The Future of Mental Health: Why Theryo AI Could Be a Game-Changer for Therapy | S2 Ep. 160

0:05Hey, everyone. Welcome back to Founder Story. Today, we have two guests. I always really enjoy two guests because I think the more people, the more experience, the more that we can talk through. And I get excited because not everyone always agrees on everything, but we have Kanji Kawanabe and Alec Lervakos, who are the co-founders of Therio AI. And you're doing something that I think will revolutionize the world. I'm very excited about when we look at AI positively impacts on mental health and therapy and all that, which we're going to dive into everything. But let's just kick it off with both of you.

0:44What was the spark that made you say, I want to be an entrepreneur? Good to be here, Dan. Thank you. For me, it started early, kind of just out of college. There was a family accident and a family business. So I jumped into that. It was an accounting. I was supposed to be an accountant. I was going for me. So that kind of, while I was doing that, I was looking for other things to do and just kind of fell into entrepreneurship and then started one business and then kept going from there. And I guess my first step in being an entrepreneur was when I was a kid, I was 14. I had a paper route that I kept expanding and expanding.

1:21And I hired people to do the work for me. And then I ended up selling it. And when I told the newspaper I sold my paper route, they flipped out because that wasn't something you were supposed to do. So that was my, I think my first taste. Mine came from the startup world. So I've been advising in the startup world, in the tech startup world, Silicon Valley out in New York as well. So just working from about 2010, brother and I just kind of started an organization to streamline innovations within startups. Currently work with some VCs as well. And so just been in the area for a long time. always interesting when we ask this question because you can normally trace back entrepreneurship back to when somebody was a teenager they sold like candy they sold their paper out which i've never heard that before i wish i knew that when i was you know all 14 i wish i sold mine but it is fascinating how you can normally trace things back and it's not like a one-time thing it's like they continue to do you know conjure like you say you've continued to help and continue to do something you don't go back and get like a corporate job.

2:24It's very hard. I'd love to understand the company now. Why were you looking at the impacts of AI and mental health? Yeah. Well, Kanji and I, we kind of fell, not fell into it, but we were kind of guided into the direction of AI and mental health. We started in starting up a company for AI and doing consulting and building bots that were helping people with different things, personal issues, family issues. We're doing it for business. And then it started with me. Three years ago, I started therapy. It was something new to me. I had no idea how it worked. Never thought it would be something that I would need or have to do.

3:07So I went into it completely blind. And I think like a lot of people, it was a friend who referred somebody. and I went through maybe three or four different therapists, never felt like I was going anywhere. I didn't know how to really participate other than go in every session, sit and talk. And I came back after a session once and Kanji and I were working and I said, you know what, this is, I'm done with this. It's not working. I have no idea what to do. And if I'm even getting better, if I'm progressing. And Kanji said, I think we can build something with AI that'll help you. So it started out as a little app to help each other.

3:45And the results, we were getting feedback. When I was using the results in therapy, the therapist was like, where are you getting this stuff? This stuff is amazing. And then Kanji was getting similar feedback. So that's where we got the idea to go into AI mental health. And then as we got more into it, started to believe we could really make a difference in how mental health is delivered, the journey, the mental health journey, and we can even help people that don't have access to, don't have the resources or access to therapists or therapy. Yeah, and I'll add on to that a little bit. I remember that day distinctly when he came into the office and was just super bummed and it hurt to watch him, like he's so frustrated and it was nice to come full circle because Alex was the one that had first brought in, like this is how we can prompt, this is how we utilize AI to really pull out the most power to assist with folks in innovations.

4:42And I told him, I was like, hey, let's use your prompting and let's build out something to look at how we can build a cohesive database structure to understand where you're at now and where you could be in the future. Now, all that to be said, this actually parlayed after we did a class or a curriculum for our kids, because we also wanted to create an AI system or a learning curriculum for them to learn to utilize AI and LLMs responsibly and ethically. And I think that's also always been a guiding post of how we develop our software. I'm seeing something, which is always common, right? Entrepreneurs find a problem and you solve the problem and you realize if you have this problem, you solve it.

5:26Most likely other people have this problem and then you build out this business model and you have this business. I think AI could solve many problems where it's almost can be overwhelming for some people who are like, I want to solve a problem leveraging AI, but there's a million problems that need to be solved. How did you go from, okay, you found that you know the problem, you build something that can fix it. Like, how do you connect those dots as to your problem to product? I'll take this one actually, Alex, because I think we worked backwards. We knew where we wanted to be. We knew there was a missing link in as far as like the stakeholders that are involved with the process.

6:05So if I look at this operationally, it's like the technology that's involved, the stakeholders that are involved, and then different times that these different pieces are showing up in the equation, right? And so we wanted a holistic view as far as like, how can we get historical context and the full context? So what was missing was all of the times between sessions. So, and we also know we didn't want to replace therapists. There's something about the humanity or the human touch within therapy that's necessary, right? And then the expertise that they have. Alex, what did you think about that?

6:42This is a very unique approach in the sense of what you just said, Kanji, is you weren't trying to replace somebody, you're trying to enhance somebody. I think there's a lot of talks around replacement versus enhancement now. So how As you progress going forward, do you think that we create these things where it's an enhancement, but the AI becomes so smart that people might use it as a replacement? How do you see this? And maybe I'm not just talking within what you're doing now, just in general, because this is something I'm thinking about. As the AI becomes smarter and it learns more, I wonder how enhancement then becomes replacement.

7:24I think as I've talked about this quite a bit recently, I think it's really important for people to learn how to use it properly, not just kind of like Google and ask a question and get a response or do some basic things. So I think learning how AI works and how to prompt and not become a specialist, but just have a better understanding of it, being able to communicate with a team. And I think that's going to be a valuable part of working for a company to know AI, be able to communicate it, kind of help the company or the organization adopt AI. So I think that's going to help. And that's OK. And Dan, one of the things that I'll kind of caution against is conflating the AI with LLMs.

8:05Right. So there's the learning section and then there's the output section. Right. And most people only have access to the LLM. And so in the utilization of that, we need to be sure or understand at least where the questions and answers are coming from. And so our software uses LLMs. And so in the future, we will be building out AIs. Right. So that we've got our own learning structure. But it's really important to understand the difference there. Right. And then to your question about the fear of replacement. it comes up a lot, man. Like that's, that's the question of the day. It's like, one, is my job going to be gone?

8:42And two, why am I here? Right. The second question has been there for everybody. Why are we here? Why are we here? And we're not here to, to say, look, we're not here to answer all the questions about our existential crises, but we're here to empower those that can help us to learn how to be introspective. Right. And so, and that's the difference between empowerment and exploitation. We're not going to take away anything that defines somebody as their job, but especially in the provider and therapy world. What we're doing is giving them more tools to be more creative and more time with their clients.

9:21That's it. Yeah. AI, if they learn how to use it, will make them more effective and be able to do more in less time. So I think in that way, people will become more productive. And that's where some people, you might need one person to do the job of two, but that position is not going to go away. And the people that know how to use it correctly, it'll be more effective, it'll be more in demand. What do you see in terms of them adopting it? I think there's a lot of industries that could benefit, but as we know, not everyone is so quick to adopting technology or adapting to new technology. What are you seeing in terms of these therapists who are using it?

9:59Are they receptive? And then if so, what do they see as the benefit to themselves? Yeah, I can actually speak to this one. Just I've spoken with a lot of the provider beta testers. There's a learning curve and it differentiates depending on the amount of experience that the individual providers have with technology. And so this isn't a field that is traditionally built off of the structure of utilizing computers, right? Like talk therapy is exactly that. We're using our mouths and we're sitting on couches versus we're sitting down and using our computers. So it's a paradigm shift, to be honest.

10:39It's very exciting. And I'm glad you bring that up around, you know, these older industries that are not really technology-based. Now they have the option to use technology and amplify the ability to make positive change and good. So can you talk through maybe some, I don't know if you, not testimonials or reviews, but what are, what have these therapists said specifically around, you know, this helped using this or that? I know, Alex, you and I touched on this before you were telling me about a few stories, but can you share anything about that? No, I was going to say that I think the first thing is for therapists to understand that AI is not there to replace them, kind of like the direction that we're going.

11:23because I think a lot of people think that AI is going to replace therapists and then therapists get worried about that. I don't think AI is there. I don't think it'll ever 100 % replace therapists. So when they better understand that AI is a tool that they can use to be more effective and gather more information and to analyze more of the data and provide feedback, they become more receptive to that when they have a better understanding. right now like most people when you use ai for the first time or don't really use it very well you don't you know a lot of people don't understand like what the big hype is and why people are talking about it so the first thought of a therapist is this is a good we can't use this in therapy well and the proof is in the pudding and to answer your question dan when they first see the outputs they're blown away and it depending on the perspective that they're coming from the blown away can be in a good way or it can be in a bad way right so when I'm working with academics on their side that are used to working with data and seeing all of this holistic information coming in where they can start to digest it and they're used to it, right?

12:27They love it. And their eyes light up and they're like, this is amazing. How did you do this? Show me how you did this. Now, if I'm working with a traditional other therapist and they get all of this information, it's like drinking from the fire hose. It's too much too soon. And with that that particular demographic, I think we need to figure out a way to kind of walk them into the water, right, and figure out a way to show them the power of utilizing the information, but in a way that aligns with their traditional practice. I was reading recently that they tested AI. I don't know the exact like software program or L.

13:08I don't know a lot of details. It's just general, but they use AI and they saw how much better AI could detect diseases in advance compared to medical doctors. And they found that the AI that they used in this study was able to detect much better than medical doctors before, like whether it was like heart disease or certain things that they were looking at. And I think about how this will be with mental health, that maybe the AI can catch things being said that the therapist doesn't hear because they're just listening in an hour-long session. I'm sure they can't catch everything. But I wonder, or maybe if you've seen this already, that it can catch things maybe in advance of happening or detecting like somebody might have depression or maybe the therapist isn't picking that up.

13:57No, I can add to this. I was at an AI conference earlier this year and there was a presentation by Bright Health as a mental health institution organization across the country. and they did a small research where they just took ChachibT out of the box. They didn't have any database hooked up to it. They didn't have any special prompting. They used it for intake of mental health patients to try to identify if there was suicidal tendencies or ideologies. And they were shocked just that how plain AI could pick up what a doctor or somebody doing the intake would take, AI could pick it up faster. And a lot of cases pick up smaller things that the doctors didn't pick up when they were doing the intake.

14:41So that was just regular AI. So I think if it's used for specific purposes and it's trained or modified, whether it's mental health or anything in health, it'll be able to pick up a lot of things and help, again, help assist mental health professionals or healthcare professionals do their job easier and quicker. And that's where this symbiotic relationship is coming into play here. Alex hit that nail right on the head when he was saying it's all about the training and what are the data sets that we're feeding into it for it to even understand what it's looking for. And as we progress, those data sets become bigger and bigger and we get to explore.

15:21The nice thing about an LLM is they are one of the most effective thought partners that you can have in the room, right? And so to your question, Dan, again, is it going to replace and is it going to find things faster? It might discover possibilities faster. And I think that's where the discussion starts to happen with between the technology and the provider. And it really is. It's a back and forth discussion that, again, it's a paradigm shift. All of a sudden, I'm talking to a computer that's talking back to me, right and so to get over that shift let's say this is now the new normal right like that's just something that we've got to get to but once we get there the thought partnership becomes super strong when you realize you need to hire it's usually already too late i've been there wasting weeks waiting for decent candidates if i had used indeed i could have hired way faster so when it comes to hiring indeed is all you need stop struggling to get your job post seen on other job sites indeed sponsored jobs help you stand out and hire fast with sponsored jobs your post jumps to the top of the page for your relevant candidates so you can reach the people you want faster and it makes a huge difference according to indeed data sponsored jobs posted directly on indeed have 45 more applications than non-sponsored jobs one thing i love about indeed it makes hiring fast because it puts you right in front of the right people without wasting time plus there are no monthly subscriptions, no long-term contracts.

16:47You only pay for results. And while I've been talking, 23 hires were made on Indeed worldwide. There's no need to wait any longer. Speed up your hiring right now with Indeed. And listeners of this show will get a$75 sponsored job credit to get your jobs more visibility at Indeed.com slash founders story. Just go to Indeed.com slash founders story right now and support our show by saying you heard about Indeed on this podcast. Indeed.com slash founders story. Terms and conditions apply. Hiring Indeed is all you need. And accessible 24-7. So I said something at a conference recently that everyone laughed at, but a friend of mine last week, his girlfriend broke up with him and he was very sad.

17:34So he He put into ChatGPT their whole last few messages of text and asked ChatGPT to analyze and see if he did anything wrong or if he could have done anything different. He spoke to ChatGPT for four hours, had a conversation, and he felt really, really good. And it told him that he didn't do anything wrong. And I thought this was hilarious, but I'm like, oh, my gosh. And if you look at younger generations, they're so used to technology. I don't think they're going to have any issue with, you know, like talking to computers or using it because they're already like into it. I am curious about the future of where do you see the company going?

18:15If you look a few years out, I'm sure you're you're always looking ahead. Where do you see the turning out to be? I really hope that this actually reaches our marginalized community. So this is one of the things that we want to be able to utilize our software is to close those gaps between who has access and who doesn't have access to therapy. And marginalized communities have always been kind of a guiding light for us as well. And so being able to build out those structures that have the information to assist those communities specifically, because right now we don't, right? Like GPT, all of the current LLMs are built off of information that's accessible on the internet, which traditionally has been weaponized or therapy has been weaponized against these communities.

18:58So how do we build out these data sets that can actually be beneficial and support to these communities? So that's kind of the future of what we're looking at, as well as educating the providers to be able to utilize technology. I think we're hoping that we can do kind of the benefit that I experienced, we can do that for people, not just here in the United States, but around the world. I think especially in countries where, again, the resources aren't available or there's a mental health is stigmatized and people won't go to therapy. So this might kind of open the doors a little bit. And then another area that we want to go into is we're hoping that we can bring our platform into schools where you'll have a school with, you know, a thousand or 500 children.

19:42And there might be one nurse or one counselor there. And especially in marginalized communities or underserved communities, they don't have the resources. So families can't see anybody or don't go to see somebody. The only person they have is that person at the school. And that person is typically overwhelmed with what they have to do. to do. So if we have Theriot that's kind of listening in the background, picking up all of the little cues, providing feedback to the counselor or the therapist that's there, we feel that that can help them do a lot more in a shorter amount of time so they could see more children and maybe be more effective just because they have more data.

20:19Yeah, this is something I'm very passionate about. Going through struggles when I was younger, now I think we've seen this mental health crisis. I think suicide is one of the biggest reasons why young people die like in their teenage years that you'd think could be one of the most solvable things that we should be able to do but we can't and it's just like getting worse and worse and pandemic I think just made it even harder and harder on people's mental health so I'm super excited for what you're doing and the ability to to give what I what I of is like if you have a lot of money you can get the best care right in most in the u.s.

21:00in most places in the world if you have a lot of money you can get the best care you can see the best doctors you can see the best therapists so if you don't have a lot of money then you can't do that but maybe maybe ai llm can help amplify people to where they are as equal as the best doctor best therapist. So we're getting this equal care, essentially, versus like you said, great care is only for those that can afford it. This has been really amazing. I know beta is closing soon, but if you want to get in touch with you, they want to be able to sign up, therapists want to find out more information, schools, how can they do so?

21:38There's a QR code, I think, on the background. You can scan that. It'll take us to the website or just go to Therio, T-H-E-R-Y-O dot AI. Well, Kanji, Alex, this has been amazing. I really, really, this is one of the most exciting things that I'm watching for the future of AI, LLM, of being able to solve real world problems that are like a detriment to our society and humans all around the world. There's like not one person that doesn't have some sort of mental health issue, crisis somewhere in the world. So this is a human issue and you guys are really solving it. So thank you so much for what you do and joining us today on Founders Story.

22:20And that was such a pleasure and just the excitement and the fact that you get it is so exciting for us, right? Like, and sharing this. So thank you for what you're doing as well. Yeah. Thank you for exploring this with us and, you know, getting no word out about, you know, AI and mental health just in general and how it could be a tool. Thank you.