Heads In Beds Show - What is the "right" AI travel planning interface?

Episode Date: July 8, 2026

In this episode, Conrad and Paul discuss Airbnb CEO Brian Chesky's bold perspective on AI, why traditional chatbots fall short, why Airbnb isn't rushing into AI, and what the future of travel planning might actually look like.Enjoy!⭐️ Links & Show NotesPaul Manzey Conrad O'ConnellConrad's Book: Mastering Vacation Rental MarketingConrad's Course: Mastering Vacation Rental Marketing 101🔗 Connect With BuildUp BookingsWebsiteBook A Call With Us🚀 About BuildUp BookingsBuildUp Bookings is a team of creative, problem solvers made to drive you more traffic, direct bookings and results for your accommodations brand. Reach out to us for help on search, social and email marketing for your vacation rental brand.

Welcome to the Heads and Bed Show where we teach you how to get more properties, earn more revenue per property, and increase your occupancy. I'm your co-host, Conrad. And I'm your co-host, Paul. All right, Paul, we're here, we're live. What is happening? What's going on? Our sports worlds have been respectively rocked. I think most people this time of year, maybe talk about World Cup. It's midsummer, which has been a compelling. Don't get me wrong. Team USA, that was a fun match to walk last night. I don't know if you caught any of that. but that was good. So for those of you who are listening, we are recording this the day after USA

advanced from the round of 32 against Bosnia-Herzicabina. We will be playing without star forward. That's a whole other issue as soccer fan. I said they might appeal it. I read an article about that this morning. I had a discussion with my wife about that before last night, like immediately following the match. So I don't know.

There's some home cook in there that I don't think I'd want. It's a bad call, but I don't think I want the home cooking either. I think I want to earn this one the right way, but we'll see. Yeah, but in basketball world, how are you feeling? Kind of how is like in your world right now? It's really interesting. Typically, when I think of podcasting with a man named Paul,

I'm pretty interested and excited to do it because it's something that I had to get to do. Now I have to watch podcast or Paul George play on the Celtics this year, and I'm not enjoying it at all. I don't think it's going to be a good experience. Yeah, I don't know. There's nothing good to say about it. They sold a vacation. rental. They should have gone for $1,000 a night for $30 for the night. And I'm not thrilled

about it. And revenue management was not good on that deal. Not bad revenue management on that deal from a man who has historically shown really good revenue management skills. So that's the disappointing part. But you know what it is? It's life. Things don't always go your way. Things change. Things shift under your sands. It's interesting ball. Travel AI planning. That's a thing now. And we're seeing that pop up more and more. So you snagged this idea out of the internet verse here by highlighting or reviewing a little bit of a new startup run by Francis Davidson, the former CEO of Sonder, which went out of business and burned billions of dollars of capital. But he's launching a new company called Odessia. I think that's how

you say it, right? So if you go to Odessia.com, you can see this AI travel concierge platform, which by the way, you and I both signed up for it and we used it to not actually, I didn't plan a trip in yet at least, but it is well done. I like the interface. I like the way you go through things. It is a fresh take on things a little bit in my opinion. But it was fascinating. this was what you uncovered. This Twitter user, X user, I should say, excuse me, by the name of Adam Weiss, tagged him and said, hey, this felt very good, just tried O'Desia. My first thought was this experience felt like the new wave design UX and trial.

He tags in Brian Chesky, CEO of Airbnb, and says, I think you should look into this. Great job, Odessia team. I always said it would be a mixture of chatbot visuals illustrative, and the only thing he hasn't seen yet as voice, which I didn't see a voice component to Odessia either on my side of things. Brian responds, Francis is the right entrepreneur. And I respect him giving this a shot, but in my opinion of chat interface is not the solution for travel. Interesting. So Brian's kind of saying Brian, who runs a platform that doesn't

use the chat interface for booking, says this is not the right solution. Let's dive into it. Paul, what do you think? Is chess key right? Is it going to turn out that Francis is right? Where do you see this transitioning and going as we go along here? So it is. I think it's, it just, it's piqued my interest because when I did go into it, admittedly, I think I've been pretty anti-AI being the driver of these chat. This is just not where we are yet for travel planning. This is the first application I've seen that makes me think, ooh, there's something, there's something here. There's a there. Now, again, how people travel what their desires are, what their needs are. I think that's going to make a difference with this. But as far as a text-based AI chat type of format goes,

this is pretty cool. This is something that. that I can see people I know using. Not me personally. Again, this is not that. I think I'm still more on the traditional side of things. But when we take a look at, I think the part I called out right away was anytime you're bringing Brian Cheskin in the conversation, you're talking about a person who has access to all of the data in the world.

So I'm going to say that his understanding and knowledge of the travel. experience and user experience. He's tiny to modify. I get that. That's running a business. But ultimately, if he doesn't think that the chatbot is the next wave right now, that's what the dad is telling him. I don't think Chesky's ever been great at being cagey with some of this stuff. Like when he knows stuff is better, when he knows has some insights behind the scenes, he's pretty good at being blunt about it, but not being the reckless. It's just if you're reading between the lines there, and maybe I'm maybe. I'm maybe. misreading and overreeding and doing everything like that.

That's what we like to do here. But I do think that right now that statement is very strong. And he said it multiple times this year specifically. I think as we get more towards AI, agentic experiences, stuff like that, we're always going to have these insights and we're going to have this understanding and these things. We're going to have to test and try and do all those things. But again, if anybody knows what the true trigger is to move on to that next level,

his chest. So I do think that this could be something folded in. I think as we went to some of the other articles that were surrounding this thread, it was. It was people weighing in of what does this mean, though? I don't think AI chat based out of the question down the road. But right now, I still think we're caught between AI-driven travel and traditional

search-driven channel and still travel agency travel and a lot of other things like that. And there's no clear direction where people are moving channel-wise. I think ultimately that marketplace does appear to be, that people are consistently choosing Airbnb. I don't know. There's a lot to impact. What are your thoughts? You're looking at all this stuff.

I pull apart a few layers here. One is I get the sense that Brian didn't actually use Odessia because to me it didn't feel like a chat interface. It felt like a very rich interface as I tested it. Yeah. So if we compare that, let's get very, let's get very in the weeds here. Because like you said, we have time to do that. When you go into chat, CBT and you say, I'm planning a trip to XYZ location or destination or whatever. The initial kind of versions of that were literally just text, like literally just text. And then you'd see a little link or citation. Hey, you can then click over and look at this hotel or this vacation rental brand or whatever the case may be. I think that's what Brian's getting at is that that is a very poor interface for planning travel. But the way that Odessia worked for my experience or just testing it for the first time was that it was a little bit more rich. There was like, as it recommended. let's say hotels even, for example, it pulled in a little preview, it pulled in a little title, it pulled in a short description, maybe the brand of the hotel is present. So to me, that feels, again, a more natural planning interface when there's like multiple steps in the journey.

And again, part of me wonders if I zoom out for a little bit is Odessia just essentially putting a fancy new UI layer on which has been around for multiple decades and they have their interface, but maybe it's just they've been doing things the same way for a long time. You come in and it gets a reimagining it with some additional AI layers inside of it. maybe it's another way to go about it. I would challenge Brian on he should actually use some of these tools and not just say, yes, version one of chat ChbT was a poor interface for booking travel. We can all agree with that. He didn't interview where he talked about it and he said the reason that they had a chat dbt app and I posted this clip on my LinkedIn before too. So this was a while ago. This wasn't a more recent comment. But he said basically to the effect of they don't have a good interface for me putting my data into this where I can actually see like a rich card. And when I think of a rich card in Chachabit or Claude, which is like my daily driver, L.m, I've seen some that to have this where you go and request it and it builds like a pretty looking dashboard or a graph or something in your chat response. So imagine a scenario where you do a request to Airbnb and it's saying, I'm picking these four properties for you and here's a picture, here's a little over you, this and that. But what I'm

noticing with all these chat interfaces and Odessia works the same way right now is a lot of time, they're just kicking you to the site to book externally anyways and a lot of different options. So not everything. To be clear, some stuff on Odessi, it looks like you can book on that platform in this smart AI interface and appears to be mostly the hotel inventory. So that that's fine. But yeah, if it does actually quote unquote recommend an Airbnb, it looks. like it does pull in the thumbnail, but you click and you book on this other interface. So if Brian's correct, I guess the question would be, okay, but then he's not a competitor to Odessia because Odessia is trying to do this end-to-end vision. That's their vision that you,

I'm assuming, would book air travel. Maybe you'd book even like a taxi. You would book. They should quote, quote, short-term rental or hotel and Airbnb through that interface. And it's just adding it to like this agenda on the right side as you go along, at least on desktop. So it's like you've got your flight booked. You've got your lodging booked. You've got your taxi or your like your transfer book, et cetera, et cetera. And then everything is just this very smooth experience, even if it was pulling from many different vendors, it would then all tied together to a trip, which is inclusive of all these different things. So I think that's interesting. And I think that maybe Ryan needs to be careful not to be too close-minded to that fact that interface may be

superior. I guess the question remains, is it going to be a layer on top of existing stuff, though, because that's what Odessa is. Or is, quote, unquote, Airbnb figures this out, where does Airbnb start and then they're sort of external vendors or whatever you want to call them hosts or providers begin? Because I don't think Airbnb getting in, like, the flight booking game is as profitable as I think it's going to be. And then it clutters up the interface. Like, why am I going to Airbnb? Am I booking a flight? Am I booking a short-term rental? Am I, now they've added silly things lately, right? Experiences. No really super traction there. They've added all these things, but it's, I just don't know. Maybe it sounds good on the surface, but are they actually able to execute all those things

well versus something like in Odessia? Maybe that's the new Expedia where it's, hey, let Airbnb be best in class of what they're doing. Let all these providers be best class of what they're doing. Let me just wrap it together in a really pretty interface and that sort of thing. So that's my thoughts on. I keep seeing, on the Twitter note, or the X note, I keep seeing all these, the influencer and everything like that saying, oh, this is what's going to end Google Flight. This is what's going to end Google Hotels. This is what's going to end this. The prompt is this, that. The problem with that. Wow, I'm going to give you these 20 prompts. They're going to get you the lowest price package ever. Okay. When you're doing something like that, really doing budget travel, my mom was a coupon. cutter on top like a super coupon cutter and was always the deal finding. So this is something that not everybody wants to travel one way. It was, I'm going to tell you, there are some times we're doing travel. It's just like, okay, this might not be what I signed up for necessary is what would do. So I do. I think that's, that is the thing that it's very difficult to

build into, unless you have that personalized customized, you're building in that personal. intelligence because I think everybody's travel motivations are different. And budget plays a very distinct value for something. It is. It's a very distinct important part, the facet of the trip planning process for some people. It's a nothing burger for other people. And trying to build that into any algorithm, any smart system, I think it's very difficult because, yeah, that's probably a genuine motivating factor for 99% of people, of some degree. But to say, to give a prompt of I want the lowest possible cost for this, for that, you are going to be flying a discount airline, perhaps going to a discount hotel.

I would assume that someone who took that trip, who was anticipating a, let's say, five-star experience, four-stars, is not going to have the experience. that they want. And is that anything to do with the prompt, the travel planner? No, not necessarily. I think that's what makes all of this so difficult. I go back to accountability a lot with AI. And I think this is one of those areas specifically, is that you may be able to get someone to plan one trip like that. The likelihood of it happening twice is this. I find I think you're going to struggle to have a retention of that traveler long term. So I do anytime we get in this like layers of this, that's always something I struggle with. Those target personas, how do you identify the generalized

target persona for X, Y, and Z? We know in the case rents of space, it's top. You build in hotels, build in the flights, the experiences, anything like that. I do. I think that when Chesky talks, he knows Agenic is going to be part of the future. It's probably why he's the future. shame by Singapore keeps hitting all of our sites another time we'll continue to see but that is I think that like I think numbers saying like nearly in one of the articles we pulled up nearly 30% of travelers now use AI for full trip planning not just research okay full trip planning we're still missing that one keyword and I think that ultimately here trip planning is great but where is the rubber going to meet the road?

Yeah, Odessia is changing people to that third party in there. It's a vendor partnership and it's an aggregator. It's an OTA. What is it really? I don't know. It's a trip planner. So that's one of these labels of what is it going to look like? Where is this all going to happen?

I still, I think that, again, that's where I'm going to keep on leaning into Vacheski's insights and trying to figure out. When are they building this into it? Because they've got to have. have per market with all the bookings that they take. They've got to have kind of the target persona for Myrtle Beach, for MSP St. Paul, for New York, every major market around the country. So how are, when is that, when they start to move the needle on making it more seamless of going from plan all the way through the booking, that's when I think of it is. We're 75% of the way there. We're not that 100% and until

there is that accountability, there is that that peace of mind that trust that Airbnb has gained with siding with travelers for as long as they have, they're going to have to do that again on the AI side of things. They're going to have to reestablish that trust

that I haven't done this before. I haven't booked to be at AI before. What does that look like? What does that feel like? There's going to be issues. There's going to be problems. They're going to have complaints and they'll probably shell out some money and make everybody happy and move on. So I think that's the part of this that's that I continue to struggle with just as we evaluate how we reach travelers right now. It's how are

they using these things. I see some locations where assistant AI traffic or AI assistant traffic is going way up. Some places it's absolutely stagnating. Content strategy is very similar. Content addition is very similar. So I don't know. There's I think there's so much more than just thinking about it from just that chatbot experience or whatever that looks like. It's how are people taking this journey of the new? We've seen the morphing of the trip planning experience. How does that really fit in the AI landscape here? I don't know.

Just for fun, I went into it and I asked it really quickly just to do a staycation, plan a staycation for me. So it removed all the flights and all that kind of stuff. And I picked Myrtle Beach. Give me a recommendation for a property that's currently rated four point 2-9 on Airbnb across 21 reviews. So that was the top recommendation from Odessia for, from an AI interface. And by the way, I clicked over to Airbnb because you can't book on the Odessia platform. So is that the best possible option for me? It's a two-bedroom for those

listening in home I have four children. It didn't ask me that either. Have fun. Have fun. Yeah, I'm good. Thanks. But yeah, I think that to go back to what you're talking about there, in theory, I feel like the version of this AI trip planning that works well, both within Airbnb and potentially within other interfaces like Odessio, he's one of the one of the one. that actually knows my preferences. It knows what I like. And it recommends based on those preferences. And maybe even it tries to send me up or down. So maybe it gives me like, let's say, imagine, imagining a search where I put in dates. It knows a little bit about what I'm looking to do. Who are you going with? That kind of stuff. And it's a little bit smarter about that.

So I type of thing. You click search and then it comes back. And maybe, let's say, gives you six recommendations that it thinks are pretty spot on to what you're looking for. And it gives you three that are niceer. Maybe. Hey, if you want to spend a little bit more, here's three that you could go upgrade a little bit. Treat yourself. That sort of thing. And it gives me three that are a little bit more better value. Hey, if you want to save a few bucks, go over here. You can use that money on an extra dinner with your family or something like that. So maybe that's a version of it where the right 10 listings for me are not the right 10 listings for you, even if we did enter the same criteria as far as dates, as far as number of guests coming,

that sort of thing. We may have different preferences. And like you said, people are going to all have different preferences all up and down the chain of value in the chain of cost even for the same dates, right? They may be coming to that destination for a lot of different reasons. And to be fair, to these chatbot platforms like an Odessia, how to know that, unless during onboarding and asking all that information and do all my preferences. In theory, something like Airbnb, be I would believe would be able to figure that out better because if you've had 10 trips in your Airbnb history and you've left a five-star review on these seven homes, like, okay, they like these seven homes. They left a five-star review here. They left a two-star over here, three-star over here.

What was the problem on those? Was it home-related? Was it location-related? Was it more property manager communication-related? In which case, I'm not going to dig the home for that if you're Airbnb. So it's like that's a very interesting challenge versus just almost like the this old-school way of thinking about things. No, I'm just going to put properties in a list. You put it to date some show you properties in a list mode. Like, what a crazy. And of course, there's this algorithm behind it, but it's the same concept, right? Like how those things work.

I do think that part of the, and that's where you're in your head, it's bottom of the head there, that part of what makes AI more effective is like it being coached along the way. And I think that this process of travel planning has not been like coaching. Why is this one wrong for you? It's just you browse your image. Go on the listing page. Go on browse the images. Go look here.

Okay, go on to the next one. That's been. But we're not. We're never telling the website, hey, why don't you like this? We have never put up an accident on that listing page. It says maybe they get some things every one spot. But to understand, why didn't this fit your match? Why didn't this match?

It'd be an interesting little thought experiment to try that on a larger site once every, if they've sat on a page for two minutes. Hey, let's pop up. Why didn't you? Why aren't you? Why not? Now, I might get a little intrusive. You might get a few complaints and ultimately it might not be worth the experiment.

But we don't know. Like we can, I look at clarity sessions way longer. And I should try to get those insights, but I don't know what's in someone's mind. When I look at they're hovering over the description and then they're going over to the calendar and then they're going back to the images and they're trying to match. I don't know what they're trying to match up. I can try to make those educated guesses. But I think that's what it is that.

if that's where I do think that Odessia with time, this is iteration one, there's some, there is some power there because we are constantly kind of educating the machine. And I think from that standpoint, if you did get that first trip nailed down, you could have future. That could be a flywheel of trip. For those, for people who are a multi-trip, a year type of group or younger generations, they want that experience and they are looking to take more trips. for you. So how do you get in front of those people? That might be the way that that might be the way

and that may be the way they're looking to plan their getaway. So I do think that's the one place where AI can provide a better experience than anything we can do right now. Yeah. There was one more concept that Francis had published on their website and on Twitter, excuse me, and I'm only really sure how this works in the website, but it was this idea of finding a new location based on your preferences. And he's about this concept of maybe overrated versus underrated or maybe undiscovered. And people always say that. I'm looking for the hidden gem. I want the place that people haven't discovered yet, whatever, which does make me laugh a little bit too always, because it's, I said this before. Sometimes the touristy things are touristy for a reason. They're really

good. You know what I mean? Could you really truly imagine going to Rome and not spending a day at the Trevi fountain or looking at the Trevi fountain? That is a very touristy thing to do in Rome. My wife and I did that. And it's like, how could you go to Rome and not do that? Now, are there other things you want to mix in your trip that are not super touristy or crowded or whatever the case may be? for sure. But again, a lot of things are busy for a reason. TPD on like, how would AI determine that, right? In theory, it's, oh, less people are going over here. Is that a good sign or a bad sign? If you're trying to find this hidden gems, in theory, that's a good sign. But in practice, that may not be a great sign. Maybe in the short term rental context, to flip it back over our way,

it'd be a property manager that's not getting a lot of bookings, but is getting good reviews in the bookings that he or she is generating. So it's, hey, this might be a hidden gem property because it's getting great reviews. It's just not had a lot of activity. It's not a lot of sale. And I've encountered over the years, many clients that I feel have been in that bucket. I'm like, it seems like you're doing everything right on paper. You're just not getting people to convert very well, but there's no very raw and obvious reason of why that might be the case. So that is a challenge that I think in theory, maybe AI could be more fair. I'm putting these in the air quotes, more fair to properties that need a little bit of boost.

It's okay, this one's got a lot of bookings. Can I give it the other way? Or is it's going to be like what we've been dealing with now for a decade, which is very algorithm driven travel. This property keeps getting booked. I'm going to keep pushing it, keep showing at the top of the search results. And it's like the winners win a ton. And then the losers, that's why you end up with all these zombie listings, a client of mine, of a zombie listing, which I find to be interesting. So yeah, very much an interesting kind of

problem. I'm going to read through some of the other things that Brian's talked about before and just kind of talk about them a little bit. So he's actually criticized Expedionbooking.com and saying, this is a quote, rushing to bolt chatbots into their existing apps without rethinking the core user experience. So his belief is that because Airbnb has all this data about your stays, about your identity or location, all these properties, the reviews, et cetera, that just putting in chatbot so that you can just summarize some of that information is not really a great way to about it. I'll be honest, when I go on the Expedia site, I don't really see a very AI-driven process, to be honest with you, in terms of booking, nor do I see that on booking.com

either. I do see a little bit of work around that with respect to the reviews and things, some of those types of things. But if I go on Expedia today and I flip this thing in Archive. org from how it was a little bit, let's say three years ago pre-AI, I feel like it's the same thing. I'm getting a where to go. I'm getting a date picker. And then I'm getting a traveler, a quantity picker. And then I do get the option to out-of-flight, a book card, that sort of thing. As I go to the Verba homepage, I pretty much see the same interface that we've seen. now for quite some time, some recommended listings, and then basically where you want to go, dates and the number of guests coming in the door there. So despite all the chat and all the

chatter, right, it's like I'm kind of seeing the same thing I've seen for quite some time now. And presumably they may be testing other things and trying to figure out, hey, what do I need someone to do in the app versus the website? Again, the goal is to get someone to book. But we want to believe that AI is going to redo everything. And then you go look at all the companies talking about spending a billion dollars on AI. And we can end up with the same interface today that we've had for a long time. I'll give one more funny anecdote and then we can change the topic a little bit, but it's all these X threads you read, Twitter sites read about how AI is 10xing the amount of code that someone's producing or 10xing the amount of tickets they're able to process or whatever.

And then you go, oh, where's 10x the profit? Where's 10x the revenue? So if all these things are so great, then surely you'd be able to point to these huge increase in revenue profit. And I laugh because I'm pretty pro AI, but I'm like, yeah, that's fair. This stuff is making us more efficient. Is it making us 10x more efficient? TBD, I think it's certain things, certain tasks, certain little jobs that you're doing,

absolutely. It can make you 10x or 100x more efficient. But does that lead to this mass, massive increase in performance or revenue or whatever the case may be? I feel like the answer is no. The interesting thing for someone like at booking.com, a verbo at Airbnb, et cetera, we'd figure out if we make this new AI travel planning interface and it works really well, does it meaningfully increase our conversion rate? You know what I mean? That's the question.

Because if it doesn't, then like, why are we even stressing about this? If the quote of both the old way still works and we'll work for a long time, then maybe we're just chasing our tail a little bit, just wanting to say that we have AI versus the actual practical use cases of AI. That's a couple months on that. I look at the people who on their websites have implemented some type of AI chat and doing things like that. When I take a look at the clarity heat mapping of how frequently those areas are getting hit, hardly at all. And when it is, I've done a little bit of unmasking of some of those chat conversations just to understand what's happening there is variable, like low usage of click the bot, click the prompt that's already in there. maybe look for like discount cheap very rudimentary chats that are taking place so I don't like

I don't think I agree with Chesky saying just bolting on a chat bot doesn't make it better like that's I think that there's an assumption that there's going to be an improved experience but that's not usually the case like pro act a proactively better experience with AI that's just AI is not that kind of proactive thing but going back to what you'd said a little bit as well, if we could make it so that AI wasn't a gamified way to be found. Like, I think that that's, if it like, it is, it's like you can gamify SEO. You can gamify AEO, GEO, you can do this. Like you, there are ways to beat the system.

I don't want that to be the case. I do because that's kind of how we do our business. But at the same time, I would prefer that the top properties get bought. That's really like it really bothers me when I work with someone who has amazing homes and they don't get booked and the reason they don't get booked is a silly reason like Orde accuracy and stuff like that. And I don't think that would be the case because there will always be some type of algorithmic whatever schematic behind the scenes that gives you an insight into how this was chosen.

And even you ask Gemini and chat GPT and Claude, they'll give you a little peek behind the scenes. not, I'm sure if we asked, we both ask the same question about that. We both get different answers. If we asked it 10 different times, we get seven different answers. So I think that's, that is the one thing about AI right now. That is scary is that there is no consistency, is that you could prime. And I haven't done that with Odessi yet, but I'm interested to, if you tried planning the same trip over and over again, what would you get results? Because nobody's going to do that.

But what does that look like? What does the AI? how does the AI respond to something like that? Because, yeah, there's no predictability to AI right now. We can't, we don't know. We can't scale it. Like I said, some sites getting hundreds of sites of site visits a week through AI. Some getting dozens.

It's all the same behind the scenes. But for whatever reason, there's, and it is. I don't know. This is where the conversation starts to get unnerving for me because, and what's going to be, it will be a conversation down the road for us is, in what order are you putting AI into your business? And I think, again, we've talked about it. I think a lot of vacation rental companies are upside down on how they're implementing it

and where they're putting it into the process. We're a little upstream, which should be a little more downstream or vice versa in some cases. So, yeah, we'll leave that for our next upcoming episode here, but what are your thoughts on all that there? maybe it goes back to what we started with a little bit, right, to come to a bit of a last chapter here, right, which is figuring out the right interface sounds easy on the surface and seems to be hard in reality. Does it actually convert better? Is there actually a meaningful improvement there? And are we actually showing what is the objective, right? What is the outcome? Is it that we're,

are we going to send 100 people through this new chatbot interface? Number one, see how they convert, but also too in the back end check to see how they like their stay. Did they like their stay booking better through an AI assistant? Because they were able to explain or give a little bit of context of it. And this is where voice is interesting. And that was one thing that was actually mentioned in that regional tweet that Chesky replied to. He said, hey, I still haven't seen anything in voice yet. Because imagine for a second, what might we have liked about the travel planning interface from a travel agent? Let's go rewind in history, 30 years or something like that. We would have liked the fact that, number one, that person would have handled up a tedium

of entering their details and things like that for you. That's nice. But we would have liked the fact if that travel agent was actually as knowledgeable as they claimed, or let's say that group was as knowledgeable as they claimed they were, that they actually sat and listened to my preferences for five minutes before they booked the vacation. Like we've all seen these older video, film TVs. I'm sure we have some older listeners as well who might have used a quote-unquote traditional travel agent. And see, it's an online travel agent, right? But that's one thing that that interface to me would be fascinating, which is essentially I yap to a AI for five minutes about what I'm hoping to do. And it goes, okay, I'm listening. I've took in all your information here,

which five minutes of talking is going to be way more information than two seconds of clicking dates and clicking five guests and clicking book now. There's very little information being pass there. If I yapped to an AI for three or four minutes about, I want to take a trip by family, I have four kids, we want to go to the beach, you want to do this, we want to do this. And it starts to look around and go, okay, I see some versions that make sense here. Because of all this rich context you provided me, I'm actually doing a whole travel itinerary for you based on information. That may be a winning interface much more so than just tapping around the screen a few times and getting something like an Odessia or something like an Airbnb into the

interface there. And I'm sure those people are thinking about that and considering that as an alternative way to enter that information in. But my goodness, you're going to get so much more information from me by just explaining for a few minutes what it is. Or I'm very bullish on this idea, too, by the way, even for like our own business, the idea of an AI interviewing you. So you have this list of information that you want from somebody. And if an AI could interview you in the same way that I do an interview show on the other podcast that I do, you can get so much information on someone in an interview if you ask the right questions. So instead of making me tap, tap, click certain buttons, this and that, tell me more about what you're looking for. And then maybe we can build a very

rich profile about a person and about their preferences from a few minutes of conversation and then recommend to them truly a property that makes sense for them based on their criteria. And it may mean that you and I go through the same thing. And we both went to the same destination, but we end up a totally different places. And we're both pretty happy with our outcomes. So I think that's like another interesting wrinkle in this whole thing, which is that everything Jesse said might be true. And then it's, yeah, maybe if they're all wrong.

And the interface of traveler or the interface of AI in the future is more like the the film her or something like that where you've got like a little thing in your ear and it's sending information to your display that you're looking at or something. And you're looking at that and going, okay, I like that one. Show me three more. And then it boom, it just shows you three more right away. that may actually end up being the interface that might end up working. I see some validity behind that.

I think that her example, I think FaceTiming with an avatar, sometime in the future is not going to be like that. I think that's the way. I don't think we can, I don't think we'll just build in audio. It will be like a video aspect. I think about like FaceTiming with an avatar travel agent or your online travel agent on the road and having that travel planning process,

I don't think that's not a question at all. And that is just connecting the dots there and making people. making it feel like it's human. I think there is that nature that people will need that. But hey, if I can look in the eye and feel like you're traveling, you're doing all the research that you had done behind a desk 30 years ago, I know, I guess. We'll see. Yeah, we'll see. It's very interesting. I'll do this, man. I'll pop the link in the show notes to people so they can check out the tweet thread for themselves from both Francis, from this guy, Adam, and from Brian Jusky. They can dig around and do some research that they'd like there and see what's

going on there. Other than that, dear listener, you know, one interface that would be awesome for you to learn how to use. It takes you anywhere from 25 to 35 seconds. It would make our day. It's the podcast review interface. So go to your podcast app of choice iTunes, Spotify, we get the most downloads. You're going to click the button that says review, leave us to review. You're going to like five stars. You're going to write as a friendly message. It's going to say, we love Conrad. We love Paul. We love listening these episodes. It's going to make my day. It's going to make Paul's day. And it's going to help us out a lot so we can get more downloads. And we super appreciate that. Thank you for listening.

And we'll catch you on the next episode. Have an awesome mystery.