Edtech Insiders
Edtech Insiders
How Buddy.ai Became the Leading AI Tutors for Kids Worldwide with Co‑Founder Ivan Crewkov
Ivan Crewkov is the CEO and Co-Founder of Buddy.ai, the leading AI tutor for kids under 12, reaching over 20 million students annually from all over the world. Buddy is not just a chatbot. It’s an AI-powered talking animated character in a mobile app teaching English as a foreign language to students all over the world. Buddy is unique because it operates in a highly regulated environment where most off-the-shelf AIs like Chat GPT can't be used.
💡 5 Things You’ll Learn in This Episode:
- From voice AI experiment to global learning companion
- Building child-focused speech recognition in a regulated space
- Blending gaming, storytelling, and education to engage kids
- Buddy.ai’s growing adoption in classrooms worldwide
- The future of Buddy.ai as a personalized, adaptive tutor
✨ Episode Highlights:
[00:02:11] Ivan Crewkov’s journey from building smart speakers to founding Buddy.ai.
[00:04:24] Designing Buddy.ai as more than a chatbot—a gamified, lovable learning companion.
[00:07:27] How kids around the world discover and champion Buddy.ai themselves.
[00:11:02] Expanding Buddy.ai into classrooms in Latin America and beyond.
[00:16:41] Building proprietary speech recognition for children where others struggled.
[00:20:56] Buddy.ai’s unexpected role in speech therapy for kids with special needs.
[00:25:50] Why Buddy is designed as a study buddy, not an authority figure.
[00:33:37] Using generative AI to accelerate content creation and storytelling.
[00:39:00] Guardrails for safe, child-centered AI companions.
[00:42:01] Lessons EdTech can learn from the entertainment and gaming industries.
😎 Stay updated with Edtech Insiders!
- Follow our Podcast on:
- Sign up for the Edtech Insiders newsletter.
- Follow Edtech Insiders on LinkedIn
Innovation in preK to gray learning is powered by exceptional people. For over 15 years, EdTech companies of all sizes and stages have trusted HireEducation to find the talent that drives impact. When specific skills and experiences are mission-critical, HireEducation is a partner that delivers. Offering permanent, fractional, and executive recruitment, HireEducation knows the go-to-market talent you need. Learn more at HireEdu.com.
Every year, K-12 districts and higher ed institutions spend over half a trillion dollars—but most sales teams miss the signals. Starbridge tracks early signs like board minutes, budget drafts, and strategic plans, then helps you turn them into personalized outreach—fast. Win the deal before it hits the RFP stage. That’s how top edtech teams stay ahead.
As a tech-first company, Tuck Advisors has developed a suite of proprietary tools to serve its clients better. Tuck was the first firm in the world to launch a custom GPT around M&A.
If you haven’t already, try our proprietary M&A Analyzer, which assesses fit between your company and a specific buyer.
To explore this free tool and the rest of our technology, visit tuckadvisors.com.
[00:00:00] Ivan Crewkov: So the path we walked was a, a multi-year and, and pretty expensive. Okay? So we had to launch our product in multiple countries, in multiple languages, collect a lot of data and train our model the largest. Available data set of kids speeches, like 500 hours. It's just pathetic. Yeah, I mean, we collect 10 x of that every month, and that's the only way to train the model.
[00:00:30] Alex Sarlin: Welcome to EdTech Insiders, the top podcast covering the education technology industry from funding rounds to impact to AI developments across early childhood K 12 higher ed and work. You'll find it all. At EdTech Insiders, remember to subscribe to the pod, check out our newsletter, and also our event calendar, and to go deeper, check out EdTech Insiders Plus where you can get premium content access to our WhatsApp channel, early access to events.
And back channel insights from Alex and Ben. Hope you enjoyed today's pod.
We're here with Ivan Crewkov, the CEO and Co-founder of Buddy.ai, the leading AI tutor for kids under 12, reaching over 20 million, even more students annually from all over the world. Buddy is not just a chatbot, it's an AI powered, talking animated character in a mobile app. Teaching English as a foreign language to students all over the world.
Buddy is unique because it operates in a highly regulated environment where most off the shelf ais like chat, GBT can't be used. Ivan Crewkov, welcome to EdTech Insiders. Hey Alex.
[00:01:43] Ivan Crewkov: Thank you for having me. It's a pleasure, an honor to be here.
[00:01:48] Alex Sarlin: I'm so excited to talk to you 'cause I feel like you're doing something that is really pointing the way towards the future of what AI learning will look like.
And I don't say that that often to people. So first off, before we get into what Buddy.ai is, tell us a little bit about yourself and how you got into the education technology world and how you came up with this Buddy.ai idea. So I am an
[00:02:11] Ivan Crewkov: immigrant parent. Voice AI developer, let's say. And that's kinda how I got there.
I moved to the US almost 11 years ago with my previous startup. We were building a smart speaker before this whole category was like created by, so there was no Amazon Echos or Google homes. And we thinking like real innovators and we. We were preparing to launch our Indiegogo campaign and two weeks before that Amazon announced Echo and it was like, like a disaster.
But anyway, great experience. We learned a lot about voice technology and the conversational ai. Back then, it was 2014, long before Chad GPTs and Geminis and yeah, so I, I brought my family with me. And my, my daughter, so she struggled to start speaking English in preschool for the first four months, and it was a real problem for her.
And we started taking lessons on platforms that connect children with live teachers via video conferencing because it was like,
[00:03:26] Alex Sarlin: mm-hmm. The
[00:03:26] Ivan Crewkov: only way to actually. Teach a child to speak. Duolingo is great, but it does not teach speaking unfortunately. At least back then it wasn't. And I just started recording those sessions with life teachers and realized that they were scripted, like teachers were literally reading from the, their screens, and most of them, they weren't actual teachers, but some random people teaching via video conferencing.
And I just realized that probably 80% of. This work could be done using virtual to eye character and speech recognition. And so we could like dramatically democratize this market because it was extremely expensive. I know even back then it was like 15, $20 per lesson. So the idea was to provide a month of learning at the cost of one tutoring session, and that's kind of how we got to where we are.
[00:04:24] Alex Sarlin: Yeah. And then of course what you came up with was beyond just a character or you know, just a, a sort of automated teacher. It's this whole world. It's this whole Buddy.ai sort of learning companion. You can dress him up. He has, you can feed him, you have adventures with him and you've sort of created this whole gamified world.
And at, at this moment, Buddy.ai, I have two different stats here. Tell me which one is right buddy. AI has reached over 50 million. Is that right? Students globally. Now. Now it's almost 60, Alex. 60 million. Wow. So we had 20 million in your bio. Now it's 50, now it's 60. You are growing very quickly, 60, almost 60 million children globally.
Tell us about what Buddy.ai actually is and how you develop this system that is so exciting for kids. It's popular and it just creates a real immersive experience that feels, I think, as. At least is exciting, if not way more so than talking to one of these human tutors that is reading a script.
[00:05:22] Ivan Crewkov: Yeah.
Alex, so Buddy, as you said, is an AI teacher. Like literally it's a, it's a talking animated virtual character, but we purposely designed it as a virtual, it is a robot, basically like a cartoonish robot. One of its, uh, key properties. We call it magic of voice. So it's basically we want children to feel the power of their voice when they speak to body.
It just like every time you say something to bodies like casting a spell. So we provide a rich visual feedback to what users are saying. You know, like say Blue and Body would change his color, say unicorn and.
Things like that. So yeah. And. We wanted also to create, you know, like a, uh, a fill of a game in many ways. Close to, you know, like tamagotchi experiences where you take care of your virtual whatever friend path and to do it, to do so, you need to progress through curriculum and that's where you face, you know, challenging educational
[00:06:35] Alex Sarlin: experiences and learn.
You're describing it in words, but I really recommend anybody listening to this should open it up and look at what Buddy actually looks like. It is an adorable character. I definitely, you have this sort of tamagotchi, it's like almost like a little flying has a rockets for feet and these big ears. It can dress up in any kind of thing.
I got Tamagotchi, but I also get Pokemon. I get a lot of these really, it looks like a very, very sophisticated, you know, children's property's. Say to use the sort of industry terms and it can do so many different things and there are all of these different games. There's mode, there's sort of a fighting mode that looks like a little Pokemon mode.
There are all of these different dress up adventures. So before we even go further, tell us a little bit about how children react to this type of learning. Especially, I'd love to say, compared to sort of traditional learning of almost any kind. What do they say when they, you know, how do they react to buddy?
[00:07:27] Ivan Crewkov: So first of all, outta this like 60 million downloads, almost 70% are done by children directly, Alex. So it's hard to believe, but I don't know, like four or 5-year-old children download. Buddy and then pitch it to their parents, especially outside of the us. Uh, so in the US we're mostly parents download the app, but outside it's, it's, it's mostly, uh, all about children.
So we have, I dunno, a whole like fandom, a bunch of user generated videos, uh, on YouTube about Buddy. So we're trying to build. Ip, as you said, like a A game world and really for children, create an impression of not just an app or in like a learning program, but more as an entity. It's an alien explorer robot.
He has his YouTube channel, which is also growing like over 20 million organic views on our YouTube channels, stuff like that.
[00:08:23] Alex Sarlin: Yeah, you mentioned Duolingo sort of in passing there, but I think, you know, we talked to a lot of EdTech companies of all different types of models and it, it always strikes me as there's such a different playbook and such a different type of strategy for EdTech companies that are going directly to students and parents than ones going to schools and districts and colleges.
And when I look at the. Sort of level of care put into the animations here, put into the character design, put into the costuming, put into the structure of it. You know very well that you're competing with YouTube. You're competing with Disney Channel, right? You're competing with many other very high production value properties for kids' attention.
So you put a lot of production value into your product. Tell us a little bit about. Those choices because I think this is something that we see a lot more on the B2C side, and I think it really pays off on the B2C side because as you say, students find it, they tell each other about it, they're gonna download it directly, they're gonna love playing it, but sometimes B2B companies just don't think like that at all.
I'd love to hear you just talk a little bit about sort of the decisions you make to say, okay, we really want this to look as polished, as beautiful as anything kids see on TV or on on YouTube.
[00:09:35] Ivan Crewkov: You mentioned competing top.
Set a goal set to be as, you know, like as engaged as top mobile games for kids. The way we chose is actually we are building, so our team combines like AI engineers on one side, but also a very strong game design team. So we're, we're like a lot of, we have a lot of people from game industry, like our head of product is Aof game person.
We head of game design who is a former. And outfit seven person. Our art director is also from game industry, so we have a whole team of like game industry professionals working to make it as engaging as possible and they're working with educators. So educators set basically like the objectives they told We need to teach this and that.
And in that order, and then our like game design team comes up with game mechanics and all of those cool things we have.
[00:10:43] Alex Sarlin: Yeah. And then you have all of this very high production value, game experts creating the games, but you do have a school product, you have a sort of classroom version of Buddy. So tell us how that works.
How does the, what do you do when you're giving teachers and classrooms access to Buddy.ai and how do you set them up for success?
[00:11:02] Ivan Crewkov: We are still working on it. So the fact is many teachers. Use in their classrooms or in their one-on-one lessons without having a proper functionality. And unfortunately we haven't had an opportunity to develop this features specifically for teachers.
So it's like an unintended that use, which means. Teachers we can create value for, for teachers and schools as well. So we have several pilot projects now running with schools, mostly in Latin America, in Brazil and Mexico, but. We plan like a real big push for next year. So we plan to raise our next round of finding, uh, next year specifically to in invest more in this features that will make body even more useful for teachers, the reporting class management and stuff like that, right in work assignment.
[00:12:04] Alex Sarlin: Your classroom product, you offer, you know, teacher packs of licenses so you can get, you know, 20 or 30 licenses with, with discounts. You obviously have the ability to work and you are being used in a lot of classrooms informally, but as you mentioned, it's always an interesting. Decision point for, for EdTech companies about this.
And Duolingo famously really never did create a true school product that they're sort of, as you know, of course, the sort of gold standard of B2C EdTech. And they are used extensively in schools and yet they never actually created a full school product. And my understanding is they never don't truly have, you know, teacher dashboards and analytics and admin controls and some of the things that people do.
So it's an interesting choice and you're obviously right at the sort of fork in the road there about how you're gonna build your teacher side. It's a very. Interesting moment. I can imagine inside your strategy room thinking about what the classroom use case looks like. So you mentioned that you're working in Brazil and one of the other things that's really interesting about Buddy and of course very modern about EdTech right now is that you have already have Buddy offered in many different languages, including Portuguese for Brazil, including Polish, including uh, Spanish and French, and Russian and Turkish.
Tell us about. That process of making sure that you're creating more and more opportunities for different students with different linguistic backgrounds to learn, especially given that you're teaching English. So you have so many different ways for students from different linguistic backgrounds to get into this English learning environment.
Tell us about how you're doing that translation and how you're accelerating, how you're using technology and your background to make sure that you have many different languages supported.
[00:13:39] Ivan Crewkov: Yeah, Alex. So first of all, we actually started from the one as a network international company, and we launched in Europe and Latin America, actually before we have in the us.
And this choice we made because of technology and regulation. So we already discussed that we operate in this highly regulated field with COPA and similar regulation. Sure. And so. In this weird regulated world, there are not a lot of off the shelf solutions that you can just take and build an MVP. So we had to build, I mean there were of course, uh, soapbox labs and solutions like that, but they were expensive and we weren't happy with, uh, with quality actually.
And so. What we did, we, we built a whole AI stack in-house and to do it, you need data. And so we just launched in countries where there were no strict regulation on data collection for children under 12. So we started collecting data there, teaching our our model. And then when we had a model that basically Google on kids.
Data. Then we launched in the us got a COPA certification and uh, things like that. So it was just a part of this like big AI long multi-year AI building strategy that we
[00:15:04] Alex Sarlin: had. Let's double click on that. You know, I was asking about the language background, but what you're answering is actually even more interesting because this is such a tricky part of the EdTech landscape right now.
Everybody, our listeners to this podcast know that, you know, soapbox was sort of the B2B. Lead in the space, but they got bought by curriculum associates and made sort of proprietary and private, and everybody has been left sort of scrambling to figure out how to create their own kids' voice model or sort of to solve this.
It sounds like you've been thinking about it from a very early on, and frankly that's a very clever move to go to places where you can collect. Kids don't voice data in a way that can train your model, but without having to go through quite as much of the bureaucratic red tape. This happens in the us.
The fact that you can outperform the Google model on kids for that is very impressive. And that, I think is a tribute to the, probably the technical prowess of your team. Tell us a little bit more about that decision, because we've talked to a few different companies that are trying to do kids voice recognition and they've all sort of had to wrestle with all the, especially since so box sort of came off the table.
So you mentioned that you are a voice engineer, that you are already working on voice recognition and smart speakers. Before you started Buddy, how would you recommend that the EdTech field sort of think about this collective problem of needing to have voice recognition software for students? You have obviously built your own proprietary model.
Soapbox was one that was offered to lots of different companies, but is now. Fully incorporated into curriculum associates. How do you think the whole field is going to be able to get over this hump and make sure that kids can actually operate AI tools like this with their voices?
[00:16:41] Ivan Crewkov: Alex, this is such a hard question.
I know. I actually dunno. So the path we walked was a, a multi-year and, and pretty expensive. So we had to launch our product in multiple countries, multiple languages, collect a lot of data and train our model. It's the largest publicly available data set of kids speeches, like 500 hours. It's just pathetic.
Yeah. I mean, we collect. 10 x of that every month. And that's the only way to train the model. But there are other providers other to soapbox now I just, I'm just unsure how good are they because we focused on building and training our own model and just. Unsure on this, like the, the status of outsourcing and uh, I mean open source and other alternatives on the market today.
[00:17:32] Alex Sarlin: Would you ever consider sort of offering your model as a service to other ed tech providers, or do you think anybody in the space is gonna consider that as an option?
[00:17:42] Ivan Crewkov: Yeah, we've been thinking about it. Alex asked a lot by other tech companies, so I think we'll need to do some work to make it actually marketable so anybody could.
Take it and plug it in their products. But yeah, we'll, we'll definitely consider this option after the next round of finding.
[00:18:02] Alex Sarlin: Yeah, it's very interesting. So let's talk a little bit more about the student experience in Buddy Dead ai. 'cause it's incredibly interesting. You really built it from the ground up with kids in mind.
Everything. There's video, there's games, there's lots of characters, there's lots of, but you also have the rigorous curriculum and you use space repetition and you use pedagogical methods. How did you think about building this entire. English language world from the ground up. And how did you use AI to really accelerate that process?
[00:18:31] Ivan Crewkov: Alex? So first of all, body is not only English as a foreign language product already, the bigger idea is, you know, like an AI tutor as a platform that a child's. Learning AI assistant that would grow with the child from, let's say age of four, teaching one subject after another while collecting learning pattern data and creating a personalized learning plan like using adaptive learning technology.
So that's the bigger vision we have. So today we have two curriculums and developing the third one. So original curriculum is in English as a foreign language, as you mentioned. That's their biggest, biggest. It's like 80% of our revenue for sure, but even more in in terms of users. Later we launched like an early learning, like an MVP of an early learning course, and about a year ago we actually discovered that about half of our.
English speaking world have been using Buddy as a speech therapy tool. Wow. And we have a lot of children with, uh, special needs on the platform. And so we started developing this new specialized speech therapy course and we know that, so I would say that in the English speaking world, so our most active and engaged users are.
In this space. So they use Buddy for speech therapy, speech and language therapy.
[00:20:05] Alex Sarlin: Yeah. That's wild. And so, and that's a tribute also to your voice model, not to get back into it, but I mean, some of the things that people wrestle with with voice models is that they have to be able to work with a wide variety of different accents and dialects, but also different speech impediments or di, various kinds of speaking.
The fact that you're. System works that well for students who are in need of speech therapy. It actually works for them, is incredibly exciting. I love that direction to go. That's a big population. It's a population that really, really, you know, needs support in this exact way. There are just simply not enough occupational and speech therapists, you know, out there in the schools to help, especially around the world.
[00:20:43] Ivan Crewkov: I agree, and I'm super proud that we started moving in this direction and it also happened organically. We just learned that. People started using our product. Speech therapy.
[00:20:56] Alex Sarlin: So when you talk about your broader vision about, and this being an AI learning companion, you know, that's how I first came across Buddy.
You came on our radar as we were looking across the landscape about who is doing what. And there's this category in Thetic Insiders Market map of AI we call learning companions. And there are not actually as many as I expected to be in there. Uh, at least so far I think they're probably growing. It's a growing category, but you guys.
We're very clearly at the very cutting front edge of doing learning companions in a really thoughtful way. I mean, buddy really is just very consistent, persistent throughout all these different experiences and games. I can only imagine that people just feel like they know him. And the the sock puppet unicorn character, what is that character's name, sparkles.
Sparkles, Mr. Sparkles. Exactly. They're fantastic. One thing that we wrestle with a lot here, and I'm sure you think about a lot too, is when you're thinking about a persistent learning companion, a core aspect of it is memory is what do you know about the student? What do they like doing? What do they, what are their speech issues?
What do they care about? What are they excited about? What's their favorite game? I'm curious how you've begun to, I'm sure you've thought a lot about this, how you've begun to get into that idea of if we're gonna have students using buddy. For months or years on end, what do we have to know about that student over time to really make the experience as successful and personal as possible?
[00:22:20] Ivan Crewkov: Yeah, Alex, so the, if the idea is to provide actual personalized tutoring, just like having an an, an actual. Then we need to understand a child's learning patterns. You know, not just like some preferences, but so let's say this character, this girl better, this kind of games, for example, work better. So this boy needs more repetition and things like that.
And so really to make it work like longer term, we will use like adaptive learning. Like AI-based adaptive learning technologies. To be super frank, we just started working in this direction. Sure. So far we've been primarily focused on just providing core experience speech, like understanding what children are saying and providing like, uh, reach, conversational, voice-driven experience that would be.
Educational. Educational, so like real educational goals, but shorter term. So today, most of our students, they just run through the curriculum that we have in under six months. And so this is one of the things that. And that's why we don't have like an annual plan, for example, today on the platform. Yeah.
But yeah, we're adding more and more and more content. I can imagine.
[00:23:43] Alex Sarlin: Yeah, that's, yeah. Yeah. Yeah. That's a great opportunity. They, when people are zipping through the entire curriculum, that means you have a lot of opportunity to expand, so that's really exciting.
[00:23:52] Ivan Crewkov: Yes. But children stick. With Buddy, even after they learned everything, just because they, they love the character.
I know they send us pictures of, of Buddy and produce YouTube videos about him. So yeah, there's definitely a huge room for improvement.
[00:24:11] Alex Sarlin: Yeah. You say this is nascent and I, the whole field I think is working on this. There's nobody has cracked this idea of how to build a knowledge base of in each individual student over time.
People are working on it in all different ways, but I think that's a really interesting aspect of this. But let's talk just about this whole concept of learning companion, right as we got on the the line today. Something that has been, that I've been thinking about a whole lot in the wake of all of these frontier model of Anthropic and Google and OpenAI all launching these.
Tutoring modes, these sort of study modes inside their core products, which is really exciting. We're really very bullish on that, you know, at EdTech Insiders. But what they are all doing is sort of trying to simulate a tutor or a teacher or you know, somebody who looks and feels and acts at least somewhat like a traditional, what would consider a authority figure, a teacher who's working with you.
And I think something that you guys are doing that is incredibly forward thinking and frankly really, especially for younger kids, incredibly powerful is. You are thinking about teaching in a different way. It doesn't feel authoritative. It doesn't feel didactic. It feels like you're playing. It feels like you're sort of collaborating and playing and helping buddy out and dressing him up and, and doing games alongside him, and it doesn't feel like, I think that that is.
Directionally something that we're gonna see a lot more of in the future with AI and ed tech because it just feels like a rich vein that we've literally never been able to do before to create really rich AI personalities that can work alongside students. So I'd love to hear how you think about sort of Buddy's personality and how you make him somebody who kids wanna spend time with and actually wanna learn from.
And it doesn't feel like a sort of teacher personality at all.
[00:25:50] Ivan Crewkov: Yeah. Yeah, Alex, thank you. That's super thoughtful. Not a lot of people dig so deep, but yes, the idea is that from the character building standpoint, buddy is more like a study buddy. So he learns with you, so he does not try to teach you. And we purposely remove all of the educational vocabulary from our product.
So like when Buddy interacts with the child, he would never say, study, learn, or things like that. He would say, explore. Go to adventure and, and things like that. We don't use any symbolic from a, I dunno, like a virtual classrooms and stuff like that. No. So we're trying to create a captivating story and teach through immersion and practice.
So it's like you study by solving. That body has so buddy is hungry. You need to feed him and you learn. So let's say you learn English or you improve your speech to help your virtual friend, virtual pet.
[00:26:58] Alex Sarlin: That puts the child in the driver's seat, right? Yep. It creates a gig. A lot of agency. It gives them the sense of authority and expertise, which children so rarely feel in their educational lives, right?
They feel like they're always the one sort of being taught, being told, being asked, but there's something very different about being in control and being like, I'm gonna help a friend. I'm gonna work. I'm gonna go on an adventure with somebody. I'm gonna play with somebody else. It's such a different way of looking at learning, and frankly, one that I just think is.
Endlessly exciting. That's why I got into ed tech in the first place. Personally, the idea of turning the didactic nature of learning into one that's much more playful and exploratory. I feel like you guys are really doing it in a way that is just, I see it as very much a vector, you know, pointing at where ed tech is gonna go even more than a Duolingo, frankly.
I mean, Duolingo also is, doesn't come across as super didactic, but at the same time, it is a little bit more of a sort of structured learning feel. The fact that you say you purposefully take the educational vocabulary out of it. That's really interesting choice.
[00:27:57] Ivan Crewkov: Yeah, yeah. Yeah. Alex. And, uh, another like important thing you mentioned is this agency.
So in buddies who have this like a story bible, you know, like games or like Hollywood movies and uh, it starts with Buddy is so main character protagonist as a player and Buddy's a sidekick. So that's kinda how we position it,
[00:28:17] Alex Sarlin: which is. A big deal. Right. I I think that's not, you know, it's interesting, I've been playing with some of the new, I, this is just hot off the presses at EdTech.
'cause the AI study mode, open AI study mode just came out a week ago and Google just announced it yesterday when we're, as we're recording this. So we're like, I've been playing with these tools and it, it's really interesting because technically the same dynamic. Technically is at play here, right? You as the user are sort of the main character.
You're driving the the experience. But at the same time, theis tend to very quickly sort of take over. So you say like, I wanna learn this. And it says, okay, why do you wanna learn it? What's of interest to you? And you say this and this and this. It goes, okay, well here's your curriculum. Here's your, here's how we're gonna do it.
We're gonna go 1, 2, 3, like four. And then it gives you a choice at the end. Do you wanna do it this or that? But immediately you're like, I don't feel like I'm sort of driving this experience anymore. It's like, it, it, it's very quick to grab the wheel and sort of be like, you told me where you want to go, you know, I'm ready to go there.
Very happy Butler feeling. And I really think that it's very interesting even though, you know, I can see the value there. From a personality standpoint, it's not a personality you wanna spend a lot of time with, right? If they just sort of do what they do, what you say kind of blindly, but then are sort of can't help it.
So it's like, I know I know more than you about this and I can't wait to sort of just like pour it on you. I just think that there's something very off about that personality. I think we're gonna evolve as an ecosystem. I think EdTech is gonna evolve towards a much more friendly. Buddy like and sparkles like attitude.
So tell us more about the YouTube channel. I'm really interested in this. So what, what do you do with the YouTube channel for Buddy? 'cause that is obviously a not an interactive medium right now. It's much more sort of broadcasty. What, how do you take your characters and put them into a YouTube environment?
[00:29:54] Ivan Crewkov: Yeah, so we use the same technology and it enables us to extremely inexpensively and fast reduce. Like cartoons basically. And we started doing this before generative AI because like we have this technology developed in house that generates by this visual behavior based on the context of conversation.
And yeah, we just use this technology to create short educational videos. So for example, body teacher colors. And it's, let's say this specific video is the most popular colors teaching video in Portuguese, for example. Yeah. And we know that teachers use our videos in their classes in preschools as this like supplementary materials.
And that was a part of this whole strategy. But also we produced like funny. Videos, but also like, I mean, every video is like educational. We still teach some vocabulary or some concepts in English. So the idea is just, it's a way for, it's a way for us to build our brand, get some traffic. Sure. The, the limited resources we have as an early stage startup, but also increase our impact like our educational impact.
And that's, by the way, Alex, one of the primary reasons we actually wanna go to schools. So the, the impact thing, the distribution. So AI is, our technology is super optimized and we are working globally. We are optimizing for a, I don't know, old cheap Android devices. So it'll work in like a village in the remote village of Guatemala, but still pretty expensive to operate.
Which means we'll still have to charge money and plus we can't use advertising and so. Reach this global coverage to really help every child in the world. We need, at least in developing countries school system to pay for it, to provide it to every child. So this is the part of the strategy. That's why we're gonna schools.
[00:32:05] Alex Sarlin: It makes a lot of sense. We hear that from a number of different ed tech founders that when you want to have impact, because that is one of the big trade offs of B2C is that you're, you're putting the, the onus of cost onto a parent that creates the, you know, that sort of feeds into the inequality in the world and, and in, yeah.
In particular countries, but especially globally. It makes it a sort of off limits. Yeah. I'm looking at the YouTube channel now. It is so much fun. You have like hundreds of videos on there, so let's talk about the sort of content creation piece of this. As well. You know, you mentioned how you use the same engine to create the YouTube videos.
One of the things that we've talked about a lot on this podcast is how, what one of the really exciting parts of what AI can do is it can significantly lower the cost of content creation, even though it doesn't necessarily lower the cost of VPI. You know, there's still like when you said, oh, the child says unicorn and buddy turns on his unicorn Yeah.
Outfit that takes an API call that takes compute. There's a cost to that. But the ability to create that unicorn outfit, you know, it might be a lot lower because you could use AI for content creation. I'd love to just hear you talk a little bit, because you create so much content in various types of, you know, media.
Talk a little bit about some of the ways you think about AI for content creation, for creating new videos, for creating new environments, for creating new assets. Has that been something you've really embraced and do you think it's going to continue to sort of go on a downward slope, or are there. Things that people should think about if they're, if they're also in that world of, of trying to make more and more curriculum, more and more assets and games, how should they think about the AI for content creation?
[00:33:37] Ivan Crewkov: We're super excited about it. I think we need to just embrace this new brave world and, uh, we're doing it a lot. So like we use AI for pretty much everything. So even the, we are now updating our visual style. Use of gen AI assets look more organic basically. So to make it a part of this whole style. So we use genai 4 0 4, everything from producing, let's say backgrounds for our videos or YouTube videos to 3D assets in the app.
And of course like speech. Things like that. It's huge. I'm in love with these tools.
[00:34:17] Alex Sarlin: They're really incredible. And I mean, do you envision a future where somebody can touch Buddy and say, antelope and even if you don't already have a preset antelope asset, it could actually sort of near real time say, antelope, great.
We're gonna make you an Antelope 3D asset and costume for Buddy and put it on him. Is that something that, is that science fiction or are we like six months from that?
[00:34:39] Ivan Crewkov: I would say we are maybe. Couple years from that. Mm-hmm. It's definitely not a science fiction and if we wouldn't be in a kid space, then probably would, it would be even easier for us.
But we need to be careful with, you know, like security wise, because who knows what. Child may ask Buddy to create, right?
[00:35:01] Alex Sarlin: Yes. Yes. Uh, so, so actually, speaking of which, I wanted to ask you one other thing about this. You know, a, a few months ago I talked to the then CEO of BrainPOP, and I asked him whether he envisioned a future where the BrainPOP characters, which are, you know, characters who have been around a long time, have a lot of IP in the ed tech world.
If they'd ever do something where you could sort of ask their characters a question and it would respond in real time using, you know, AI generated. Voice and AI generated content and he said we'd love that in theory, but it's a little scary. We've spent all this time building up this IP and the idea that children could try to trick our robot and our characters into saying things that are unto or problematic in any way, and then suddenly we own that.
Basically the idea of, you know, connecting generated content to their, their IP was very scary to them and I thought that made to tons of sense. It was a really interesting take. So I'm curious how you approach that, right? You're creating new IP and Buddy, the robot and Sparkles Unicorn are sort of AI native characters, right?
They're made in the AI era, but how do you handle that idea of how do you make sure that the child can never trick buddy into saying something that's a problem, and then recording it and then creating a scandal out of it?
[00:36:12] Ivan Crewkov: Yeah. Alex, first of all, I fully support fellow, fellow entrepreneur. Of course. Yeah.
I believe that like bringing existing ips to life, using general rate of ai, even if you have tons and tons of data for this specific characters is. Risky.
[00:36:33] Alex Sarlin: Yeah.
[00:36:34] Ivan Crewkov: And I'll know it won't still most probably be feel authentic. I just imagine talking to a joker. Right. So it wouldn't be like, it should not be a pleasant experience.
Right.
[00:36:46] Alex Sarlin: That's an interesting example. Right?
[00:36:48] Ivan Crewkov: Yeah. And so Buddy, as you said, is an AI native character, right? We're building IP specifically for this gen ai. World. This concept of magic of voice is native to this gen AI world. How do we build guardrails? So first of all, our approach is not to try build guardrails, but to cut everything that.
So maybe build a less sophisticated and versatile model, but this model would be hundred percent focused on this like curriculum.
[00:37:24] Alex Sarlin: Mm-hmm. And
[00:37:24] Ivan Crewkov: will be just completely free of like dangerous. Material. And same with our visuals today. So we are, body understands like AI controls which asset to use for magic of voice, but we still have like a pre-generated library of assets.
Yeah. So a child would not ask body to, to create something weird, scary, or
[00:37:48] Alex Sarlin: inappropriate. Right. It feels like an interesting, it almost feels like an uncanny valley moment in AI right now where like, as you say, you know, if you're Disney and you have 500 characters that are like world class ip, right? From everything from Marvel, everything from Star Wars, everything from your Disney Library, every Disney movie, and you're like, well, would we ever wanna have Ask Mickey something or ask.
Spider-Man something, and suddenly the hackles go up. But for all, for all the right reasons, it's like, oh boy. But at the same time, the power of generative AI is that it can improvise and that it knows everything and that you really can go as deep as you'd like. Right? If you had a kid who wanted to learn who's doing their English vocabulary and doing fruits, and they say, actually, my favorite fruit is a durian, and it happens to not be in your library, but it would be awesome if it could say, durian, that's a great fruit.
This is what it tastes like. Let's talk about it. Yeah. Yeah. There's, there's this funny moment where you're sort of like LLMs know everything, but when you put it in front of, especially kids, you have to make it a little dumber. You have to dumb it down. You have to sort of create a around it, dumb down.
You have to dumb it down a little bit. Exactly. Yeah. I'm curious how you think that's gonna evolve, especially when it comes to persistent learning companions that really don't wanna be limited, like super limited in how they respond.
[00:39:00] Ivan Crewkov: Yeah. First let me say that we should be excited about the new generation of IP born in this wonderful new gen AI universe.
And it's an opportunity for storytellers too, right? Yeah. So it's like every new technological shift. Like, remember all of those, like great new games created for mobile phones, smartphone. Yeah. Stuff like that. So let's, let's be excited about that, about new stuff we can build with it and talking about.
Models and educational models. So our take and our like approach is AI companions actually as we believe they could be dangerous. They could create, you know, like part of social relationship you, you heard probably about, you know, this terrible stories about, I won't name the names, but people suffer. So we believe that AI companions for children should be strictly educational.
So they should be built around curriculum, so, and should be stripped off everything else. We don't want Buddy to talk about politics, right? Or we also don't expect Buddy to write an essay for you or help you with homework. Our approach to building AI assistance for children is to make them strictly educational and allow building this like social relationship.
Only as a part of the curriculum, right, only to make it learning more effective, more engaging.
[00:40:35] Alex Sarlin: It makes sense. And I mean, I, I think that can still be a very rich experience. 'cause a curriculum can be quite vast, right? It doesn't mean that it, it has to only talk about something very, very proximal to what's being talked about.
If your curriculum is English language learning and you have hundreds and hundreds and hundreds of, of vocab words and. Yeah, I'm looking at your videos, you know, daily rituals and alphabets and animals and letters and traveling and all this stuff. You know, you could still have a very rich and interesting conversation with Buddy, even if Buddy is limited to the curriculum.
You've mentioned mobile gaming a couple of times and how B2C tools as EdTech tools like Buddy.ai have to compete with a wide variety of mobile games and videos. Those types of developers, game developers and video developers are also getting incredible AI tools and they're able to accelerate their content creation and their creativity and their ideas as well.
I'm curious how you think the ed tech world. Should think about keeping up with the entertainment world as close as we can in a world where everybody suddenly has this AI acceleration.
[00:41:42] Ivan Crewkov: I just believe we need to do our best and also learn from the entertainment industry. We need to learn from the game industry.
So we have, for example, showrunners among our investors who help us with like a storytelling and stuff. So we just need to learn from the best.
[00:42:01] Alex Sarlin: Yeah, yeah. Fast followers, right? Whenever interesting things happening in entertainment, how do we make sure that we're not left behind in education? We make it.
We use them in the service of impact. I, I like that answer a lot. This has been such a fun conversation. Sorry for a couple of technical glitches on my side. Ivan Crewkov is the CEO and Co-Founder of Buddy.ai, the leading AI tutor 60. Million children globally, they're approaching, it's for kids under 12 for learning English, and it is, I highly recommend checking it out.
It is a really, really exciting program. Something that I think points the future of where AI and education really are gonna go. Thanks so much for being here with us on EdTech Insiders.
[00:42:39] Ivan Crewkov: Thank you, Alex. Thank you everybody. Best of
[00:42:41] Alex Sarlin: luck. Thanks for listening to this episode of EdTech Insiders. If you like the podcast, remember to rate it and share it with others in the EdTech community.
For those who want even more, EdTech Insider, subscribe to the Free EdTech Insiders Newsletter on substack.
