[
  {
    "start": 0.14,
    "end": 2.66,
    "text": "Hello friends, this is the AlphaList podcast."
  },
  {
    "start": 3.04,
    "end": 4.08,
    "text": "I am your host Tobi."
  },
  {
    "start": 4.4,
    "end": 11.5,
    "text": "The goal of the AlphaList podcast is to empower CTOs with the info and insight they need to make the best decisions for their company."
  },
  {
    "start": 11.84,
    "end": 17.76,
    "text": "We do this by hosting top thought leaders and picking their brains for insights into technical leadership and tech trends."
  },
  {
    "start": 18.16,
    "end": 24.06,
    "text": "If you believe in the power of accumulated knowledge to accelerate growth, make sure to subscribe to this podcast."
  },
  {
    "start": 24.4,
    "end": 32.22,
    "text": "Plus, if you're an experienced CTO, you will laugh at the discussion happening in our Slack space where over six hundred CTOs are sharing insights."
  },
  {
    "start": 32.4,
    "end": 33.6,
    "text": "or visit one of our events."
  },
  {
    "start": 33.88,
    "end": 36.2,
    "text": "Just go to alphalus.com to apply."
  },
  {
    "start": 51.46,
    "end": 53.3,
    "text": "Welcome to the Alphalus podcast."
  },
  {
    "start": 53.48,
    "end": 54.2,
    "text": "I'm your host Tobi."
  },
  {
    "start": 55.0,
    "end": 59.32,
    "text": "And to prevent questions, I'm in my new studio here, which is"
  },
  {
    "start": 59.94,
    "end": 60.68,
    "text": "better, not"
  },
  {
    "start": 60.78,
    "end": 60.98,
    "text": "alpha."
  },
  {
    "start": 62.98,
    "end": 63.48,
    "text": "It's a bit rough."
  },
  {
    "start": 64.56,
    "end": 69.56,
    "text": "There is a motion sensor there, which is hanging out of the wall and everything's half finished."
  },
  {
    "start": 69.6,
    "end": 70.2,
    "text": "So bear with me."
  },
  {
    "start": 71.72,
    "end": 85.4,
    "text": "And to not talk about me, but my guest today, I'm super honored to have Matthias Keller with me, a German guy living in the US, working for a company that everyone here knows called Kayak."
  },
  {
    "start": 85.64,
    "end": 87.28,
    "text": "And he's the Chief Product Officer."
  },
  {
    "start": 88.5,
    "end": 90.54,
    "text": "So not the CTO, but the CPO."
  },
  {
    "start": 91.68,
    "end": 105.58,
    "text": "But he has a long lasting history and tech studied at, or has a PhD in computer engineering from ETH Zurich, and a master in computer science from TU Munich, which is crazy."
  },
  {
    "start": 107.42,
    "end": 112.4,
    "text": "You have almost all education that a computer scientist could think of."
  },
  {
    "start": 114.0,
    "end": 117.24,
    "text": "And yeah, you're with Kayak for twelve years already, Matthias."
  },
  {
    "start": 118.84,
    "end": 119.8,
    "text": "How did you survive that?"
  },
  {
    "start": 121.02,
    "end": 124.52,
    "text": "First of all, thank you for having me and thank you for a very warm introduction."
  },
  {
    "start": 125.72,
    "end": 127.14,
    "text": "How did I survive twelve years?"
  },
  {
    "start": 127.2,
    "end": 133.86,
    "text": "You know, the interesting thing is that in good things and maybe scary for some is that no year has been like the other."
  },
  {
    "start": 134.26,
    "end": 146.04,
    "text": "So it never got boring in a way because every year was not only like some changes in roles and responsibilities, but also like how the environment changed within these twelve years."
  },
  {
    "start": 146.14,
    "end": 148.4,
    "text": "We had the pandemic, of course."
  },
  {
    "start": 149.36,
    "end": 157.02,
    "text": "When I started at Kayak, Google flights barely was a thing and Google just purchased this company not known to many called ITA software."
  },
  {
    "start": 159.22,
    "end": 161.86,
    "text": "Airbnb I think became big in between, right?"
  },
  {
    "start": 161.92,
    "end": 171.04,
    "text": "Like travel moved a lot and so these these twelve years have never been boring and I'm also not afraid that the next years are going to be boring either."
  },
  {
    "start": 172.14,
    "end": 180.48,
    "text": "Yeah, I hope it will be a bit more boring for you, but I don't know if there's anything that can scare you these days."
  },
  {
    "start": 181.7,
    "end": 189.14,
    "text": "And before you did travel, you actually started measuring wheel vibrations on high-speed trains before you joined."
  },
  {
    "start": 189.2,
    "end": 190.04,
    "text": "Is that correct?"
  },
  {
    "start": 191.96,
    "end": 199.96,
    "text": "Yeah, so I think for my professional background, to explain a little bit before going... too bad i i i guess."
  },
  {
    "start": 200.02,
    "end": 212.22,
    "text": "when i really started into so i've always been building things with desire for distribution having people who can who can use these great creations and the web was of course a wonderful place to be there."
  },
  {
    "start": 213.16,
    "end": 217.42,
    "text": "and so i actually started off as many of our age i would say as a php developer."
  },
  {
    "start": 218.26,
    "end": 222.56,
    "text": "so i i did in my life as a student a lot of php."
  },
  {
    "start": 223.8,
    "end": 231.94,
    "text": "We also had this category of web pages that would take pictures from people who went to clubs because they didn't have smartphones with cameras yet."
  },
  {
    "start": 232.72,
    "end": 252.12,
    "text": "I did that for a while and then when I got a little bit more serious about electrical engineering I did some internship and working as a student with back then Daimler Chrysler and through that I got connected to folks who just all other Germans who just moved down to Australia to work for this company who was doing true."
  },
  {
    "start": 252.76,
    "end": 254.2,
    "text": "train maintenance and robotics."
  },
  {
    "start": 254.5,
    "end": 262.34,
    "text": "so the company itself was doing things like trains that drove itself through the desert to carry coal from the mines to the ports."
  },
  {
    "start": 263.54,
    "end": 273.76,
    "text": "i was on a still interesting but but less terminator sounding project which was helping the the some someone in i think the british government or some british city."
  },
  {
    "start": 273.82,
    "end": 282.14,
    "text": "they just had ordered a bunch of high speed trains and with that they ordered the package of maintenance and predictive maintenance and and so i was really working on lasers."
  },
  {
    "start": 283.44,
    "end": 295.7,
    "text": "lasers to project on the parts of the train and taking pictures and basically measuring the shadows of these lasers to find out if this train needs to go on maintenance or is still good to go."
  },
  {
    "start": 296.22,
    "end": 297.04,
    "text": "That sounds super deep."
  },
  {
    "start": 298.64,
    "end": 303.46,
    "text": "I mean, deeper as like the normal PHP developer kind of experience in his lifetime, right?"
  },
  {
    "start": 303.94,
    "end": 306.54,
    "text": "So, very cool."
  },
  {
    "start": 311.94,
    "end": 313.88,
    "text": "And then you decided for travel like."
  },
  {
    "start": 313.94,
    "end": 315.36,
    "text": "what was the idea behind that?"
  },
  {
    "start": 315.42,
    "end": 317.04,
    "text": "like where you'd like?"
  },
  {
    "start": 317.72,
    "end": 319.68,
    "text": "which challenge did you see back in the days?"
  },
  {
    "start": 320.82,
    "end": 327.22,
    "text": "Was it that like as you started in analytics that you were like filled by the amount of data floating through?"
  },
  {
    "start": 327.4,
    "end": 328.62,
    "text": "or what was the idea?"
  },
  {
    "start": 330.1,
    "end": 331.82,
    "text": "Yeah, I really I got in through the data."
  },
  {
    "start": 332.04,
    "end": 334.8,
    "text": "so to some I guess some more perspective here is."
  },
  {
    "start": 336.7,
    "end": 341.82,
    "text": "When I was in Switzerland in Zurich we were putting wireless sensors on mountains to."
  },
  {
    "start": 342.96,
    "end": 345.22,
    "text": "Idea was to predict rock falls."
  },
  {
    "start": 345.48,
    "end": 349.4,
    "text": "interesting enough ten years later we sadly have a lot of these happening right now if you read the news."
  },
  {
    "start": 349.46,
    "end": 352.88,
    "text": "but but like predicting earthquakes we were trying to predict rock falls."
  },
  {
    "start": 352.98,
    "end": 353.78,
    "text": "so it was wireless."
  },
  {
    "start": 353.84,
    "end": 354.98,
    "text": "it was a lot of data but."
  },
  {
    "start": 355.6,
    "end": 359.48,
    "text": "But there was not really an opportunity to just continue doing it necessarily."
  },
  {
    "start": 359.64,
    "end": 363.98,
    "text": "maybe we could have founded a company which which some of my co students did but."
  },
  {
    "start": 364.54,
    "end": 367.6,
    "text": "There weren't like a lot of wireless sensor network jobs in Switzerland at the time."
  },
  {
    "start": 368.32,
    "end": 370.46,
    "text": "And I'm not sure how many of them are around today."
  },
  {
    "start": 370.6,
    "end": 372.92,
    "text": "So I had to readjust."
  },
  {
    "start": 373.04,
    "end": 375.54,
    "text": "And I thought think about a new area."
  },
  {
    "start": 375.62,
    "end": 377.02,
    "text": "And I always liked travel."
  },
  {
    "start": 377.3,
    "end": 389.88,
    "text": "And at the time, there was the opportunity to join this little not known to many company called Kayak, which at the time had just closed their acquisition of a very well known German page called Swudu."
  },
  {
    "start": 391.36,
    "end": 397.82,
    "text": "which used to have its office in Munich and, yeah, that's what I would say was my beginning in travel."
  },
  {
    "start": 398.48,
    "end": 407.8,
    "text": "and coming back to my PHP developer times is that, you know, there was definitely a very different scale than anything that I had touched before in my life."
  },
  {
    "start": 408.68,
    "end": 413.14,
    "text": "It was not PHP anymore, but a massively different scale."
  },
  {
    "start": 413.56,
    "end": 414.34,
    "text": "and also learning."
  },
  {
    "start": 416.14,
    "end": 427.24,
    "text": "United States American way of thinking about engineering and how these organizations are built, which twelve years ago is something I didn't really have insight into and still excites me today."
  },
  {
    "start": 428.22,
    "end": 433.78,
    "text": "If you look at your current current job as a CPO at Kayak, what's your biggest"
  },
  {
    "start": 433.94,
    "end": 434.24,
    "text": "CPO?"
  },
  {
    "start": 434.72,
    "end": 437.04,
    "text": "CPO, not CTO."
  },
  {
    "start": 438.12,
    "end": 440.32,
    "text": "What's your biggest challenge in business right now?"
  },
  {
    "start": 441.44,
    "end": 455.32,
    "text": "You know, I think that there are a number of challenges that have been around for a long time and one is that buying travel for the consumer has always been becoming more complicated than easier."
  },
  {
    "start": 456.12,
    "end": 466.58,
    "text": "So like twelve years ago when i started i would say when you try to buy an airfare you had to pick between economy business and first and today with the unbundling that happened."
  },
  {
    "start": 467.78,
    "end": 471.96,
    "text": "Now there's suddenly a ticket where you can't pick your seat anymore or a ticket where you can't bring it back."
  },
  {
    "start": 473.48,
    "end": 480.66,
    "text": "If you buy it on this side the back cost fifty dollars but if you buy it over there cost eighty dollars and it got really complicated."
  },
  {
    "start": 480.86,
    "end": 481.02,
    "text": "right."
  },
  {
    "start": 481.1,
    "end": 485.66,
    "text": "it has of course to do with how the industry evolved and how they found new revenue streams and."
  },
  {
    "start": 486.54,
    "end": 493.1,
    "text": "And you know selling more customized to the needs of the consumer but it hasn't made the life easier for our users."
  },
  {
    "start": 493.38,
    "end": 496.58,
    "text": "it has made the life easier for us in terms of the data we have to process."
  },
  {
    "start": 497.1,
    "end": 501.2,
    "text": "Like if instead of three tickets you now have maybe an average i don't know fifteen or so."
  },
  {
    "start": 501.36,
    "end": 503.96,
    "text": "right like all the different combinations and how they can be bundled."
  },
  {
    "start": 504.0,
    "end": 505.56,
    "text": "that makes a lot more data of course."
  },
  {
    "start": 506.84,
    "end": 515.34,
    "text": "But it's also very confusing for the consumer especially if they don't at the time of booking if they don't understand something like they don't understand that there is no back included."
  },
  {
    "start": 516.419,
    "end": 520.039,
    "text": "If they feel kind of like betrayed when they come to the airport and suddenly there's a fee for a back."
  },
  {
    "start": 520.78,
    "end": 528.2,
    "text": "And so the ongoing challenge here is that we don't only do flights, we also do hotels, rental cars and packages, which is flight and hotel bundled."
  },
  {
    "start": 529.52,
    "end": 540.54,
    "text": "They all have very unique challenges of new regulations, all kind of twists, a rental car that you need to fuel up before you bring it back, a rental car that you don't have to fuel up, that you have to fuel up to the amount you got it."
  },
  {
    "start": 540.66,
    "end": 548.92,
    "text": "And all in all, there's so much data and just allowing our customers to make that comparison is really challenging in this."
  },
  {
    "start": 549.44,
    "end": 550.74,
    "text": "constantly changing environment."
  },
  {
    "start": 550.88,
    "end": 561.54,
    "text": "And then from a newer challenge, just to certainly not to have to start talking about that is now certainly the advent of AI and what that does to all of us."
  },
  {
    "start": 562.04,
    "end": 571.12,
    "text": "And for us, it is a super interesting opportunity because we now can use also AI to make our products smarter and better."
  },
  {
    "start": 572.44,
    "end": 575.92,
    "text": "But certainly, there are now also completely new players entering the arena."
  },
  {
    "start": 576.7,
    "end": 587.64,
    "text": "that are also all understanding that travel is a very interesting business and that basically travel is one of the most sought after categories in anyone's online searches."
  },
  {
    "start": 589.76,
    "end": 594.22,
    "text": "And how do you feel about the competitive landscape?"
  },
  {
    "start": 595.24,
    "end": 598.68,
    "text": "I guess that also increased pressure quite a lot."
  },
  {
    "start": 599.98,
    "end": 601.7,
    "text": "I mean, you just mentioned Google Flights, right?"
  },
  {
    "start": 601.8,
    "end": 605.94,
    "text": "The small acquisition that turned out really as a monster, I can imagine."
  },
  {
    "start": 607.48,
    "end": 608.68,
    "text": "or kind of a monster."
  },
  {
    "start": 610.26,
    "end": 611.64,
    "text": "How does that play in?"
  },
  {
    "start": 613.68,
    "end": 623.66,
    "text": "You kind of tend to have the hybris of being very on top of the bar."
  },
  {
    "start": 625.44,
    "end": 626.66,
    "text": "How do you keep track with that?"
  },
  {
    "start": 627.26,
    "end": 630.38,
    "text": "How do you maintain your position there?"
  },
  {
    "start": 632.02,
    "end": 641.14,
    "text": "Yeah, so certainly our focus on only meta search makes us innovate faster and thinking of problems earlier than others."
  },
  {
    "start": 641.26,
    "end": 643.78,
    "text": "and just one example is a fun fact."
  },
  {
    "start": 646.16,
    "end": 656.72,
    "text": "Today when you go to many products including other products, you might look for a ticket and instead of one round trip, they give you two one ways on two different airlines and that saves you money."
  },
  {
    "start": 658.28,
    "end": 660.96,
    "text": "When we introduced that, we call it since today, we call it a heck of air."
  },
  {
    "start": 661.76,
    "end": 665.16,
    "text": "But that's just one example of the innovations that we had first."
  },
  {
    "start": 666.28,
    "end": 671.36,
    "text": "And there are many more such as having the first baggage recalculator out there."
  },
  {
    "start": 672.48,
    "end": 673.66,
    "text": "Having the first product out there."
  },
  {
    "start": 673.7,
    "end": 682.24,
    "text": "There was a time in Germany when in other places where suddenly there was a random fee depending on what kind of credit card you used to pay with something."
  },
  {
    "start": 682.44,
    "end": 686.98,
    "text": "This was the time when everyone had to go on support because everything else was so expensive."
  },
  {
    "start": 687.04,
    "end": 694.44,
    "text": "So it's definitely innovation is really in our DNA because we look at that specific travel problem so much."
  },
  {
    "start": 696.14,
    "end": 702.38,
    "text": "I think where a challenge comes in is then certainly on the other side, we are facing a lot of players."
  },
  {
    "start": 702.9,
    "end": 707.18,
    "text": "First of all, differentiation certainly only keeps up for a certain time."
  },
  {
    "start": 707.86,
    "end": 710.74,
    "text": "If you have good ideas, other people may implement the same good ideas."
  },
  {
    "start": 711.72,
    "end": 713.5,
    "text": "And there's, of course, a challenge of distribution."
  },
  {
    "start": 715.38,
    "end": 722.92,
    "text": "It is obvious to everyone that Google.com is a lot larger than any other website on the world, including ours."
  },
  {
    "start": 723.94,
    "end": 728.88,
    "text": "And we see chat GPT coming up as one very new of these big sites."
  },
  {
    "start": 729.9,
    "end": 733.94,
    "text": "And certainly a lot of users start all their questions there."
  },
  {
    "start": 735.9,
    "end": 751.56,
    "text": "And on Google, I think all travel players, there has been a healthy ecosystem of SEO and paid advertisement on how Google takes the original intent, maybe refines it and sends it off to others like us with who can give the specific answers."
  },
  {
    "start": 752.62,
    "end": 754.9,
    "text": "We don't know yet today how chat GPT is going to work."
  },
  {
    "start": 755.6,
    "end": 759.28,
    "text": "to my understanding There is no paid advertising model or anything like that right now."
  },
  {
    "start": 759.4,
    "end": 766.66,
    "text": "It is more of a black box who shows up where This this may or may not change over time."
  },
  {
    "start": 766.7,
    "end": 768.48,
    "text": "I mean, that's something for open AI to ask."
  },
  {
    "start": 769.32,
    "end": 771.86,
    "text": "um, but but certainly there there is the."
  },
  {
    "start": 774.04,
    "end": 785.1,
    "text": "There is the one part of the equation which is being really deep in meta search really innovating around that problem connecting hundreds of different partners to have the most comprehensive comparison."
  },
  {
    "start": 787.0,
    "end": 802.1,
    "text": "On the other side, then there is the distribution of the users understanding this differentiation, understanding this value and using the product because they understand how it helps them to save money by comparing in the biggest scope possible."
  },
  {
    "start": 804.06,
    "end": 812.88,
    "text": "How do you, we also today want to talk about platforms and maybe we can even a few of the CPU also deep dive a bit more on challenges and travel and tech."
  },
  {
    "start": 814.26,
    "end": 821.3,
    "text": "But if you zoom out and move like, I don't know, when was it like, twenty sixteen, twenty fifteen?"
  },
  {
    "start": 822.4,
    "end": 830.9,
    "text": "When Alexa first hit, was that like, I guess it wasn't the similar feeling before because adoption was never."
  },
  {
    "start": 832.22,
    "end": 836.32,
    "text": "was never that much happening, I think, at least from my perspective."
  },
  {
    "start": 836.76,
    "end": 842.48,
    "text": "I mean, I remember there was some folks jumping on it, and you also jumped on it, right?"
  },
  {
    "start": 843.0,
    "end": 847.6,
    "text": "You published your Alexa skills, but I think it never took off."
  },
  {
    "start": 848.26,
    "end": 849.12,
    "text": "How does that compare?"
  },
  {
    "start": 850.88,
    "end": 863.02,
    "text": "So, you know, it was super interesting from a perspective that, so first of all, you know, I agree with you that that any sort of platform that requires you to buy a device is challenging."
  },
  {
    "start": 863.16,
    "end": 865.52,
    "text": "right and we right now see that with the apple vision pro."
  },
  {
    "start": 866.76,
    "end": 875.24,
    "text": "it's a gorgeous device but just not a lot of people have it and that really limits your distribution when people need to buy a three and a half thousand dollar headset."
  },
  {
    "start": 876.18,
    "end": 878.08,
    "text": "we have other examples with devices like the iphone."
  },
  {
    "start": 878.2,
    "end": 878.78,
    "text": "everyone has it."
  },
  {
    "start": 879.8,
    "end": 883.32,
    "text": "so this one worked out now with alexa specifically."
  },
  {
    "start": 884.26,
    "end": 886.8,
    "text": "so the first start of this interesting time was that."
  },
  {
    "start": 887.52,
    "end": 892.42,
    "text": "Um, we actually before we started developing the skill or anyone could develop a skill."
  },
  {
    "start": 893.4,
    "end": 897.12,
    "text": "Actually a lot of people around us in the office or neighbors and whatnot."
  },
  {
    "start": 897.32,
    "end": 898.16,
    "text": "They all had this device."
  },
  {
    "start": 898.2,
    "end": 899.32,
    "text": "I'm very excited about it."
  },
  {
    "start": 899.96,
    "end": 901.8,
    "text": "Like it was very entry level price point."
  },
  {
    "start": 902.32,
    "end": 905.16,
    "text": "It wasn't like the time where you would, you know, get it with every washing machine."
  },
  {
    "start": 905.2,
    "end": 908.64,
    "text": "You still had to buy it, but it was, it was entry level price point."
  },
  {
    "start": 908.74,
    "end": 912.9,
    "text": "It was the first time where you had the feeling that voice actually works after all the."
  },
  {
    "start": 913.52,
    "end": 922.1,
    "text": "bad experiences in car navigation and people got really excited about it for the use cases of the weather, the music, the reminders."
  },
  {
    "start": 922.94,
    "end": 929.38,
    "text": "And the challenge now is that if we get to the end of the story, unfortunately, that were the only use cases that really worked."
  },
  {
    "start": 930.44,
    "end": 932.42,
    "text": "The timer, the music, and the reminders."
  },
  {
    "start": 933.58,
    "end": 940.02,
    "text": "But in between, thinking of when the iPhone was there, it could have been sort of boom or bust."
  },
  {
    "start": 941.94,
    "end": 945.74,
    "text": "we had really great success being there early when the ipad was there."
  },
  {
    "start": 946.02,
    "end": 947.36,
    "text": "nobody knew who needs a tablet."
  },
  {
    "start": 947.74,
    "end": 953.0,
    "text": "we were there early and we think that it was really really great for for getting momentum and so."
  },
  {
    "start": 954.78,
    "end": 956.16,
    "text": "There was around twenty sixteen."
  },
  {
    "start": 956.22,
    "end": 958.2,
    "text": "so there was this whole group of alexa."
  },
  {
    "start": 958.3,
    "end": 959.26,
    "text": "look pretty interesting."
  },
  {
    "start": 959.48,
    "end": 964.46,
    "text": "um facebook messenger came out with their first way of writing your little bots."
  },
  {
    "start": 965.58,
    "end": 967.56,
    "text": "And it was only alexa."
  },
  {
    "start": 967.72,
    "end": 970.7,
    "text": "right like i was google out there with assistant."
  },
  {
    "start": 971.8,
    "end": 972.34,
    "text": "i can't remember what."
  },
  {
    "start": 972.58,
    "end": 974.38,
    "text": "i think microsoft was there as well."
  },
  {
    "start": 974.44,
    "end": 976.2,
    "text": "and then samsung had this thing called bixby."
  },
  {
    "start": 976.94,
    "end": 978.28,
    "text": "apple of course is serious so it."
  },
  {
    "start": 978.62,
    "end": 982.5,
    "text": "it looked like a lot of momentum and actually we."
  },
  {
    "start": 984.06,
    "end": 986.62,
    "text": "I still love the products we built back in the day."
  },
  {
    "start": 986.92,
    "end": 989.88,
    "text": "so it was a lot of travel research."
  },
  {
    "start": 991.08,
    "end": 992.02,
    "text": "it was a little bit of."
  },
  {
    "start": 993.84,
    "end": 996.24,
    "text": "Yeah just like you know how much would it cost to fly to turkey?"
  },
  {
    "start": 996.52,
    "end": 997.86,
    "text": "you know find me a hotel here or there."
  },
  {
    "start": 998.68,
    "end": 1002.1,
    "text": "Um we did some things that today you would call a genetic like."
  },
  {
    "start": 1002.16,
    "end": 1009.78,
    "text": "at the end of ours at the end phase of our skill you could go there and say something like i'm Alexa tell kayak and is already like it was clunky right."
  },
  {
    "start": 1009.92,
    "end": 1010.7,
    "text": "tell kayak to."
  },
  {
    "start": 1011.64,
    "end": 1016.92,
    "text": "Tell kayak to remind me twenty minutes before delta flight one two three lands."
  },
  {
    "start": 1017.66,
    "end": 1025.22,
    "text": "and so interesting thing here was like when this plane lands is it is a kind of dynamic time right it may be delayed it may be early it may be all that."
  },
  {
    "start": 1025.319,
    "end": 1027.68,
    "text": "so we had already the genetic process of."
  },
  {
    "start": 1028.24,
    "end": 1032.42,
    "text": "Something checking for you what's going on with that airplane and then making the estimation."
  },
  {
    "start": 1032.48,
    "end": 1038.72,
    "text": "Okay, it seems like that's in twenty minutes and Then you know the ring would light up and then there's notification for you."
  },
  {
    "start": 1039.64,
    "end": 1042.119,
    "text": "Um, I think it was super exciting."
  },
  {
    "start": 1042.18,
    "end": 1046.0,
    "text": "even still today We picked people up from the airport and we don't really know when to go there."
  },
  {
    "start": 1046.16,
    "end": 1048.4,
    "text": "so the the demand still exists."
  },
  {
    "start": 1048.5,
    "end": 1056.72,
    "text": "but But I think what was really challenging in that first generation to me was was not the amount of devices out there at least in the US it was."
  },
  {
    "start": 1057.5,
    "end": 1062.66,
    "text": "it was kind of clunky to personalization there and people just wouldn't understand that."
  },
  {
    "start": 1063.44,
    "end": 1066.28,
    "text": "they wouldn't understand that it worked like a phone book that you connected."
  },
  {
    "start": 1066.4,
    "end": 1067.72,
    "text": "now i'm connected to kayak."
  },
  {
    "start": 1067.82,
    "end": 1068.64,
    "text": "now i'm connected to."
  },
  {
    "start": 1068.8,
    "end": 1069.62,
    "text": "you know the car wash."
  },
  {
    "start": 1070.1,
    "end": 1078.76,
    "text": "people just wanted to ask for another song in between and i think there were a lot of early voice to x challenges together with."
  },
  {
    "start": 1079.72,
    "end": 1082.6,
    "text": "It really being hard for us to find out something like where you are."
  },
  {
    "start": 1082.86,
    "end": 1086.82,
    "text": "you like you would go into the app first to to give us a permission to see your location."
  },
  {
    "start": 1086.92,
    "end": 1101.34,
    "text": "and You know at the end of the day we had the hotel booking on voice running which which was really a great demo and and I really still missed the flight notification, but yeah, it feels like that right now."
  },
  {
    "start": 1101.72,
    "end": 1102.98,
    "text": "I don't know anyone anymore using it"
  },
  {
    "start": 1106.26,
    "end": 1109.04,
    "text": "and How."
  },
  {
    "start": 1109.5,
    "end": 1116.3,
    "text": "how does that compared to today and the LLM world and what is your take on that?"
  },
  {
    "start": 1116.64,
    "end": 1132.44,
    "text": "I saw that you actually released an LLM based search in your product, which allows basically like natural language queries and real LLM queries basically."
  },
  {
    "start": 1134.34,
    "end": 1156.98,
    "text": "What's your take on you potentially or the world changing towards players, like you potentially having to connect to other user interfaces or having to deliver it into other user interfaces instead of serving things on your own UI."
  },
  {
    "start": 1159.56,
    "end": 1164.7,
    "text": "Yeah, that's really the million dollar question right now, so to speak."
  },
  {
    "start": 1165.02,
    "end": 1172.88,
    "text": "I think, first of all, if I think about the situation in general with LLMs today, the entrance barrier for the user is extremely low, right?"
  },
  {
    "start": 1172.96,
    "end": 1174.36,
    "text": "It's only as a browser."
  },
  {
    "start": 1175.42,
    "end": 1176.26,
    "text": "Everyone has a browser."
  },
  {
    "start": 1177.5,
    "end": 1185.6,
    "text": "I think the second thing that is much better to date than with Alexa is that with Alexa, you always have to think about if I ask this thing, is it going to work?"
  },
  {
    "start": 1186.28,
    "end": 1188.24,
    "text": "You often had, it didn't work, right?"
  },
  {
    "start": 1188.3,
    "end": 1190.56,
    "text": "Which also made you sensitive about it may not work."
  },
  {
    "start": 1190.66,
    "end": 1191.34,
    "text": "How do I ask?"
  },
  {
    "start": 1191.4,
    "end": 1192.06,
    "text": "Can I ask that?"
  },
  {
    "start": 1193.38,
    "end": 1197.22,
    "text": "While I feel for most consumers on the LLM experiences, it just works."
  },
  {
    "start": 1199.0,
    "end": 1200.48,
    "text": "And so that whole barrier is gone."
  },
  {
    "start": 1201.48,
    "end": 1202.4,
    "text": "And the destination"
  },
  {
    "start": 1202.44,
    "end": 1203.56,
    "text": "is often ignored, right?"
  },
  {
    "start": 1205.76,
    "end": 1208.54,
    "text": "And I feel like people don't maybe accept them, right?"
  },
  {
    "start": 1208.58,
    "end": 1210.58,
    "text": "But I don't see so many people talk about them anymore."
  },
  {
    "start": 1212.2,
    "end": 1213.04,
    "text": "Definitely exists."
  },
  {
    "start": 1213.08,
    "end": 1214.5,
    "text": "There are some really great examples there."
  },
  {
    "start": 1214.54,
    "end": 1227.44,
    "text": "Now, for our journey with LLMs is that we actually tried that input text box also back in twenty sixteen with the same technology like Alexa and users were just not ready to enter free text."
  },
  {
    "start": 1228.78,
    "end": 1236.52,
    "text": "But so right now, I think it's about two years ago, we... OpenAI launched these... Back in the day, they were called plugins."
  },
  {
    "start": 1236.9,
    "end": 1238.46,
    "text": "Later, they were renamed into GPTs."
  },
  {
    "start": 1239.26,
    "end": 1254.98,
    "text": "And I don't know what their future is now that they are the apps, but we added a plugin, where basically on chat GPT, you could go in there, say that you want to use Kayak as like a tool, and search for flights, hotels, and cars."
  },
  {
    "start": 1257.36,
    "end": 1270.12,
    "text": "We then later brought... This was early this year, we launched kaya.ai, which is a from the ground up build AI first conversational travel experience, focused on travel."
  },
  {
    "start": 1271.56,
    "end": 1281.62,
    "text": "If you think about the early times of using chat GPT is that you would ask for flight prices, it would usually come back with some price that it found on some SEO page."
  },
  {
    "start": 1282.02,
    "end": 1285.36,
    "text": "Could be ours in many cases or one of some of our competitors."
  },
  {
    "start": 1286.86,
    "end": 1295.64,
    "text": "Whereas when you go to Kayak AI and you ask for prices, you get actual life itineraries with a life price that are one click away from being booked."
  },
  {
    "start": 1298.24,
    "end": 1301.46,
    "text": "To get to your question, we believe a lot in travel."
  },
  {
    "start": 1302.08,
    "end": 1309.4,
    "text": "We believe a lot in the dual experience of having real availability, real prices, all the information and the ability to book."
  },
  {
    "start": 1310.68,
    "end": 1320.76,
    "text": "And then the latest thing that is out there, what we call AI mode is the input box everywhere on our, you don't have to go to this AI corner anymore, it's on all"
  },
  {
    "start": 1320.86,
    "end": 1321.32,
    "text": "our products."
  },
  {
    "start": 1323.42,
    "end": 1343.74,
    "text": "Now, the question with the distribution, right, is that in a way I start thinking of, first way I start thinking of when I think about Google, right, Google would have their native hotel widget, which is a little map that shows all the hotels with a lot of Google data, but actually all the, the prices behind would be backed by by hotel chains."
  },
  {
    "start": 1343.9,
    "end": 1344.86,
    "text": "right so they have."
  },
  {
    "start": 1345.08,
    "end": 1351.9,
    "text": "they drew the line about making the widget, then the content in the widget you still click out and finish your booking in another place."
  },
  {
    "start": 1354.26,
    "end": 1361.1,
    "text": "When we go into certainly into the complexity or the gpt model we definitely see that check out happens more on their sites."
  },
  {
    "start": 1361.9,
    "end": 1368.48,
    "text": "Let's say if you look at the hotel booking on complexity or if you look at the walmart integration and the Shopify integrations they all happen on their sites."
  },
  {
    "start": 1369.34,
    "end": 1371.26,
    "text": "A few things here."
  },
  {
    "start": 1371.54,
    "end": 1379.66,
    "text": "I think the first thing to realize is that selling a t-shirt or a tea kettle is much easier than selling a flight."
  },
  {
    "start": 1380.74,
    "end": 1391.46,
    "text": "Coming back to the original problem of all these customizations that are happening there and having miles or not having miles and all these different categories and how you merchandise them."
  },
  {
    "start": 1391.6,
    "end": 1396.94,
    "text": "It's also really important for airlines that they don't want you necessarily to go with the cheapest ticket."
  },
  {
    "start": 1396.98,
    "end": 1399.4,
    "text": "They would like you to get a nice receipt for you and for them."
  },
  {
    "start": 1400.14,
    "end": 1404.34,
    "text": "Upsell is simply said, upsell is extremely important."
  },
  {
    "start": 1404.76,
    "end": 1411.52,
    "text": "So everyone who is there online selling something, even if it's a t-shirt, they also want to sell you a second t-shirt or maybe a sweater."
  },
  {
    "start": 1412.42,
    "end": 1418.56,
    "text": "And we have to see how well that is going to happen when the agent or that interface goes buying for you."
  },
  {
    "start": 1419.8,
    "end": 1423.44,
    "text": "We have the additional complexity with the flights."
  },
  {
    "start": 1426.58,
    "end": 1432.88,
    "text": "I think another point is that especially for higher purchases, people actually like to compare."
  },
  {
    "start": 1433.74,
    "end": 1444.34,
    "text": "So I think a place that has, I think it would have to be a meta experience where you can compare different flights from different sites instead of just saying, okay, eight hundred dollars, let's go."
  },
  {
    "start": 1446.1,
    "end": 1456.62,
    "text": "And the third part I want to bring up here, and I'm not saying that it's never going to happen, but I think we also shouldn't underestimate the commercial complexity behind all these decisions."
  },
  {
    "start": 1457.34,
    "end": 1461.6,
    "text": "So for example, American carriers are most of their content."
  },
  {
    "start": 1461.64,
    "end": 1464.5,
    "text": "You can only find them on their actual apps and web pages."
  },
  {
    "start": 1464.68,
    "end": 1466.16,
    "text": "That's the only place where you can do that."
  },
  {
    "start": 1466.86,
    "end": 1475.3,
    "text": "And they do that out of the commercial decision that they want you to get into their ecosystem."
  },
  {
    "start": 1476.78,
    "end": 1483.28,
    "text": "If you have seen these articles, a lot of American allies actually make almost more money with their credit card programs than with selling airfares."
  },
  {
    "start": 1483.8,
    "end": 1485.88,
    "text": "So there's a whole ecosystem that they want you to get into."
  },
  {
    "start": 1486.24,
    "end": 1496.74,
    "text": "And if you now think about people buying on JetGPT, you know, starting from subscribing to your newsletter, you really don't want to lose that ecosystem."
  },
  {
    "start": 1497.66,
    "end": 1505.02,
    "text": "But aren't people, like, isn't the core problem that people are lazy and whenever you make it the easiest for them, then at a certain point they change their behavior?"
  },
  {
    "start": 1505.66,
    "end": 1507.6,
    "text": "That's at least how I see it."
  },
  {
    "start": 1508.56,
    "end": 1529.8,
    "text": "And then to your point with the commercial barriers, I think, or I think what is often to like observable is that at a certain point, like you first have like the firewall that like no one gets passed, right?"
  },
  {
    "start": 1530.24,
    "end": 1553.14,
    "text": "That no one can pass and all companies like protect themselves from basically changing behavior and at a certain point someone like one of them is actually so attracted by that ecosystem that on the other side that He or she Adopted and jumps in and then all the others also follow right and I don't know."
  },
  {
    "start": 1553.5,
    "end": 1563.16,
    "text": "I think like Apple CarPlay for example has this new new interface that I don't know how it's called actually"
  },
  {
    "start": 1563.42,
    "end": 1563.9,
    "text": "but"
  },
  {
    "start": 1564.02,
    "end": 1569.14,
    "text": "like it takes over all instruments in the car and the car industry obviously doesn't want that."
  },
  {
    "start": 1569.4,
    "end": 1578.72,
    "text": "but at a certain point I see that coming or someone like one player maybe from china or whatever actually wants to get the user the best experience and actually uses that."
  },
  {
    "start": 1578.82,
    "end": 1589.64,
    "text": "and then all of a sudden all of us follow and i could imagine that this also could happen in that industry but i don't know the industry well enough to predict that."
  },
  {
    "start": 1591.34,
    "end": 1592.24,
    "text": "So certainly our."
  },
  {
    "start": 1592.78,
    "end": 1599.72,
    "text": "in a recent interview a few months ago our our ceo was was asked a similar question and he described it as a dam breaking."
  },
  {
    "start": 1600.34,
    "end": 1603.14,
    "text": "is that like that's gonna be a dam breaking as soon as the first deal is done."
  },
  {
    "start": 1603.88,
    "end": 1605.78,
    "text": "And I agree with you from a perspective."
  },
  {
    "start": 1605.84,
    "end": 1614.48,
    "text": "It's nice to say like, I'm never going to have my stuff out there until your competitor who may have, you know, investor pressure or whatever reasons they have is out there."
  },
  {
    "start": 1615.2,
    "end": 1617.92,
    "text": "And suddenly you start wondering, hmm, this is interesting, right?"
  },
  {
    "start": 1618.5,
    "end": 1620.86,
    "text": "Do I want them to be the only thing on the shelf?"
  },
  {
    "start": 1622.18,
    "end": 1624.26,
    "text": "or maybe should I look at this shelf as well?"
  },
  {
    "start": 1626.4,
    "end": 1627.4,
    "text": "Yeah, absolutely."
  },
  {
    "start": 1627.74,
    "end": 1630.96,
    "text": "And then I don't know."
  },
  {
    "start": 1631.0,
    "end": 1644.74,
    "text": "the gates to hell open up right but i don't know if it has to be so bad and i am also with you that like the complexity of flights must be so high that it there."
  },
  {
    "start": 1644.78,
    "end": 1649.6,
    "text": "I sense that there will be space for mediators that kind of."
  },
  {
    "start": 1650.46,
    "end": 1651.98,
    "text": "Manage that complexity right."
  },
  {
    "start": 1653.24,
    "end": 1656.08,
    "text": "and i mean it's also."
  },
  {
    "start": 1656.68,
    "end": 1665.52,
    "text": "unrealistic that one platform has an easy chance to go against all commercial barriers."
  },
  {
    "start": 1666.46,
    "end": 1674.26,
    "text": "I just read yesterday that Wikipedia is basically charging LLMs for their data."
  },
  {
    "start": 1678.18,
    "end": 1679.94,
    "text": "Wikipedia has a reason to exist."
  },
  {
    "start": 1681.1,
    "end": 1689.94,
    "text": "It has a really big reason to exist and to manage this sheer amount of content and moderate that."
  },
  {
    "start": 1690.48,
    "end": 1692.92,
    "text": "And I think this is potentially what will change."
  },
  {
    "start": 1693.04,
    "end": 1718.34,
    "text": "Like maybe you don't get the traffic anymore, like the entries are traffic, but there is a reason to actually pay you for the data and pay you for the accuracy of the data and not the accuracy, but the freshness of the data to maintain the whole beast and deliver it to the LLMs continuously."
  },
  {
    "start": 1718.44,
    "end": 1726.58,
    "text": "But yeah, I mean, it's really hard to keep track of who's using your data when and how often it's being displayed and whatever."
  },
  {
    "start": 1726.72,
    "end": 1728.78,
    "text": "And the currency for that still doesn't exist, right?"
  },
  {
    "start": 1730.46,
    "end": 1732.28,
    "text": "Yeah, that's true."
  },
  {
    "start": 1732.62,
    "end": 1735.44,
    "text": "So I think a few things here to unpack."
  },
  {
    "start": 1735.52,
    "end": 1741.04,
    "text": "So the first certainly is like, I think it starts with news organization, right?"
  },
  {
    "start": 1741.32,
    "end": 1748.02,
    "text": "If they are just here to get a little bit of free traffic until the new lm training happened that's not gonna be very sustainable for them."
  },
  {
    "start": 1748.1,
    "end": 1751.28,
    "text": "right like if if you're if you're unique knowledge tomorrow is a commodity."
  },
  {
    "start": 1751.9,
    "end": 1758.06,
    "text": "it will be a hard game catching up delivering always unique content and it's it's you won't see the reward here."
  },
  {
    "start": 1758.74,
    "end": 1761.56,
    "text": "but but out of sight of these asio challenges i think what is."
  },
  {
    "start": 1762.76,
    "end": 1769.14,
    "text": "What we definitely see is that consumers use chat gpt and civil tools a lot for doing research."
  },
  {
    "start": 1769.3,
    "end": 1769.4,
    "text": "so."
  },
  {
    "start": 1770.02,
    "end": 1775.5,
    "text": "The consumer right now today who clicks out to your site and you know could be a tv could be a car could be a flight."
  },
  {
    "start": 1776.18,
    "end": 1777.7,
    "text": "Who has been on these platforms?"
  },
  {
    "start": 1778.24,
    "end": 1782.08,
    "text": "this consumer is gonna convert faster because they already did all their research right."
  },
  {
    "start": 1782.7,
    "end": 1787.56,
    "text": "they went through all the different types of screens and you know airlines and connections and this and that."
  },
  {
    "start": 1788.36,
    "end": 1796.0,
    "text": "And so they're gonna convert faster than someone who let's say comes from a from like a more simple google search like a one liner so to speak."
  },
  {
    "start": 1799.66,
    "end": 1801.58,
    "text": "I think this is."
  },
  {
    "start": 1801.9,
    "end": 1809.78,
    "text": "Still shows that it can coexist with researching what I want, and then I want to actually still compare and find out where to get the best price."
  },
  {
    "start": 1810.28,
    "end": 1811.44,
    "text": "That's two parts."
  },
  {
    "start": 1811.96,
    "end": 1817.34,
    "text": "I want this television and the next question is, how can I get it tomorrow at the best price?"
  },
  {
    "start": 1818.56,
    "end": 1829.82,
    "text": "I think the second angle here is that, and this is really interesting, I think for someone like us and many others, when you think about LMS, they certainly operate these new front doors of the internet."
  },
  {
    "start": 1830.28,
    "end": 1835.46,
    "text": "But they also give us these completely new tools and rails to supercharge our products as well."
  },
  {
    "start": 1836.6,
    "end": 1845.6,
    "text": "Coming back with the theme of comparison, if you think about it right now, let's say you are comparing hotels, right?"
  },
  {
    "start": 1846.28,
    "end": 1847.08,
    "text": "And there is a hotel."
  },
  {
    "start": 1847.84,
    "end": 1850.4,
    "text": "All of them have four stars because that's what you want."
  },
  {
    "start": 1850.7,
    "end": 1855.82,
    "text": "All of them are somehow in a similar place, similar price, you know, fifty dollars up and down."
  },
  {
    "start": 1858.1,
    "end": 1876.9,
    "text": "You know, you can now start reading all these reviews and really find out where you like the pool more or where you like the View more or whatever it is, right and out of restaurant and and this is where LLMs really come in handy and and us being able to support the user with Like one sentence that tells you all about this hotel."
  },
  {
    "start": 1877.16,
    "end": 1877.92,
    "text": "What's the vibe here?"
  },
  {
    "start": 1878.12,
    "end": 1878.48,
    "text": "You know, what?"
  },
  {
    "start": 1878.62,
    "end": 1879.86,
    "text": "what do the reviews say in general?"
  },
  {
    "start": 1880.56,
    "end": 1886.62,
    "text": "and and of course we have all been doing sentiment analysis and clustering of reviews and all these things even before LLMs."
  },
  {
    "start": 1887.38,
    "end": 1893.76,
    "text": "But this makes it now a lot easier which also puts a lot of value on having data having unique data."
  },
  {
    "start": 1895.3,
    "end": 1898.86,
    "text": "And it becomes certainly this interesting data strategy as if you mentioned Wikipedia."
  },
  {
    "start": 1900.66,
    "end": 1907.86,
    "text": "Do I want to keep that unique data to me or do I want to give it to everyone and should that be for free or should that be for a fee."
  },
  {
    "start": 1911.3,
    "end": 1921.02,
    "text": "Coming back to you and your path there with AR mode now being on your home page and being like the I think the preferred way of searching, right?"
  },
  {
    "start": 1922.98,
    "end": 1925.98,
    "text": "How did that change the patterns in user behavior?"
  },
  {
    "start": 1926.06,
    "end": 1932.62,
    "text": "You mentioned that like people that come from Google search basically need a longer time to convert while the others like convert quicker."
  },
  {
    "start": 1934.38,
    "end": 1937.02,
    "text": "And how does that work on your own system?"
  },
  {
    "start": 1937.12,
    "end": 1937.76,
    "text": "Like how many?"
  },
  {
    "start": 1938.82,
    "end": 1938.94,
    "text": "How?"
  },
  {
    "start": 1940.42,
    "end": 1944.78,
    "text": "I don't know which KPIs you measure but I guess you'd like kind of the time to book."
  },
  {
    "start": 1944.9,
    "end": 1948.92,
    "text": "basically then I don't know what your north star is I mean."
  },
  {
    "start": 1950.7,
    "end": 1954.44,
    "text": "You tell me like how did that change those those patterns."
  },
  {
    "start": 1957.12,
    "end": 1958.0,
    "text": "And the behavior in general?"
  },
  {
    "start": 1959.84,
    "end": 1960.48,
    "text": "so I have to say."
  },
  {
    "start": 1962.14,
    "end": 1964.5,
    "text": "We don't have specific numbers to share here but."
  },
  {
    "start": 1965.14,
    "end": 1972.98,
    "text": "But we are well in line with what other people, including Google, are sharing that say that the engagement with AI features is still single digit percent."
  },
  {
    "start": 1974.34,
    "end": 1976.38,
    "text": "So most consumers are still living."
  },
  {
    "start": 1976.58,
    "end": 1987.0,
    "text": "It's also a very interesting part when we have this conversation about open AI becoming the new store is that it takes a really, really long time to change customer habit."
  },
  {
    "start": 1987.58,
    "end": 1990.78,
    "text": "Consumer habit, it may go quick at some point, but it takes a very long time."
  },
  {
    "start": 1991.58,
    "end": 1994.74,
    "text": "So right now it's a small group of users doing that."
  },
  {
    "start": 1995.62,
    "end": 2004.54,
    "text": "I think AI mode as it stands today, AI mode has a bit of the challenge that, it's not yet, we need to do a better job of explaining the user when they need it."
  },
  {
    "start": 2004.84,
    "end": 2011.06,
    "text": "And what I mean with that is, if you know that you want to fly to Berlin tomorrow, I'm not sure why you should deal with AI."
  },
  {
    "start": 2012.14,
    "end": 2015.74,
    "text": "you could just go to the form and you would probably be done by the time you finish typing."
  },
  {
    "start": 2017.54,
    "end": 2024.62,
    "text": "But if you have, you know, we distributed into the upper funnel and the mid funnel and the lower funnel, right?"
  },
  {
    "start": 2024.68,
    "end": 2036.52,
    "text": "And upper funnel being, if you're someone who just knows that I really should get out, but I don't know, like some people are fully flexible, doesn't matter to them when or where, and there are different levels of flexibility."
  },
  {
    "start": 2038.48,
    "end": 2040.2,
    "text": "AI is really powerful in that, right?"
  },
  {
    "start": 2040.4,
    "end": 2044.92,
    "text": "And it's also really powerful and will get better when we think about complex trips."
  },
  {
    "start": 2045.26,
    "end": 2051.38,
    "text": "Like you want to do this one world tour of six cities in eight days or something like that."
  },
  {
    "start": 2051.88,
    "end": 2053.26,
    "text": "There's a lot of searching for you right now."
  },
  {
    "start": 2056.06,
    "end": 2058.28,
    "text": "And our job is to educate the user."
  },
  {
    "start": 2059.26,
    "end": 2061.28,
    "text": "when is this sidekick going to help you?"
  },
  {
    "start": 2062.179,
    "end": 2066.04,
    "text": "What is the right time to go here versus just saying I need a hotel in Paris tomorrow?"
  },
  {
    "start": 2067.62,
    "end": 2072.82,
    "text": "Kayak AI, which is our native AI product, there's no forms, it's just a text box."
  },
  {
    "start": 2074.82,
    "end": 2082.96,
    "text": "What we really see there, what is super interesting is that it's really eye-opening that suddenly every user intent is possible."
  },
  {
    "start": 2083.58,
    "end": 2087.08,
    "text": "So you can now see things like gaps in your offering."
  },
  {
    "start": 2088.28,
    "end": 2106.32,
    "text": "For example, users asking about trains, for example, we know the train side, or asking about some... We are conversations where users are trying to get, they want to plan everything around that one last ferry that brings them to a certain island, somewhere up in Sweden."
  },
  {
    "start": 2108.32,
    "end": 2119.62,
    "text": "And it's really first of all interesting to see what users are trying to solve, but it's also interesting now to see how they feel about it in terms of the behavior that they would ask."
  },
  {
    "start": 2119.72,
    "end": 2120.64,
    "text": "why a lot, right?"
  },
  {
    "start": 2120.7,
    "end": 2122.58,
    "text": "So you say like, you should take this hotel."
  },
  {
    "start": 2123.12,
    "end": 2126.28,
    "text": "So we all know it, like, you know, we go to Amazon, we enter something, we get a list."
  },
  {
    "start": 2126.84,
    "end": 2128.74,
    "text": "You know, yeah, somehow Amazon has figured it out."
  },
  {
    "start": 2128.82,
    "end": 2133.1,
    "text": "Maybe it, you know, gets sold the most or pays them the most or whatever it is."
  },
  {
    "start": 2133.3,
    "end": 2134.54,
    "text": "And then the chat."
  },
  {
    "start": 2134.6,
    "end": 2135.3,
    "text": "people now ask, why?"
  },
  {
    "start": 2135.72,
    "end": 2136.36,
    "text": "Why is it this one?"
  },
  {
    "start": 2136.42,
    "end": 2137.36,
    "text": "Why not that one, right?"
  },
  {
    "start": 2137.74,
    "end": 2139.6,
    "text": "I found another flight that is a similar price."
  },
  {
    "start": 2140.28,
    "end": 2141.5,
    "text": "Why do you recommend me this one?"
  },
  {
    "start": 2142.52,
    "end": 2152.1,
    "text": "And the last bit here that goes hand in hand is that as soon as you give users a chat window, also coming from chat GPT and Proplexity, they expect that you know everything."
  },
  {
    "start": 2153.7,
    "end": 2160.98,
    "text": "So this is something to really be Prepared for and and and you know to keep the conversation going to have them come back."
  },
  {
    "start": 2161.6,
    "end": 2164.42,
    "text": "So so coming back to your original question how do we measure it right now?"
  },
  {
    "start": 2164.5,
    "end": 2166.26,
    "text": "It is really at the moment."
  },
  {
    "start": 2166.62,
    "end": 2168.48,
    "text": "We're looking at engagement."
  },
  {
    "start": 2168.72,
    "end": 2170.4,
    "text": "How many people engage with AI features?"
  },
  {
    "start": 2171.48,
    "end": 2177.84,
    "text": "Over time we have to transform that into more and of course our lower level metrics such as you know How many turns in every chat?"
  },
  {
    "start": 2178.42,
    "end": 2179.44,
    "text": "How often do people come back?"
  },
  {
    "start": 2180.06,
    "end": 2182.4,
    "text": "Classic retention metrics, but over time."
  },
  {
    "start": 2182.78,
    "end": 2188.74,
    "text": "I you know, this is It's a hard one to measure, but I think over time we have to then think about valuable AI interactions."
  },
  {
    "start": 2189.16,
    "end": 2196.2,
    "text": "People used AI because it really was the best tool to use, and it really gave them a value over doing it the old way."
  },
  {
    "start": 2198.88,
    "end": 2199.08,
    "text": "Cool."
  },
  {
    "start": 2200.08,
    "end": 2207.28,
    "text": "And what's your take on when to actually jump in?"
  },
  {
    "start": 2208.62,
    "end": 2214.12,
    "text": "I mean, you mentioned that you were early on with custom GPTs."
  },
  {
    "start": 2215.94,
    "end": 2219.14,
    "text": "and that like most likely would also jump in on app."
  },
  {
    "start": 2219.32,
    "end": 2226.18,
    "text": "but are you still in the first mover mode?"
  },
  {
    "start": 2227.18,
    "end": 2228.9,
    "text": "Do you want to be the first?"
  },
  {
    "start": 2229.6,
    "end": 2230.76,
    "text": "Do you want to test the first?"
  },
  {
    "start": 2230.92,
    "end": 2241.86,
    "text": "Or are you maybe a bit reserved because of the potential, let's say, neck breaking competition that arises from all of that?"
  },
  {
    "start": 2245.0,
    "end": 2255.24,
    "text": "So first of all, we actually, for example, we have a large network of affiliate partners that consume our APIs in different ways."
  },
  {
    "start": 2256.04,
    "end": 2262.06,
    "text": "And we actually also have other native travel bots, so to speak, as part of our clients."
  },
  {
    "start": 2262.22,
    "end": 2265.78,
    "text": "So it's not really the neck breaking part of the point of view here."
  },
  {
    "start": 2267.0,
    "end": 2272.18,
    "text": "I think it comes down to available resources and things you can do with these resources."
  },
  {
    "start": 2274.18,
    "end": 2283.94,
    "text": "And then maybe, you know, as good as your crystal board is like everyone else's early indication of what this may become and maybe also the phase in which your company is in."
  },
  {
    "start": 2284.78,
    "end": 2298.64,
    "text": "Like, let me give you the example is that I think, you know, we optimized our kayak app to reasonably work on the Vision Pro, but from out of the gate, it didn't look like something to us that we should really pursue."
  },
  {
    "start": 2300.78,
    "end": 2311.08,
    "text": "And so I think it's a mixture of How we like what we think about the technology ourselves what the buses how we see how it fits to travel."
  },
  {
    "start": 2313.56,
    "end": 2318.14,
    "text": "As we say like it is really like we saw we learned that in Alexa that that also like selling flights."
  },
  {
    "start": 2318.44,
    "end": 2318.92,
    "text": "it was great."
  },
  {
    "start": 2318.98,
    "end": 2323.28,
    "text": "in the end you could install Alexa or you could buy a TV that Alexa installed on the big screen."
  },
  {
    "start": 2323.34,
    "end": 2330.54,
    "text": "it was actually a cool flight selection experience but not every technology necessarily fits to travel and."
  },
  {
    "start": 2331.48,
    "end": 2335.36,
    "text": "We don't have a fancy tech radar or anything like that for this type of problem."
  },
  {
    "start": 2337.24,
    "end": 2348.08,
    "text": "But on a case-by-case basis, we are really evaluating what we are thinking, how many people there are going to look for travel, what's the effort to get there."
  },
  {
    "start": 2348.9,
    "end": 2356.22,
    "text": "And we think that this is the experience that has the right level of experience at the right point."
  },
  {
    "start": 2357.94,
    "end": 2365.6,
    "text": "Sometimes mixed in with, let's be honest, For instance, it was quite onerous to be in the first batch of open AI plugins."
  },
  {
    "start": 2367.04,
    "end": 2373.42,
    "text": "Not to ignore that sometimes the bust around the first group is quite exciting as well."
  },
  {
    "start": 2375.22,
    "end": 2376.04,
    "text": "Yeah, I can imagine that."
  },
  {
    "start": 2377.62,
    "end": 2381.02,
    "text": "It's then a bit tiring if you don't see traction or"
  },
  {
    "start": 2382.26,
    "end": 2382.32,
    "text": "the"
  },
  {
    "start": 2382.38,
    "end": 2384.7,
    "text": "amount of traction that you anticipated."
  },
  {
    "start": 2385.44,
    "end": 2388.1,
    "text": "But on the other hand, you also get used to that, right?"
  },
  {
    "start": 2389.7,
    "end": 2404.18,
    "text": "Unfortunately, the case with many platforms, some kind of are more like the outliers when it comes to like quick user adoption and real traffic and real real real business generating that being generated from that, like like IOS you mentioned, right?"
  },
  {
    "start": 2405.28,
    "end": 2405.62,
    "text": "But but"
  },
  {
    "start": 2406.34,
    "end": 2409.8,
    "text": "reality is that it's more like the opposite that."
  },
  {
    "start": 2410.34,
    "end": 2417.84,
    "text": "and then then you think initially like everyone wants to jump on them and then in most cases it's it's."
  },
  {
    "start": 2418.02,
    "end": 2425.46,
    "text": "it's a bit of a Yeah, not tiring, but it's much slower than you anticipated."
  },
  {
    "start": 2426.96,
    "end": 2428.28,
    "text": "Let me give you another cycle."
  },
  {
    "start": 2428.94,
    "end": 2430.54,
    "text": "that is interesting in this conversation as well."
  },
  {
    "start": 2431.54,
    "end": 2433.14,
    "text": "It doesn't have to be always a new platform."
  },
  {
    "start": 2433.38,
    "end": 2435.48,
    "text": "It must be like three or four years ago."
  },
  {
    "start": 2436.1,
    "end": 2439.02,
    "text": "You can look that up when Apple started to have live activities."
  },
  {
    "start": 2440.58,
    "end": 2447.56,
    "text": "When you initially looked at it, so it's on your lock screen of the phone, the phone is closed and there is a soccer game going and you see the live score without having to open your phone."
  },
  {
    "start": 2448.6,
    "end": 2453.38,
    "text": "And so originally when they released that, it was all about tracking your Uber."
  },
  {
    "start": 2454.0,
    "end": 2460.94,
    "text": "It was about tracking basketball games and it was all meant to be sort of short term, maybe an hour or two, not too long, right?"
  },
  {
    "start": 2460.98,
    "end": 2462.34,
    "text": "You don't want to use that lock screen forever."
  },
  {
    "start": 2463.02,
    "end": 2471.52,
    "text": "So originally when you look at the documentation, so like flight tracking didn't appear to anyone that this would be a great fit because the flight can be eight hours and all that."
  },
  {
    "start": 2472.14,
    "end": 2477.1,
    "text": "Anyways, us and many others went out with a live activity for flight tracking."
  },
  {
    "start": 2478.24,
    "end": 2484.72,
    "text": "We still have it today, but now with the latest iOS release, actually there's a native flight tracking life activity, right?"
  },
  {
    "start": 2484.78,
    "end": 2486.08,
    "text": "So there is from Apple."
  },
  {
    "start": 2486.62,
    "end": 2488.2,
    "text": "So it's a different game, right?"
  },
  {
    "start": 2488.26,
    "end": 2502.44,
    "text": "Like if you're on the right track, sometimes you get how we call it show locked, where you're now competing against the canvas yourself and having different access barriers than what they have, right?"
  },
  {
    "start": 2502.54,
    "end": 2504.98,
    "text": "Because if you think about flight tracking, right?"
  },
  {
    "start": 2505.98,
    "end": 2507.16,
    "text": "At least every flight ticket I"
  },
  {
    "start": 2507.22,
    "end": 2507.48,
    "text": "have,"
  },
  {
    "start": 2507.88,
    "end": 2509.24,
    "text": "I already put it into the Apple wallet."
  },
  {
    "start": 2510.62,
    "end": 2518.3,
    "text": "And so logically, the native version of flight tracking is a little bit easier to access than any app."
  },
  {
    "start": 2518.7,
    "end": 2526.84,
    "text": "Unless, and that's again the opportunity here, you can provide better and deeper data than the native one that people will still appreciate."
  },
  {
    "start": 2527.18,
    "end": 2528.02,
    "text": "and come back to your app."
  },
  {
    "start": 2528.62,
    "end": 2531.62,
    "text": "But it definitely levels the game up a little bit."
  },
  {
    "start": 2532.44,
    "end": 2532.94,
    "text": "Can you imagine?"
  },
  {
    "start": 2532.98,
    "end": 2533.38,
    "text": "Can you imagine?"
  },
  {
    "start": 2535.64,
    "end": 2543.28,
    "text": "Slowly coming to the end, one question that really like these days keeps me busy a lot."
  },
  {
    "start": 2544.54,
    "end": 2550.04,
    "text": "What's your take on AI engineering and the agentic engineering?"
  },
  {
    "start": 2551.6,
    "end": 2552.96,
    "text": "Do you use it like personally?"
  },
  {
    "start": 2554.16,
    "end": 2555.5,
    "text": "How much do you use it in your company?"
  },
  {
    "start": 2556.24,
    "end": 2563.0,
    "text": "What are your thoughts like just in a few sentences around this really changing the game?"
  },
  {
    "start": 2565.62,
    "end": 2569.38,
    "text": "Hi, yeah, let me look at it from two ways."
  },
  {
    "start": 2569.46,
    "end": 2576.0,
    "text": "So first of all, certainly we're using curls like co-pilot and tools like co-pilot and cursor."
  },
  {
    "start": 2578.0,
    "end": 2595.22,
    "text": "I think when you talk to folks who have longer standing, large enterprise code bases, it's not exactly so easy that you just send off ten agents that work on ten different tickets and you just, so we're not in the mode that our engineers are just reviewers of agents."
  },
  {
    "start": 2597.26,
    "end": 2611.76,
    "text": "I think it is the more realistic view is probably to have a really, really powerful autocomplete at your hands and really a different way how you interact with documentation with the first capabilities to take some chores away from you."
  },
  {
    "start": 2613.36,
    "end": 2626.54,
    "text": "What I find On the other hand, what I find super interesting now, that is the tooling side, what I find super interesting with LLMs is that you today no longer need to engage data scientists to build AI features."
  },
  {
    "start": 2627.5,
    "end": 2654.06,
    "text": "So one example here is that when one of our also our first JetGPT based products was something called price check, which works in a way that You're just in a checkout pass of a flight somewhere and you want to make one more check that this is the best price so you take a screenshot Uploaded to the kayak app and we find you the exact same flight and compare it to the price on your screen."
  },
  {
    "start": 2655.52,
    "end": 2658.96,
    "text": "disinvolve the vision and a big part of chat."
  },
  {
    "start": 2659.02,
    "end": 2663.82,
    "text": "GPT to solve problems are like, you know understanding it reading it and it."
  },
  {
    "start": 2665.42,
    "end": 2667.64,
    "text": "This can basically be built without data scientists, right?"
  },
  {
    "start": 2667.74,
    "end": 2673.12,
    "text": "Like you're no longer yourself in in tuning weights You may be in fine-tuning models."
  },
  {
    "start": 2674.34,
    "end": 2688.26,
    "text": "And there is now this huge leverage that great software engineers, call them back-end, full-stack, whatever you like, are really able to now build features based on LLMs, which we've also seen that."
  },
  {
    "start": 2688.58,
    "end": 2696.58,
    "text": "But this is for us, I think, is a major unlock that there is a whole category of problems that now can be done by software engineers."
  },
  {
    "start": 2696.96,
    "end": 2700.18,
    "text": "The ones that haven't been in touch with AI, at first, right?"
  },
  {
    "start": 2700.3,
    "end": 2700.88,
    "text": "Or at all."
  },
  {
    "start": 2702.0,
    "end": 2716.92,
    "text": "That's like for me really amazing that you can, I don't know, write like a software that analyzes conversations in minutes at a level which you never would have dreamed of before, right?"
  },
  {
    "start": 2717.5,
    "end": 2719.18,
    "text": "That's like for me really amazing."
  },
  {
    "start": 2721.28,
    "end": 2726.24,
    "text": "As long as you kind of manage the amount of slop that comes through, right?"
  },
  {
    "start": 2727.32,
    "end": 2733.54,
    "text": "But I think that gets better and better as well in the recent months and years."
  },
  {
    "start": 2733.7,
    "end": 2735.44,
    "text": "So really excited about that one."
  },
  {
    "start": 2736.76,
    "end": 2752.74,
    "text": "Now, coming to my outro question, I have to admit that I found a bug in your AI mode, which, like crazy enough, allows you to time travel physically."
  },
  {
    "start": 2754.72,
    "end": 2762.0,
    "text": "And i don't know about that back if you if you try it out yourself and bought i don't know big amounts of bitcoin a few years back."
  },
  {
    "start": 2762.12,
    "end": 2763.84,
    "text": "but i hope you didn't."
  },
  {
    "start": 2765.22,
    "end": 2768.0,
    "text": "i hope you're not."
  },
  {
    "start": 2768.22,
    "end": 2769.52,
    "text": "You're not rich through that now."
  },
  {
    "start": 2769.78,
    "end": 2773.14,
    "text": "but imagine we would now use that feature."
  },
  {
    "start": 2773.26,
    "end": 2777.14,
    "text": "we type in like slash slash slash time machine."
  },
  {
    "start": 2778.42,
    "end": 2780.14,
    "text": "matthias munich."
  },
  {
    "start": 2780.24,
    "end": 2781.12,
    "text": "two thousand and six."
  },
  {
    "start": 2781.98,
    "end": 2790.4,
    "text": "And we now observe yourself, like in that year, you're wiring train wheel sensors for your computer science project."
  },
  {
    "start": 2791.84,
    "end": 2797.76,
    "text": "And you now have the chance to whisper one or two sentences in your younger self's ears."
  },
  {
    "start": 2798.12,
    "end": 2799.28,
    "text": "What would you whisper?"
  },
  {
    "start": 2802.32,
    "end": 2802.84,
    "text": "Fix the bug."
  },
  {
    "start": 2804.32,
    "end": 2805.94,
    "text": "It's funny that you mentioned Bitcoin."
  },
  {
    "start": 2806.08,
    "end": 2807.42,
    "text": "I'm not an early Bitcoiner."
  },
  {
    "start": 2809.4,
    "end": 2811.8,
    "text": "When I think about this question, it's actually a good way there."
  },
  {
    "start": 2811.88,
    "end": 2824.14,
    "text": "you know like when the bitcoin start ten twelve years right i would joke with friends oh wow these people must really have a warm apartment while they're mining these bitcoins right when the bitcoin was at five euros."
  },
  {
    "start": 2826.82,
    "end": 2830.94,
    "text": "and so i think the you know learning always has to do with."
  },
  {
    "start": 2831.0,
    "end": 2837.0,
    "text": "maybe something that you would do better in hindsight right is is that i?"
  },
  {
    "start": 2837.46,
    "end": 2841.66,
    "text": "and then but sometimes it works out right is like touching to our conversation today."
  },
  {
    "start": 2841.76,
    "end": 2844.96,
    "text": "is that um is is it somewhere in between?"
  },
  {
    "start": 2846.14,
    "end": 2848.68,
    "text": "If you can you have to take certain risks, right?"
  },
  {
    "start": 2848.74,
    "end": 2858.18,
    "text": "Like if you come to Bitcoin when it's at a hundred fifty thousand I'm not saying I'm not giving financial advice here, but it sounds like worse than when you get to it at five at five dollars, right?"
  },
  {
    "start": 2858.3,
    "end": 2867.82,
    "text": "And and the same I feel about about LLMs I felt the same about wireless sensor networks and and I felt in some places I I could have maybe stayed longer at it."
  },
  {
    "start": 2868.92,
    "end": 2876.7,
    "text": "So like, you know, it's this moment when you realize that that oh wow this idea that we had five years ago Suddenly, somebody has a business, right?"
  },
  {
    "start": 2876.78,
    "end": 2879.2,
    "text": "And if we would just have stayed at it, right?"
  },
  {
    "start": 2879.98,
    "end": 2888.32,
    "text": "And back in the day, we rejected it because we felt like, you know, it wasn't... We didn't get the distribution or the feedback quick enough, right?"
  },
  {
    "start": 2888.38,
    "end": 2892.96,
    "text": "It was the frustration you mentioned, like, you have all these great things there, and maybe they don't come."
  },
  {
    "start": 2894.3,
    "end": 2906.58,
    "text": "So, you know, to steal a quote from the Nvidia CTO that I recently heard at a conference, this is like... If you build it, they may come."
  },
  {
    "start": 2907.02,
    "end": 2908.54,
    "text": "If you don't build it, they will not come."
  },
  {
    "start": 2910.5,
    "end": 2912.4,
    "text": "So I think this is what I would whisper to myself."
  },
  {
    "start": 2915.08,
    "end": 2915.64,
    "text": "Okay, yeah."
  },
  {
    "start": 2916.58,
    "end": 2919.24,
    "text": "Really cool conversation, Matthias."
  },
  {
    "start": 2919.58,
    "end": 2922.22,
    "text": "Really, really excited about that."
  },
  {
    "start": 2922.76,
    "end": 2923.26,
    "text": "Thanks a lot."
  },
  {
    "start": 2924.08,
    "end": 2928.76,
    "text": "And yeah, I think, I mean, the hard thing is really to choose your balance, right?"
  },
  {
    "start": 2929.48,
    "end": 2932.74,
    "text": "Also, like when I look at what you just advised yourself."
  },
  {
    "start": 2934.16,
    "end": 2952.3,
    "text": "i mean i remember a time when i read about i don't know i was in a hotel and read about like blockchain and what like the amount of things it unlocks and how it makes the world how it changes the world in total and i thought like i have to get into that back in the days."
  },
  {
    "start": 2952.8,
    "end": 2953.14,
    "text": "I didn't."
  },
  {
    "start": 2953.96,
    "end": 2958.26,
    "text": "and i have to say like right now i'm happy that i didn't waste my time there."
  },
  {
    "start": 2958.9,
    "end": 2961.66,
    "text": "so it's it's it's really hard to like an insight."
  },
  {
    "start": 2961.74,
    "end": 2965.88,
    "text": "it seems easy but it's it's really hard to choose your battles right in reality."
  },
  {
    "start": 2967.54,
    "end": 2978.62,
    "text": "Yeah yeah we you know we're all hoping that we're all thinking that that ai is not the bubble but that it really unlocks a lot of great technology for us."
  },
  {
    "start": 2978.96,
    "end": 2979.22,
    "text": "but."
  },
  {
    "start": 2980.7,
    "end": 2992.98,
    "text": "Yeah, you know, I I also just here I I remember I think at the time actually when we presented Alexa the hotel booking and maybe or maybe the year after it doesn't matter It was this conference in San Francisco."
  },
  {
    "start": 2993.08,
    "end": 3000.78,
    "text": "And actually that was I think it was Sam Altman and Greg Brockman were coming there Talking but back then, you know, they were a GPT one or two."
  },
  {
    "start": 3001.0,
    "end": 3005.08,
    "text": "So I was like, wow, these these people they they talk weird stuff."
  },
  {
    "start": 3005.66,
    "end": 3007.86,
    "text": "They're talking about software that is writing itself."
  },
  {
    "start": 3008.44,
    "end": 3009.88,
    "text": "This all makes no sense, right?"
  },
  {
    "start": 3009.96,
    "end": 3014.72,
    "text": "And I will closely follow other people who kind of like tried a little bit with GPT-II and stuff."
  },
  {
    "start": 3015.24,
    "end": 3016.22,
    "text": "But it was all there, yeah, right?"
  },
  {
    "start": 3016.26,
    "end": 3017.64,
    "text": "There's this kind of like weird eye-lap."
  },
  {
    "start": 3018.28,
    "end": 3024.1,
    "text": "And now, let's call it ten years later, that's where we are, right?"
  },
  {
    "start": 3024.22,
    "end": 3031.48,
    "text": "So these guys were hanging in and seemingly had a really big vision ahead of all of us."
  },
  {
    "start": 3032.62,
    "end": 3033.56,
    "text": "Yeah."
  },
  {
    "start": 3033.96,
    "end": 3034.86,
    "text": "I really admire that."
  },
  {
    "start": 3035.82,
    "end": 3038.62,
    "text": "So, Matthias, thanks a lot for the conversation."
  },
  {
    "start": 3039.26,
    "end": 3041.42,
    "text": "Hope to see you soon in real life somewhere."
  },
  {
    "start": 3045.32,
    "end": 3048.9,
    "text": "Have a good evening, have a good day, and I hope to see you soon."
  },
  {
    "start": 3049.94,
    "end": 3050.36,
    "text": "Thanks, you too."
  },
  {
    "start": 3050.94,
    "end": 3051.08,
    "text": "Bye."
  },
  {
    "start": 3053.04,
    "end": 3055.24,
    "text": "Thank you for listening to the Arthur List podcast."
  },
  {
    "start": 3055.46,
    "end": 3057.92,
    "text": "If you liked this episode, share it with friends."
  },
  {
    "start": 3058.14,
    "end": 3059.04,
    "text": "I'm sure they love it too."
  },
  {
    "start": 3059.36,
    "end": 3066.08,
    "text": "Make sure to subscribe so you can hear deep insights into technical leadership and technology trends as they become available."
  },
  {
    "start": 3066.5,
    "end": 3073.66,
    "text": "Also, please tell us if there is a topic you would like to hear more about or a technical leader whose brain you would like us to pick."
  },
  {
    "start": 3074.06,
    "end": 3080.96,
    "text": "Alphalist is all about helping CTOs getting access to the insights they need to make the best decisions for their company."
  },
  {
    "start": 3081.32,
    "end": 3085.12,
    "text": "Please send us suggestions to cto at alphalist.com."
  },
  {
    "start": 3085.68,
    "end": 3087.44,
    "text": "Send me a message on LinkedIn or Twitter."
  },
  {
    "start": 3087.82,
    "end": 3092.22,
    "text": "After all, the more knowledge we bring to CTOs, the more growth we see in tech."
  },
  {
    "start": 3092.52,
    "end": 3096.54,
    "text": "Or, as we say on Alphalist, accumulated knowledge to accelerate growth."
  },
  {
    "start": 3097.1,
    "end": 3098.74,
    "text": "See you in the next episode."
  }
]