning models, which actually then are recursive large language models where the large language model will actually think about a problem, break it down into pieces, how to solve that problem, and then go back into the large language model or different ones and solve each of those pieces and then give you the aggregated answer. That's what that's about. And now with the agentic part, it's really interesting because all of that previous stuff was done with the data that the large language model or large reasoning model had within itself from its training. 

  (06:06): 

  What agentic allows you to do is actually use data from other systems. So the large language model or large reasoning model actually could use these other systems and they actually call it as a tool. And MCP, model context protocol, is basically the connector between the large reasoning model and that other system. So now you can have up-to-date information and you can do actually transactional things like book a ticket or make a payment or that sort of thing through the large language model, through the generative AI platform. So now you can use tools like Gemini to say, "Oh, show me all the flights and book me a ticket," et cetera, and connect through to systems like Sabre to actually do that work. 

  Melissa Harrison (06:51): 

  And the idea is that at some point it will know me, the customer, so well that it'll also know that I prefer an aisle seat or a window seat and that it'll start booking according to my preferences. 

  Joe DiFonzo (07:02): 

  That's right. The key here is context, right? So you want to think about all of the context of the industry, what are all the airlines, what are all the hotels, what are all the flights, what are the rooms available, that sort of stuff. But then also context is your preferences and your history. And oh, when you fly to Los Angeles, do you usually like to have an aisle seat or do you usually like to have an extra leg room or do you usually like to stay at the Hyatt instead of the Hilton? All of these things that can remember and now you can just say, "Hey, book me my regular weekly ticket to Los Angeles," and it will do mostly the right things or hopefully everything correct and you might only have to give it little bits of input after that. 

  Melissa Harrison (07:41): 

  As someone who travels a lot, I was really interested in Sabre's recent white paper, which was, it's called Chat as the New Influencer. It suggests that the next travel influencer isn't actually a person, that it's more a conversation. What does this shift look like for both travelers and brands? 

  Joe DiFonzo (07:58): 

  The interesting thing about this is if you think about the experience that most people have today with booking travel, it's about hunting, right? They're doing a bunch of Google searches. They're looking up hotel websites and airline websites and searching for different ticket prices. Maybe they're going to some sort of other aggregator on the internet, but they're still assembling that whole package by themselves and they don't really get much information outside of what's actually being offered. They don't see the connections between things. 

  (08:27): 

  But what if you were able to say, "Hey, this is what I like. This is the experience that I want." You might be going on a business trip, but maybe on that business trip, "Hey, the Red Sox are in town. Can I catch a game?" That sort of thing. Or maybe you're going on a personal trip and it's like, "Well, I don't know exactly where I want to go, but I think I want to go somewhere in Italy and I want to have these experiences. I want to go to a cooking class. I want to see some Greek ruins," whatever. And you can basically start having a conversation with the AI platform and then it can then hone in on exactly the right itinerary for you with all the right components. 

  (09:01): 

  Then the other cool thing now is with the agentic solution, with MCP and the integration we can have with a system like Sabre, you can actually not only get the itinerary, but you can actually book that itinerary, right? You can actually buy that itinerary, and more than that, it can actually follow you through the journey. So you can still be interacting with your AI during the journey and say, "Hey, I got to this hotel. I've stayed here one night. I don't really like it. Can you give me an alternative in the same town that's better for these reasons?" And it can give you other options and so on and so forth, right? So this is the way that now you're starting to be interacting with that AI and it's actually influencing your purchasing decisions and your choices based on all the data that it can give you. 

  Melissa Harrison (09:47): 

  I definitely know that for myself and for others, we've been using different types of chat to plan vacations to start to get an idea of when you're going somewhere, what could we look at? But I think we're just scratching the surface. 

  Joe DiFonzo (10:01): 

  That's correct. The way this is going to go, now imagine the situation where these generative AI tools can be producing images and videos and things like that. It can almost prepare a whole sketch of your entire trip for you. It could show you the places that you're going to go, not just talk about them. It could make recommendations, and this is the interesting part, that you might not come across in normal web searches because, again, it's connecting things, it's making inferences that you might not make or you and travel agent might not make that are still interesting to you and gives you choices. And you could keep interacting with it again and again and say, "Well, I don't really like this, but I like that." 

  (10:41): 

  What's interesting about this is people interact with the AI in much different ways than they would interact with a person. If they're talking to a human, they may get steered by that. They may be shy about asking too many questions. They may be like- 

  Melissa Harrison (10:54): 

  Or saying, "No, I don't actually like that"? 

  Joe DiFonzo (10:56): 

  Exactly, exactly. But they're not that way with the automation. So with the AI, they might get more to the exact right trip that they want to take. 

  Melissa Harrison (11:05): 

  We've focused a lot on the consumer side of this. How are brands responding to this change? Because it means that brands on the other side are also going to have to be prepared for how this is going to work. 

  Joe DiFonzo (11:14): 

  Oh, absolutely. And this is really critical because think about it. What are the purposes of those brands, especially travel brands? Is to serve those consumers and give them what they want. They want happy travelers. They want happy customers. So by getting a customer closer to the exact right trip, the exact right itinerary, that's going to help get them there. So we're working with several companies now in the travel space, specifically some airlines that are looking directly at this problem and we're helping them build these experiences with basically AIs that they're essentially rebranding to represent their airline. For example, this could also happen with a travel agency or with a hotel that's then available to customers to use directly. And it's really cool because one of the nice things about these experiences is that they can be tuned to your specific brand. They can make it sound like if you're an Australian airline, they could give that flavor to it versus an American one. 

  (12:11): 

  By the way, the one other cool feature is they can naturally converse in many languages at once. So if you have people that are coming, let's say from Japan or Korea or China or wherever, it's not just about English or whatever the local language is that's there and all of this happens pretty much automatically. So it's a really cool experience, but it definitely then, again, it increases that customer trust and increases the customer focus. It's essentially something that gets you to a more satisfied customer at the end of the day. 

  Melissa Harrison (12:43): 

  Well, and it allows brands to hyper personalize, which is what consumers are really demanding now. 

  Joe DiFonzo (12:49): 

  Absolutely. So now think about it. You think about your frequent flyer program or your affiliate program or your- 

  Melissa Harrison (12:55): 

  Yeah, your loyalty... Right, right 

  Joe DiFonzo (12:56): 

  Loyalty program at a hotel. Exactly. And they can actually use that data constructively, right? It's not just about sending blind offers, but maybe now attaching this to your experience to say, "Oh, hey, and by the way, you could stay at Hotel X or Hotel Y, but you have a loyalty program at Hotel Y. And if you stay at this particular property, you can get two nights for free." So all of these things now can get to a much more connected experience. And by the way, do it all in one shop. This is really key because if you ever try to book a complex trip with multiple flights and multiple hotels in multiple cities, it can take hours, it could take days to get it all right. And you're like, "Oh, I got the perfect hotel for these days and oh, shoot, I can't get the flight I need for that time or I got a flight, but the hotel I want to stay in." Now you can actually put the whole package together at once and book it. 

  Melissa Harrison (13:48): 

  I love it. Okay. I want to nerd out with you just a little bit because you recently launched the first comprehensive agentic APIs, which you're calling the model context protocol, which is being described as a universal translator. So for everyone listening, can we unpack that? What does it really mean and why is this such a huge breakthrough? 

  Joe DiFonzo (14:11): 

  Yeah, this gets a little bit wonky. 

  Melissa Harrison (14:13): 

  I love it. We're here for it. 

  Joe DiFonzo (14:15): 

  Okay. What's interesting is a system like Sabre has had these APIs for a long time and we've evolved them over time. There've been different kinds of API standards, the most recent of which have been web service APIs, where these things are basically just open to the web, but you really have to be a computer programmer to know how to use them and take advantage of them. So what's interesting about what's going on with AI now is there's a way that you can glue together that AI large language model or large reasoning model with those wonky technical backend functions like a system like Sabre provides by using this technology called model context protocol. And what that is basically a way to wire those technical APIs that you have on the backend to the AI on the front end so that basically the AI can speak to your adapter in regular language and then the adapter can turn that into the specific technical requests that have to happen on the backend. 

  (15:21): 

  And so with a system like Sabre, because we can do the shopping, the booking and all the support of the travel, we can make all of those APIs available through MCP and thus make them all available to the front end AI, but you as a user don't have to see any of that wonky stuff anymore. And even if you will, the AI adapter, if this company uses this kind of front end AI and this company uses a different front end AI, they can both use the same MCP adapter to talk to Sabre and they'll get the same results from given requests. 

  Melissa Harrison (15:57): 

  That's incredible. Okay. So how do your components like Sabre IQ, Travel Data Cloud, and then the partnerships with Google's Vertex AI, and Gemini all come together to power this next generation of travel retailing? 

  Joe DiFonzo (16:12): 

  Yeah, this is where the story gets interesting because if you build a fusion of technology, there's a lot of dependency on various components in this story, as we've talked about. We've got the basic technology of the large language model, but the services on the backend... And by the way, the other AI that we've put into them, that we talked originally about that statistical AI, we have that built into a lot of our services on the backend to give you better answers than it would normally give you from just a regular database lookup, for example. 

  (16:45): 

  So the fact that we have been doing this business for so long and we have all of these customers around the world, airlines, hotels, travel agencies, car rentals, et cetera, we have all of this information stored, of course, in a very anonymized way, in a very safe way, but gives us all this background on preferential information. And we can use that, right? This is the Travel Data Cloud you talked about. We basically keep that in a large, what's called a big query database from Google. And we use that data to drive both our statistical AI and it's also used by that generative AI, those NCP adapters when they need to get the right answer for customers. 

  (17:27): 

  So basically we can take all of this work that we've done to move our systems into Google Cloud, to move our data into Google's BigQuery, and now to use Google's Gemini product, which is the large reasoning model that's behind it, but also what's the Vertex AI platform, which Gemini is kind of a part of, but also that old statistical AI function is still in there. All of these technologies come together into this one great solution that allows us to give those customers of ours exactly what they've always wanted, which is a uniform, simple interface to Sabre that works with almost any human language out there that's very easy to integrate and that gives the customers what they want. 

  Melissa Harrison (18:07): 

  It's incredible to think about how integrated everything is on the backend and just how much data it takes, scale to make everything so simple for the user. But it seems like that scale data richness... I mean, in the intro, I mentioned 50 petabytes. I had to ask a colleague how big a petabyte was. That was a new term for me. These seem like this is your superpower though at Sabre. How do these assets really position you as the first mover in AI-driven travel technology? 

  Joe DiFonzo (18:42): 

  It gives us tremendous advantage because we can use this data in ways that it wouldn't have been practical to be used before. Many times we would use this data in very targeted applications. For example, how do we help an airline figure out how to better plan their routes for maximum efficiency and fuel and at the same time maximizing their revenue intake? So that's a very specific question that we look at slices of that data to get, but now we can actually use that data more in the aggregate, look at many different elements of it altogether. 

  (19:16): 

  Here's an interesting thing. When you think about the shopping function, people are looking for airlines or looking for flights or looking for hotel rooms. Sometimes you get more data out of what they're not selecting than what they're actually selecting. So we have that kind of information that lets us know, hey, and by the way, there are events that happen that drive travel and you want to know. And if you're going to New York City for a particular thing that happens that time of year, then that's a choice. If you're going to New York City for some other reason, but flights are really expensive that week because of the thing, but somebody's not going for that thing, you might tell them, "Hey, you might want to go a different week because the flights are going to be more expensive because this particular event is happening." Maybe the Super Bowl is happening with the Jets or something. 

  (20:01): 

  So that's the kind of thing that's interesting there. But you got to think about it that way. It's really about being able to use all this information and all of our services in the aggregate, but use them extremely effectively where essentially AI is guiding you right to the target you want to hit. 

  Melissa Harrison (20:17): 

  That's incredible. We do have some colleagues here who are Jets fans. So thank you for the shout out to possibly making it to the Super Bowl. 

  Joe DiFonzo (20:25): 

  I hope that they get there someday, but I don't know. 

  Melissa Harrison (20:27): 

  Someday. I know. Well, I'm a long-suffering Cleveland Browns fan, so we are in safe company. 

  (20:34): 

  We talked quite a bit about the consumer interface and we talked about rebooking disrupted flights, confirming hotel check-ins, even handling visa applications. So I'm going to just ask selfishly, which of these cases is the closest becoming reality and addressing these common pain points for travelers? 

  Joe DiFonzo (20:51): 

  Well, certainly selecting booking flights and hotels is getting extremely close right now. That's the one that's probably the most obvious, the most common experience people have, but the other ones are getting very close too. The thing that's interesting about this technology, if you've noticed, it's moving really fast. ChatGPT really just came out in a public way in very, very late 2022. So we're talking about three years later, and look how quickly it has moved. We're seeing major developments on the AI front almost every 90 to 180 days now. 

  (21:26): 

  The way this is going and the ease of rapid integration that we're seeing from these technologies is really driving progress at a rate that I haven't seen in a long time. I mean, I would probably be going back to when web technologies first came out in the late 1990s, right? Now all of a sudden, everybody had to have a website. Now, everybody's got to have some sort of AI presence with your technology provider like Sabre, you've got to get on there right away. But if you're providing any other kind of business, people are going to use the AI like they use the web. And instead of having to define these fairly static or slightly dynamic webpages that you have to trundle through to get to your answer, now you can just naturally interact. You could speak, you can type it, you could speak to it could speak back to you and it can dynamically and much more rapidly get you to your answer than you could do through any normal website. I like to say AI will become the new UI. 

  Melissa Harrison (22:27): 

  I love that. You should probably trademark that now. I mean, Sabre really is on the forefront of all of this. What are the early insights or feedback that you've gathered thus far from the testing and the programs that you're working on with all these new AI capabilities? 

  Joe DiFonzo (22:43): 

  Well, the first thing I'll say is that everybody's really excited about it because I think it's been a long time since our business people and our customers' business people have seen such promise in a technology. It really is something that we think will change the landscape of the travel industry. So that's one thing. And another thing is I really believe, and our customers believe, that this is now the vehicle that they can use to satisfy their customers more than they ever have before. There's a lot of frustration in travel, and certainly the whole experience of booking and servicing travel is difficult, it's challenging. And there are lots of things that can happen. Externalities, right? Flights can get canceled for weather or something like that, for example. So it can be very difficult to deal with these challenges. And now this is a technology that can help the traveler deal with those challenges very effectively and efficiently. 

  (23:45): 

  And by the way, in a lot of cases, we believe it'll deal with them even automatically because nothing's stopping, for example, that AI from reaching out to you directly, even if you're not asking it a question, but it's in the background working for you the whole time you're on the trip saying, "Hey, I've noticed your flight is delayed. I'm going to reach out to your hotel and tell them to expect a late arrival." 

  Melissa Harrison (24:06): 

  Wow. I'm ready for it. Bring it on. 

  Joe DiFonzo (24:10): 

  I think we all are. The people that are working on this are super excited about it too, because we all travel. I'm traveling now, for example. 

  Melissa Harrison (24:16): 

  Right, right. 

  Joe DiFonzo (24:16): 

  And so this is a thing that we are all just super excited about. We think it's going to be a game changer. 

  Melissa Harrison (24:23): 

  We would be remiss in this conversation around AI if we didn't talk about trust and transparency and privacy. As AI takes on more responsibility for managing travel data and decisions, how is Sabre approaching this and making sure that you're ensuring transparency and reliability and privacy? And then also, there are some challenges like latency or hallucinations. How is Sabre approaching all of this? 

  Joe DiFonzo (24:53): 

  Okay. There's a few different dimensions here. Let me start with the outer shell of it. Sabre has spent a lot of time and a lot of money on our security profile as a business. We realize that we are touching a lot of very sensitive information about people, about the traveling public, about our customers and their businesses. So we invest a lot. We invest tens of millions of dollars every year in our security program and making sure that that is just top flight, best in the industry, keeping that data safe, making sure that only the proper people or organizations have access to certain pieces of data. That's very important. That investment has to be at the core of everything that you're doing. 

  (25:39): 

  And then all of that now applies to everything that we're doing in the AI space. So we have very careful rules about how things can be accessed, for example, how those NCP adapters can access the data so that they don't accidentally allow information to leak that shouldn't be leaked out through those interfaces. And everybody really has to focus a lot on that. There's a lot of unique technological challenges around that that we face. 

  (26:05): 

  Then the other part of your question is interesting. It's more the inner part of it, which is we've all seen when the AI makes mistakes and it will actually assert things that we know are not true. Now, I will say that the technology has gotten much better over the last couple of years about that. The companies that are providing this tech are spending a ridiculous amounts of money trying to eliminate that problem, and we do things as well. So there's a lot of work that we do inside of our applications, inside of those adapters, to specifically tune the grammar that we use in those things, to tune all of those queries and prompts to make sure that we get as close to absolute accuracy as possible, and even to, if you will, loop around and recheck answers multiple times in cases to make sure that the data that's coming out of the system is absolutely accurate and not a hallucination. 

  (26:58): 

  And by the way, this is another good place where the agentic stuff comes in because a lot of those hallucinations that people have experienced come from the place where the AI is trying to generate an answer just from its past training. But now with the agentic interfaces, it can actually get live data to generate the answer. So there's no need for it to hallucinate to make something up to give a person the answer that they want. They can actually go get the actual data from the system that's really managing the transactional state of things, and that really cuts down on that issue a lot. 

  Melissa Harrison (27:31): 

  I know that Sabre has previously said that trust is the new currency in travel retail. What are the most important steps that you're taking to keep that trust front and center as AI becomes more conversational and autonomous? 

  Joe DiFonzo (27:43): 

  Sure. First of all, I'd say we have a lot of compliance discipline at Sabre and we share our compliance activity with our customer. Of course, by compliance, I mean, are we complying with all of the regulations and standards around how we manage data and keep it private, how we manage access to our systems and keep that appropriately controlled? So that's one thing. And we will interact with customers at that level. This is really important. Almost every one of our major customers has regular interactions with our chief information security officer, as well as our leading design specialists and architects to discuss these issues from time to time. Because by the way, every one of these issues that we face at Sabre, every one of our customers is facing as well in their own systems and interfaces, right? 

  (28:32): 

  What we're thinking about is we're trying to raise the entire industry up with us at this level. And that's a way, again, that we build that trust with our customers because we don't hide anything in terms of how we do the things we do, because that gives them a level of trust. If they understand how we're keeping things secure, if they understand how we're building our interfaces, if they understand how we're managing our identities, they don't need to be able to go into our systems, but they will be able to understand that we understand what we're doing and we're doing things the right way. 

  (29:06): 

  And by the way, we also take feedback if they request that we work harder in certain areas or improve certain things. We treat these things just like any other feature requests on our product. So the customers see that as well, that we're looking at that aspect of trust as essentially a core capability, not just something that's underneath there and hidden and don't worry about that, but basically something that's exposed and visible because it's important that everybody feel like they can trust these systems and that they can trust their business to them. 

  Melissa Harrison (29:37): 

  Well, I really appreciate your business perspective, which is this idea that a rising tide lifts all boats and that the work you're doing will bring the rest of the industry along, which leads us into a nice segue to talk about CES 2026. One of the things I love about CES is that you just never know what you're going to see. It really is a melting pot and it brings together innovators from across all different types of industries. So I'd like to get your perspective on how you see travel technology and Sabre's work in agentic AI fitting into that broader conversation around the future of connected and intelligent experiences. 

  Joe DiFonzo (30:10): 

  It's all about the consumer and how the consumer's being served. What's interesting to me about this related to travel and Sabre specifically is Sabre is not really a consumer brand. We're a B2B type company. But what I'm hoping is that what we will be pushing on now will help improve that consumer experience around travel and bring it to a place where it's never been before, bring it to a place where people are confident, comfortable in their travel experience, that they feel like they've got somebody there helping them when they're on that trip, helping them prepare for the trip, helping them manage the trip. It's something that will touch you everywhere. It will be on your tablet, on your laptop, on your watch, on your phone. Anywhere you are, that travel data can get to you and you can interact with AI to help you drive the experience in a good way and in appropriate ways. Eventually it will get to things like virtual reality and stuff like that as well. But all of these technologies basically can reach back in and get to any of that critical data that they need. 

  Melissa Harrison (31:20): 

  I want to pretend like it's the year 2030 where we're five years ahead from now. How do you see the traveler experience evolving as AI becomes more embedded? I mean, you just talked about, I think it's going to be at the point where it's going to be so embedded in our daily lives that we may not even notice it working in the background. 

  Joe DiFonzo (31:39): 

  To me, that's the key element of this, which is it's almost a set it and forget it solution. It's a level of comfort in the experience. When you're traveling and when you're planning travel, it's a thing that people don't do terribly often unless you're a real road warrior for business. It can be a disconcerting experience. It can be something that's complex, but you basically want to feel like you've got somebody in your corner helping you every step of the way. And the feeling that if something ever goes in a way that you're not expecting, you've got something that's helping you solve that, helping you close the gap, right? There's a lot of possibility there that I think we're just scratching the surface of right now. And I think over the next couple or three years, we're going to see a number of developments that get us much further down the road on this than we could even think about right now. 

  Melissa Harrison (32:36): 

  I always save the hardest question for last. I would love to know what you're most excited to see at CES 2026. 

  Joe DiFonzo (32:42): 

  Oh, wow. It's so hard. I'm such a tech geek. I love everything. I could look around that place forever. I'm always excited about the interactive technologies, the visual stuff. I still think we're having a hard time getting where we need to be in terms of the augmented reality devices. I think that's going to be a really big thing, which by the way, and I think that's going to be great for travel too, because again, guiding you around, helping you get to the right place, explaining things that you're seeing, all of those things, especially when you're in unfamiliar territory, but that's what I'm looking forward to most, what developments are coming out on the AR front. 

  Melissa Harrison (33:20): 

  That's great. Well, I can't wait to see you in Las Vegas. Joe, thanks so much for joining us today and for sharing your insights on the rapidly advancing technologies that are shaping our world. And thank you, our listeners, for tuning in. 

  (33:31): 

  That's our show for now, but there's always more tech to talk about. Be sure to follow, subscribe, comment, and whatever else you need to do to keep those algorithms happy. You can get even more CES and prepare for Vegas at ces.tech. That's C-E-S dot T-E-C-H. Our show is produced by Nicole Vidovich with help from Paige Morris and Doug Weinbaum, recorded by Andrew Linn and edited by Third Spoon. I'm Melissa Harrison on the special Inside Innovation episode of CES Tech Talk.