Letstrip.ai: Fully Personalized Travel Assistant (one simple AI Prompt)

Let's Let's Trip generates personalized trip itineraries by pulling data from social media platforms like TikTok to showcase current trends and hotspots. Users simply describe where and when they want to travel, and Let's Trip provides venue ideas, events, restaurant recommendations, playlists, and more to build a full trip plan. 

Alon explains how he got the idea to combine artificial intelligence with social media data to create an effortless way to plan vacations and weekend getaways. He also discusses the development of Let's Trip, current features, and future capabilities in the works like budgeting and exporting trips to Google Maps. With powerful AI continuing to evolve rapidly, Alon sees a future where AI assistants can create full travel videos and share them on social media automatically. He shares his vision for AI integration across various industries as the technology advances and humans continue finding new applications for it.


LET'S TRIP AI

https://letstrip.ai/


FRY-AI.COM

https://www.fry-ai.com/subscribe
https://www.youtube.com/@TheFryAI
https://twitter.com/lazukars
https://twitter.com/

Creators & Guests

Host

Ryan Lazuka

The lighthearted Artificial intelligence Journalist. Building the easiest to read AI Email Newsletter Daily Twitter Threads about AI

What is Behind The Bots?

Join us as we delve into the fascinating world of Artificial Intelligence (AI) by interviewing the brightest minds and exploring cutting-edge projects. From innovative ideas to groundbreaking individuals, we're here to uncover the latest developments and thought-provoking discussions in the AI space.

Alon Zilberman: Yeah, so my name is Alon. I started my IT journey very, very long time ago in Ukraine, even before moving to Israel. When I moved to Israel, I joined the army. I used my IT background there. It was IT related army job and after three years in army having the development experience, I joined an early stage startup in Israel as a full-stack developer with more mobile background and worked there for a few years.

It was acquired by a bigger US company and a year ago with this company, I moved to US and in US as a CTO and in US, of course, it's a new country. I started to travel a lot and it also was the time when charge gpt was just announced and different AI tools started to pop up here and there and I thought how the AI can be used to plan the journeys. So one thing was AI, the first thing I tried to do is just to ask charge gpt to build a plan for weekend trip or for a weekend in Los Angeles and this kind of stuff and it worked nice, but it was really missing the visual component. So another part of it that in social media, when you just scroll Instagram or TikTok, you often see a lot of really cool travel videos and I thought that it would be a great idea to connect two of this company together. The power of AI and social media to give AI the lack of visual part and this is how I started to work on Let's Trip a half year ago.

Hunter Kallay: Very cool. So yeah, I stumbled upon this this tool that you have, letstrip.ai and it is just very simple to use, very cool plans and anywhere you want to go in the world for really any occasion it seems like. So very cool. Can you just give us an overview of what the tool is and how it works?

Alon Zilberman: Yeah, basically you just describe where you want to travel. It can be like a two-day trip with my wife to Seattle or one week in California with my friends and it builds you a basic trip plan with places to see, with events nearby, with some restaurants and hotel recommendations and show it everything on the map.

So it's more visual and then you can sign in and make this trip plan more personal. You can edit dates for your hotel space or you can add more places. You can add some notes for example, I was just having a two-day trip in Los Angeles and I built a basic trip plan with Let's Trip and then I went to read it to find what locals are recommended to see. So I took some recommendations from read it and just edited to my trip plan which was built with Let's Trip. And on top of it, I think the big benefit of Let's Trip is that it has a social media component for each place. It gives you a recent TikTok video so you can very quickly see if you want to visit this place or not. You can just click on the video, see if you short videos and it doesn't really, like it's too much tracking. I want something different or this place looks really cool. It works to hours tracking. I will do it. And yeah, that's basically the concept.

Ryan Lazuka: Cool. The thing that struck me when I first tested it out is one of the really cool things it does is, it does two things that stood out, is that it gives you a playlist of songs so that like sort of compliment where you're going. So like say if you're going to the Caribbean or Caribbean or wherever you say it and you are going to the Bahamas or something, Let's Trip will put out a bunch of different songs that you can play to sort of get you hyped up for your trip.

I know, but I went on cruise as before and I like to download a bunch of reggae songs to listen to. So this sort of does that for you automatically. And then it also does something that's really neat is how it will list out events in the area. So if there's a concert going on or maybe a sports event, it will put that in the itinerary from the days that you want to go on your trip, which is really neat.

Alon Zilberman: Yeah, the playlist idea I got from the trip I had with my wife in West Virginia. As you know, there are a lot of cool county songs about West Virginia. So and my wife made the playlist manually before going to the trip. Like she added all the different songs in her Spotify playlist. And then I thought it will be really cool. Like if you can just get this playlist automatically.

Ryan Lazuka: Yeah, I mean, it had to have John Denver's Take Me Home Country Roads.

Alon Zilberman: That's a great karaoke song.

Hunter Kallay: Yeah. So just tell me a little bit about, you know, we'll go into a little bit more of the details, the cool stuff that it does. But before that, tell me what inspired you to do this project? Was it just as you talked a little bit about Chad T.P .T. But why travel? You know, was it just because you were traveling a lot or?

Alon Zilberman: Yeah, I love traveling. I visited a lot of different countries. And when I moved to U.S., I also, U.S. is very different from Israel. Here you can travel a lot by car.

In Israel, it's a small country. So usually you fly to Europe. Here you can just, I'm near the Seattle, so I can just drive to Portland for a weekend or to Vancouver, B.C.

for a weekend. And I made a lot of trips like this. And each time you kind of need to spend time to decide what you want to see. And you really don't want to invest too many effort because it's just a weekend trip. It's not some big trip which you plan for a year.

But from one side, you don't want to invest too much time. From the other side, you do want to have a cool trip. And this is how I got the Let's Trip idea. The initial idea was to generate the small weekend trips really, really fast.

Ryan Lazuka: One cool thing I see a lot of people could use this for is like, that I would use it for is if you're sitting on the couch and you're watching a travel channel and you see some cool like destination come up, like, man, that'd be cool to go there. And you just type it in, let's trip and see what it comes up with. It's almost like a kind of a game to play, to see what kind of cool exotic trip it's going to find for you. And then it will give you more information about where you want to go as well. So you might learn some new things about whatever Let's Trip puts out for you.

Alon Zilberman: Yeah, it's also a nice use case.

Hunter Kallay: Yeah, I wanted to talk about how you're getting some of these recommendations because when I looked up Let's Trip, I went to some of the places that I've been to. I live in Knoxville, Tennessee. So I wanted to see, well, what does it say about, you know, Knoxville and what to do there?

It was really good. Some of the restaurants that were coming up for like some of my favorite places, you know, really interesting, like I'm aware of what's going on. And I'm like, how's it getting these recommendations? Is it off of like Google reviews or something? Where's where's it pulling this data from?

Alon Zilberman: It's a combination of chat GPT and TikTok together with a little bit of Instagram, but mostly TikTok. So basically, it collects data about popular places on TikTok, and then uses a chat GPT to be kind of a moderator or be the layer on top of this data and combining two of these things together, allow you to get really nice lists.

Ryan Lazuka: Say if you're typing in, I want to go to San Diego for two days. It pulls in, how does it, it works by pulling in data from TikTok. So it sort of pulls in like the hottest videos about San Diego at the time and then reads all the, maybe reads the description of the TikTok video in the comments or is that how it works in general?

Alon Zilberman: Yeah, yeah, it uses the descriptions, it uses the hashtags and also the amount of likes and shares the video gets. Basically on the background, it makes a search on TikTok. It receives the list of the places, the potential places from chat GPT, then does searches on TikTok to see which ones from these places have any social media activity and then returns back to chat GPT. This place has X social media activity, this Y and chat GPT decides which ones will go to the final list as a user see.

Ryan Lazuka: Awesome. That's really, it's really cool because TikTok's like the avant-garde tool to get whatever the hottest thing out there is. So probably why you're finding so great recommendations.

Hunter Kallay: It seems to stay very current too. Like it, will it adjust because of the trends on TikTok then? Like if something opens up or gains popularity, will it then pull that in?

Alon Zilberman: Yeah, it should right now. It does have some cash period. So if it's something super fresh, for example, something like a new restaurant which was opened a month ago, probably it will not catch it. But if it two months ago, it should already catch it because yeah, it's pretty difficult to do it in real time. But if you have a small delay, like a two weeks delay or one month delay, especially for popular destinations, it's pretty doable to get the fresh data.

Hunter Kallay: So it looks like once you put in your trip that you want to have, the prompt can be simple or complex. It seems like it'll also give you like right on Google Maps exactly where everything is, how far everything is from everything. You can just click it, go right to it, share the trip, save the trips. Some very cool features.

One of the cool features that I saw was it'll, if you say I'm taking a two day trip to Orlando, Florida, it'll actually give you the summary of what you should do for day one, do this, day two, do this. So you don't even have to think about it. How does that happen? How does it pull in the recommendations? Is that ChatGPT putting it all together?

Alon Zilberman: Yeah, it's basically a ChatGPT. I also experiment a lot with other models with latest Google One, with Gemini and some open sourced ones. So it's basically the model use the trip structure to build this trip overview with a day's breakdown. And you can also ask if you really want a detailed breakdown, you can include it into your request and say, give me a detailed, my day breakdown and it will do it.

Hunter Kallay: So I want to ask you kind of retroactively what you're talking about before. How did you land on TikTok? That's not really the first place I would think is like, this is where we need to pull our trip information from. Did you experiment with some other stuff and figure this doesn't work or how did that idea come in?

Alon Zilberman: Yeah, so initially I really wanted to use the Instagram because this is what I use in myself daily and I found a lot of nice trip ideas from Reels in Instagram. But when I tried to use the Instagram, there were too many technical challenges. So I decided to try this TikTok and figure out that if you personalize your TikTok feed, if you let it know what you like and what you don't, you can get really, really nice personalized feed. And also TikTok is more friendly to pull the data from. So that's why I went to the TikTok. But I still want to add the Instagram component later.

Hunter Kallay: Very cool. So kind of bring in all different kinds of social media. So one thing I did notice about it is it does pull in like pricing for different activities around the area. Is there any way that you can put in some sort of budget into your itinerary or have you ever thought about that kind of feature?

Alon Zilberman: Yeah, it's a cool idea. I had it and I even have the UI ready for it already, like the mock-up. So I'm now experimenting what will be the best way to do it. And because the budget has a lot of variables, are you traveling by car or are you flying?

In what hotel will you stop? So I think what we will have is the basic budget and then allow you to make simple choices like your transportation and this kind of stuff. And it will adjust. Yeah, but I think it will be cool. I experiment a lot with the budgeting.

Hunter Kallay: Very cool. Do you have any other features that you're working on currently to improve? Yes.

Alon Zilberman: So the export to Google Maps is coming. I got a lot of feedback that people want to import the trip plan from Let'sTrips and create Google My Map with all the points.

Because they use it for navigation and they just used to use the Google Maps app for navigation. So this will be the soonest feature which is coming. I also want to do some UI tweaks based on user feedback. What I'm basically doing right now, I measure the user's feedback with the hotjar to see where they are struggling with the UI.

And I want to improve these parts. And then it will give users the freedom to experiment with different AI models for the trip plans because they behave a bit differently. And probably it will also, for different people, different models will give trips which are more personalized for them. So this is also in my to-do list. And this will be probably the first paid feature I've led.

Ryan Lazuka: It sounds like you're developing this or do you have a team of people? Yeah. It's just you right now.

Alon Zilberman: Right now it's just me. I do have the designer who does all the mock-ups and UI parts. And I have a person who helps me with SEO component a little bit. Okay.

Ryan Lazuka: Awesome. Well, the website looks great. I've done web development for the last 15 years. And you have a very nice looking website. It's a nice work. Do you have an app right now? And if not, do you plan on having one? I would love to have an app.

Alon Zilberman: I don't know on which stage because I don't know yet what will be my financial plan. Will it be raising money or just adding paid features and see if I will have enough money from the paid features to develop it. But I do want to have an app. And I also have an idea to go to the really narrow niche and have a Mac app, like desktop Mac app.

Ryan Lazuka: Well, I didn't even think about that. So you would do that just because you could get more market share because no one else is doing that on the Mac. That would be the reason. Right.

Okay. I didn't check fully, but can someone book a flight on your site? And if they do, do you get a small percentage commission fee off of that? Or is it just a subscription membership that you're trying to monetize eventually for the site?

Alon Zilberman: Right now it's totally free. I don't have any monetization, neither via affiliate links or subscriptions because I launched it a few months ago and what I wanted to do initially is just to measure the interest. Will people use it? Will they sign in?

Will they play with strips? And now I do see, like I have around 100 users per week from organic traffic and I do have 25 registered users. So there is some interest here.

And I will start. I don't really want to go to the ads direction because I hate how a lot of travel sites have the screens full of ads. So I would try to add paid features first, like pay $1 to import your trip to Google Maps or pay $5 to get access to another model for the trip generation. And then, yeah, probably I will add some affiliate links. For example, if you click on the hotel and want to book it, it can direct you to Expedia or to booking.com and I don't get some small fare. So yeah, this is the plan actually.

Ryan Lazuka: Yeah, but right now the site is more focused on just giving you an itinerary things to do because you're right. Can you book a flight or a hotel on the site right now?

Alon Zilberman: No, no, it will give you the links to the hotels. But right now it doesn't have a booking component.

Hunter Kallay: It seems like if you click on a link to a hotel, it just takes you to the hotel's website or something like that. Yeah, yeah. So it's very convenient, but on the business side, eventually you might work in.

Alon Zilberman: Right, probably on some point I will add some man in the middle, like it will be a link to booking.com or link to Expedia page of the hotel. So I will be able to get some booking share.

Ryan Lazuka: Yeah, it's always hard to build. You could have a great idea, but if you don't mark it at the right way, it sounds like you're doing a lot with SEO, which is good because it doesn't that doesn't cost a ton of money compared to ads. But it's really hard to see if something's going to gain traction without spending a ton of money in ads or, you know, right, like Facebook ads or Twitter ads or stuff like that. So that's is that your primary target? That's your primary strategy right now is just to go with the SEO route to push it?

Alon Zilberman: I want to, I started with SEO because I have a lot of background, like the company where I work full time. We are all about local businesses SEO. So I have a lot of backgrounds there. I think my next step will be more social media stuff. I am looking for a partner or somebody who can help me with social media because I think we can try a lot of cool stuff with AI generated content or mix of human and AI generated content related to traveling in social media. And then yeah, the third step probably will be trying some ads, but I will add ads only when they have, when they will have some clear monetization title. So I will be able to measure how much I spent to the ads and how much I earned from from the monetization channel.

Hunter Kallay: Very cool. So one of the things that we like to ask a lot of the people that come on, is. We just like to ask your opinion, you know, outside the project, you know, AI is a crazy thing that's taken our world by storm. There's no doubt about it. I mean, everybody's got their opinion on it now. Good, bad, whatever. What's your general impression of AI in a general way, and how do you see it influencing society?

Alon Zilberman: Yeah, I really loved it. It's the first time I started to play with GPT, GPT3, GPT4, and I think it will, it already changes the world a lot, and I think it will change it even more. Probably the big part of it will be that it will replace a lot of human jobs.

But it's not the first time in human history when something new replace, replace humans in certain niches. So I think right now it's mostly about creating the content, but I think in a year, for example, we will start seeing some effect from AI on the medications, health related stuff, maybe education more, more like, right now I think it's still very technical related. So it's great, a big buzz in technical community for developers, for social media creators, and for IT related jobs. But with time, I think it will slowly go to all the different other different niches. And it also, it looks like that Open AI and Google are moving very, very fast. Like Open AI just announced the Sora video model. So it looks like that maybe in a year or two, we will have even more powerful AI models.

Ryan Lazuka: That's Sora. That was unbelievable. It's like, you know, runway ML is doing their text to video right now, and then you see what they do if people are familiar with them. And then Sora came out yesterday from Open AI, the text to video, and they do 60 seconds. I mean, I think runway did maybe 10 seconds or 12 seconds at the most, even if that's even that might even been long, I don't belong.

But they do 60 seconds. Now that Sora, it hasn't been released to the public yet, but there's a bunch of different examples out there, like a lot of you've, I'm sure you've seen, and they're incredible. So it's like incredible for the same token, same token. You know, I'm a computer programmer. It's like, well, if that can do that for video, I mean, developers could be a thing of the past in the next five to 10 years, you know. So it's just a real kind of very exciting moment and scary one at the same time, because it's going to start going into all these different niches, like you said, artificial intelligence. You know, there could be a robot to do a plumber's job one day or a robot to take care of your kids. I don't know how far it's going to go, but it seems like it's all evolving very quickly.

Alon Zilberman: Yeah, yeah, I agree. But I think that people will also evolve with it. Like, even if some jobs will be replaced by AI, it will definitely create more jobs which now can't imagine, but which will be very real in a few years.

Ryan Lazuka: Do you think that's going to be like those types of jobs? I'm trying to envision what the world's going to look like, and that's probably impossible to do, but is it going to be those jobs where, like, say, for example, you work at a corporate job in a big city in the United States, and you go to work every day, and maybe you work remote a few days a week, and it's mostly data entry or, you know, stuff that AI can do. So maybe your job gets replaced, right? You're a business analyst, and AI replaces your job. But then what would your job be after that? Is it more like you would just maybe never go into the office, and you're a manager of AI instead of doing the job, instead of doing the business analyst job, AI now does it, and you're more like a prompt manager for all these different AIs? I mean, can you see it being like that, or do you have your own take on things?

Alon Zilberman: Yeah, I think that when we will have AI spread out to different industries, the human touch will be much more valuable, because everything will be done with AI. You will be ready to pay more if you're a waiter, you will be a real human, or if your teacher will be a real human. So I think there will be a lot of room for, like, it will go to a little opposite direction. Right now we have a lot of Zoom classes, for example. I think that real life classes will be more valuable. And so I think there will be more room for jobs where human interact, are interacting with human, because the more AI we have, the more human interactions we would like to get.

Ryan Lazuka: Your vision might be the opposite of what a lot of people think. We're going to be, instead of getting more technical, maybe it's going to help us more focus on our human relationships more.

Alon Zilberman: Right, it's similar to, I don't know how to say it in English, but you know this effect when you're looking at a very realistic robot, which is so realistic that it looks scary. So I think this, we will arrive to the stage where there will be so many AI that you will be ready to pay to interact with humans.

Ryan Lazuka: Got it, because that makes sense. Yeah, it's going to be fun to see how all this plays out, how quickly it happens. I mean, you see these robots from Tesla, and there's a few other companies that are doing robots, but I mean, those days where you have a robot that comes to do your chores for you, it could be right around the corner.

Hunter Kallay: Yeah, I wanted to ask you on, because you have some background in the market in Israel as well, how do you see the US versus other countries, particularly Israel in terms of AI development? A lot of people think US is way ahead of the curve. Other people say, well, US is behind the curve, at least in terms of regulation and things. What's your take on that?

Alon Zilberman: Yeah, I think US is definitely ahead, and regulation is one of the reasons Europe, well, Israel is less than Europe. It's Israel is somewhere between Europe and US. Like Europe love to regulate. US usually much slower on the regulations, and Israel somewhere in the middle. So I think the thing is that the regulations doesn't really make sense in global world, because you can regulate it in US or in Europe, but in Russia or China, you can't force them not to develop it. So the more you regulate it, the more your competitors from another countries where it's not regulated, will run faster. So I think US, by my opinion, have pretty good balance between having the basic regulations or working on basic regulations, but give more freedom and not try to regulate it too much and stop companies from innovations.

Ryan Lazuka: Yeah, I think you hit the nail on the head. As long as one country in the world doesn't abide by regulations or just says we're not going to regulate AI, then what's the point of regulation?

Because all those companies are going to move to that island or country or wherever it is. So maybe the best regulation is to have the AI regulate itself, if that's even possible.

Alon Zilberman: Yeah, you can compare it with a nuclear weapon. Even this very sensitive topic is super hard to regulate. You still can't force countries to have it. Even in such super complicated topic, AI, it's made by developers.

You don't have too many physical components included. So I don't believe you can stop it or regulate it. You can regulate it in Italy or in France, but developers or companies will move to US or if it will be regulated to US, they will just move to China. So you can't stop the progress. That's my vision.

Hunter Kallay: It seems like whoever has the AI has the power. You think about a standoff. I'm thinking of the wild West standoff when everybody's got their guns up and I'm like, okay, well, if I drop my gun, then I'm the most vulnerable.

If everybody drops their guns, maybe that's better for everybody, but as soon as one person picks it up, now they have all the power over everybody. So I think that's kind of an analogy for AI in this sense of, yeah, if somebody's, kind of like Ryan said, if somebody's got AI going full steam ahead, then everybody else is behind the curve. So Ryan and I tend to have a more positive view towards AI. We try to see it as something that opens up a lot of opportunities for people who can help them elevate their level of performance and whatever field they're in. And we're hopeful for a great future of AI.

Alon Zilberman: Yes, same for me. I also think it will do much more good things than bad things.

Ryan Lazuka: Yeah, it's gonna, there's, I just, I haven't dug into it, I just found out there's people out there like respected people on the flip side of all this AI hype, like they have computer backgrounds that say AI is just completely overhyped, all this compute power. Yeah, it does cool things, but it's really gonna be a big waste of time.

And it's all overblown. So I'd like to dig in more to that to see why those people say that. But it's gonna be, is there anything that you, Alain, see out of AI that maybe most people don't or you see something happen that most people don't think is gonna happen with AI over the next, you know, five years?

Alon Zilberman: Yeah, on your topic, what is a little bit scary for me, or I understand from which points these people you mentioned are coming from. I also, I have a I don't really technically fully understand how, for example, charge APK works. So, and yeah, it's, it's a little bit of scary when you don't almost any technology, if you have a lot of background and experience, you can dig in and understand. But when it's, you don't fully understand that you afraid it a little bit. So maybe this is how some people are feeling about it. And another point, what I read a lot about that it can go under control, like it can become, it's, we can use an analogy with a Google search algorithm, like, nobody really understand, even inside the Google itself, how it works.

So, but this is just a search algorithm, if it's an AI model with access to the internet, and nobody really understand how it works, I understand why it makes people scary. My vision for the next five years, yeah, I think it will be more kind of evolution, like we just saw with Google or OpenAI, like more powerful models, more context, ability for models to work with different tools, like right now it's images and videos and voice. Maybe in two years, they will be able to make not a minute video, but a full movie, or maybe they will be able to not just make videos, but upload them to the TikTok and Instagram themselves. So I see an evolution, I don't think we will see in the next five years, like something magical, which will just replace the human, but we definitely will see a lot of small evolutions. And then when you will compare AI today versus in five years, you will see that there is a huge gap, like between iPhone and the first Nokia, this kind of stuff.

Ryan Lazuka: That's a good analogy. One of the things that's sort of playing out right before our eyes right now is, you know, three major companies in the United States are controlling, potentially controlling AI. So you got Nvidia, you know, Meta, he's investing a ton in AI and OpenAI and Microsoft. So we got these three big players and they take up, like for the NASDAQ, they're the reason why the NASDAQ has hit all time highs lately, the only reason. Like a lot of other companies in the NASDAQ have actually lost value, but since these AI stocks have done so well, they pulled everybody up by the bootstraps and caused the also is that, you know, if we get all these companies that are just a few handful of companies that control all of AI, well, it can almost be a national security risk for the United States, because say God forbid, some country bombs Nvidia plants, you know, in Taiwan or something like that, because those companies control so much of the stock market, if Nvidia goes down and loses 50% of its value overnight because some terrorist attack or something, that would cause a huge global crisis because the United States economy would just collapse overnight. So what do you, how do you feel about like all these big players, these three big players controlling AI at the moment, and do you think it's going to get worse? Like is there going to be, you think there's going to be like one big winner at the end of this, or it's going to be shared between three or four companies like it is right now?

Alon Zilberman: I think there will be very soon nice open source competitors in this space, because right now it's super hard, lots of developers are working on it. There are some pretty promising open source AI models. And I think that open source is something that can make it less centralized, like with operation systems, for example. Yeah, you have big players also here, like you have Microsoft with Windows, and you have Apple and Android, but you have open source Linux, which kind of takes some power from them. And I have the same, I think it will be the similar future with AI, but it doesn't solve the problem you mentioned with the Taiwan, where all the cheap manufacturers are there. So if there will be some issue, and yeah, I don't really see any movement of these companies to US or any distribution of this manufacturer process, and this, yeah, this is a little bit scary. So I think it will be, the problem with centralization is more about hardware parts and software part, because in software world, I do think we will see some nice open source solutions soon, but with hardware, yeah, it's problematic. Yeah, we'll see.

Ryan Lazuka: It's like you don't need to watch TV anymore, you just watch the news, you know, changes every day.

Alon Zilberman: Yeah, Mistral is really cool. We always talk about that on podcast. Are there any other open source LLAMs that you know of? LLAM from the one Facebook outsourced, it's pretty cool. Yeah, they are both are not as capable as ChagPP for now, but open source always have a big benefit.

You have a huge community which works on it. So even if it won't be the same as good as ChagPP, maybe it will be 80% as good or 90%, it will be competitive.

Ryan Lazuka: It's gonna be crazy when ChagPP 5 comes out, because if they release Sora, which is like night and day better than their competitors, like Runway. So when they release ChagPP 5, is it gonna be 20, 30, 100 times better than ChagPP 4? Who knows, but it's gonna be awesome when it releases, you know?

Alon Zilberman: Right, it could be that they will just wipe all the competitors.

Ryan Lazuka: Maybe those open source companies just perpetually play catch up, you know, they're perpetually at 80% of what the competition is, but I hope not. I hope open source wins out.

Hunter Kallay: Yeah, I mean, it's insane. You didn't see any video creation from OpenAI for months, and now all of a sudden here comes Sora. And if you go on OpenAI, I was on OpenAI's Twitter today this morning, watching some of these videos. They have some demos out there right now, and it's like the Snowfall one, they've got like a movie trailer, they've got this lady walking through New York City, and it's like the most video I've seen is like Ryan said, kind of grainy, you can definitely tell it's AI. It's good, you know, but you can definitely tell it's AI for like 10 seconds, you know, or like a picture that moves a little bit, you know, and now we got like this whole, they're like making a whole movie trailer now. It's like, it's insane.

Alon Zilberman: Yeah, I got the same impression. I played with some other video making models before, I said maybe I can somehow automate the social media creation process, like create travel videos and upload them to the Instagram, like so each trip plan will have kind of a trailer video, and all, all solution I tried was, they weren't really there, it was bad quality video. But yeah, with Sora, it will probably change.

Ryan Lazuka: Definitely, you got to check out, we just interviewed osa.ai, O-S-S-A .ai, and they make short form videos, they're AI generated, and all you do is like enter a two to three sentence script, and they produce a short video for you, and it does an incredible job. I mean, it's all AI generated pictures, but a lot of the times you can't tell.

And even if you can't tell the the shorts so good that you're like, wow, that was really cool. So yeah, if you want to reach out to them, we can hook you up with them, but they're really cool, cool guys that started it, actually Cole, the guy that started it, he has like 7 million social media followers and 7 billion views on TikTok and Instagram. So I've used their tool, and it probably be a good fit for you guys, because the the shorts that I created on it have done pretty well on my channel, and I was like very surprised.

Hunter Kallay: Yeah, I will give it a chance. Yeah, it's an interesting project because Cole, like Ryan said, has, he's got 5 million TikTok followers, and some of the, he's been in this industry for a while, and some of the secrets that he's put into his videos, he's actually embedded that process into the model, like just the way the photos move, the way the text jumps at you. So you don't even have to, you don't even have to know how to make videos, you know, it's kind of embedded into the model, which is pretty cool. And that gives it a little unique spin over some other video creations.

Ryan Lazuka: Yeah, we're not, we're not like paid or affiliated with them or anything like that, it just, it was a cool tool that works well. Yeah, I will give it a try.

Hunter Kallay: One of the things we like to do is we just like to explore some of the cool stuff in the space, and that's how we came across Let's Trip, is we're just like, you know, we like, we like to explore cool tools and projects, and just like, I stumbled across, you know, I looked through the tools for our newsletter, because we feature tools in our newsletter every day, and Let's Trip came up, and I was like, this is one of the, this is one of the coolest projects I've seen in a while.

Thank you. Very, very practical, you know, a lot of times, like, you got to go through like a painstaking sign up process to even get the thing running. You can just type it right in and go right ahead. And I was like, whoa, like, I couldn't believe it was free, actually, I was like, I can't believe this is actually a free tool.

Alon Zilberman: I really wanted to gather the user feedback. So, I tried, that's why I, I tried to make it as simple as possible to start. Yeah, and I got some nice traffic from your newsletter. That's why when you emailed me, I recognized, I recognized the name, because I saw it in the traffic reports.

Ryan Lazuka: Cool, cool, glad to help you out. It's interesting to see, like, we don't study the analytics that much. I mean, we do a little bit, but we're just trying to grow the newsletter as fast as possible right now. So, it's good to hear that it's actually useful, you know, and helping get some traffic for who's ever listed on it. So, cool. It's, it's Letstrip.ai, and anything else you want to promote, feel free to do it now.

Alon Zilberman: Yeah, I would just mention that if you have any ideas, if something didn't work or something is missing, I will really appreciate that you just submit a contact form on Letstrip.ai. I read them all, and I'm getting some nice feedback from people who submit them. So, it will be very helpful.

Ryan Lazuka: That's, that's the fun part, to be critical. My wife loves being critical of me, so I'll tell her to go on there and she'll give you some good feedback.

Hunter Kallay: Yeah, and then be sure to subscribe to Ryan and I's newsletter too. It's fry-ai .com, so that's subscribe. We've got weekday news updates, cool tools. Sundays, we do deep dive articles as well into different developments and developers, like Letstrip, and give you the rundown of what's going on in AI. So, thank you so much, Alon, for being here. It's always great to have cool guests like you, people who are interactive and passionate about what they do.