Scaling Trust: What Mediktor's AI Triage Tool Actually Does

0:03So, when it comes to healthcare delivery, and we are talking about healthcare companies providing healthcare delivery, you need a professional tool. You cannot use a general purpose tool, and these tools need to be integrated in the flow of healthcare. So, this is a very niche way of approaching this, not a general way. And for that, of course, you can blend technology, but you need to put in place all the safeguards for healthcare. So I think that all big companies have realized this. I'm talking about big companies in healthcare, and they look for professional tools that maybe they can embed LLM or not, whatever.

0:48I mean, I don't think that LLM is a need. I think LLM is an option that comes also with cost and cost efficiency in some places of the world. It's extremely important also. And of course, it needs to be in this healthcare delivery totally integrated. So I really believe that, yes, it's a huge disruption. It's something that is there and has changed the mindset. But when it comes to healthcare delivery, you need a professional tool.

1:25Christian Pascual, CEO and co-founder of MedicTor. Thank you for joining me here at Startup Health. No, thank you for allowing me to share where we are. Yeah. So you and I have had the benefit of getting to talk about the MedicTor story over many years. And I remember the first time I heard about this idea of virtual triage, this idea of helping people know where to go for care. A lot of people, they get hurt or they get sick and they immediately just sort of start casting about for the nearest emergency department or they just don't know what to do. And you created this initial tool that has evolved over many years, but really helps people.

2:06I remember you telling me, find the right level of care at the right time, the right place. And it's something that was needed a decade ago and it's still needed now. The technology has really changed. So I wanted to call you up and talk to you about how Medictor has evolved, kind of where triage tools sit today in 2026, the role of AI in that process. So why don't you just start by giving me a kind of an overview of what the Medictor product looks like now in 2026. Of course, we have evolved a lot, but I think that the core technology sits in the same place. We just add refinement and more refinement to try to be more easy to use and more safe and also more accurate.

2:52So that has been the evolution. But at the end, as you said, we are solving a global problem, which is the first step in health care. When people start to feel symptoms and decide where to go, typically nobody trains you on what services are available. Nobody tells you how to decide where to go. And people just guess. or Google or touch GPT. So I think that step, that very first step, is extremely important to not only for safety, but also convenience and efficiency. So what we do is we solve this problem. We assess symptoms, we guide people to the most appropriate next step, and that drives a lot of value and also convenience to the user.

3:40How has your technology shifted? I'm guessing that originally it was more of a rules-based system than maybe it is now as we've gotten into this world of LLMs and layering on conversational intelligence. So tell us about how that evolution has changed the product. So we have never been rule-based. We have always let AI decide where to launch. I mean, our game is what is the next best question to ask you. Our mindset is how we use technology to make this question as a physician would. That's the objective. Since the beginning, we did one thing that is to allow natural language processing. We allowed users to express how they want it.

4:28From there, I think that we refined a lot the technology through the years, and maybe the last evolution has been to add an LLM layer to all these technologies. So we combine both worlds. We allow natural conversation at the same time that we provide the same clinical safety that we had with the non-LLM technology. But I have to say that it's a client choice to use LLM or not. that has, of course, some convenience on one side, which is as you were talking to somebody, but also people like buttons. And for easy to use to get these buttons away and need to ride, some people don't find it so much, let's say, easy.

5:22So, well, there's ups and downs on the LLM. And if the client uses it or not, it's a choice. Okay. Okay. Talk to us about who your clients are, who you're working with, insurance companies, employers, kind of what does the market look like for Medictor? So we are present in more than 30 countries. Mainly we work with insurance companies. The typical big names in the industry has been clients for a long time now. We also target private hospital chains. We have also big names as clients. And then we have a solution that we We have been working with pharma. That's a very specific use case. And finally, we are working with governments now.

6:07And this is a side of the business that now we see a huge opportunity. And we are also working well on that and chasing those opportunities. Tell us more about why the government contracts are such a big opportunity. We talked to a lot of digital health companies that are moving more consumer focused. So why start at the top? Why go to governments? Well, because we have been in the market for a long time. We have the reputation, the clinical validation. We check all the boxes. So at the time that you need or that you engage a government, first, what has changed is that now I see interesting governments to solve, let's say, countrywide problems on the healthcare space.

6:55And that is, I see, increasing interest on that side. So these, of course, when they look for solutions of our kind, the shortlist, there's not much technology in the world that has all the requirements. So now we are on that point, right? We are a solution that can provide real value with all the safety that you need and solve a government-wide problem that now governments see as a problem, which maybe some years ago they did not even thought about solving that. So it's starting to happen. Take off your Medictor hat for a moment and help us understand the moment that we're at in terms of consumer adoption of AI tools.

7:44You started Medictor and started doing natural language processing back when that was a differentiator. And now people have become so attuned to ChatGPT and Claude and Gemini that they expect to be able to express themselves in natural language. Having been there prior and seeing the current moment, how do you think about what's happening in healthcare? maybe give us a glimpse of where you think that's heading in the next couple of years. So of course, ZGPT launch was a huge disruption in technology. I would say that prior to that was iPhone launching. I mean, to put it into perspective, so it has been really an earthquake to everything.

8:30And of course, I mean, people have been getting used to a more natural interaction with technology. But I think that more and more people is getting more conscious of the, let's say, limits of this technology, especially when it comes to some things. And these things, of course, are one is healthcare and can be potentially extremely dangerous. And I think this has been In public cases, there has been people, of course, learning on the bad way that this is the thing. So when it comes to healthcare delivery, and we are talking about healthcare companies providing healthcare delivery, you need a professional tool.

9:18You cannot use a general purpose tool, and these tools need to be integrated in the flow of healthcare. So this is a very niche way of approaching this, not a general way. And for that, of course, you can blend LLM technology, but you need to put in place all the safeguards for healthcare. So I think that all big companies have realized this. I'm talking about big companies in healthcare, and they look for professional tools that maybe they can embed LLM or not, whatever. However, I mean, I don't think that LLM is a need. I think LLM is an option that comes also with cost and cost efficiency in some places of the world.

10:04It's extremely important also. And of course, it needs to be in this healthcare delivery totally integrated. So I really believe that, yes, it's a huge disruption. It's something that is there and has changed the mindset. But when it comes to healthcare delivery, you need a professional tool. Part of our audience are founders of early stage startups with teams under 10 people, and they are trying to figure out their go-to-market strategy. You have figured out how to be this professional tool that can approach government at that sort of level of enterprise security and robustness. What's your advice to another founder who would like to consider that go-to-market strategy?

10:51Maybe they just haven't even considered it before. What are some of the things that you've done in order to find the right champions, in order to find the right inroads? What should a founder be thinking about? No, first of all, I think that there's a lot of opportunities right now because the tools that are available for developers right now are really game-changing from years ago. You can build just putting things together. And that's something that changes also the framework. Do you mean the ability to scale your engineering team? No, tools that are available. I mean, voice recognition. Or, I mean, I'm talking about tools that maybe 10 years ago were not in the market.

11:33And then you can plug pieces together and build a solution that was not possible 10 years ago to build up so easily, so scalable. So now I think that there are a lot of pieces of technology that you can put together and build out something interesting for the market. But once you do it, I think that a must is to build a trust. I think that's the most difficult thing in this market, to build a needed trust for the client to buy your technology. You need to prove it's safe, prove it works, and also comply with other regulations, which in some parts of the world are starting to be really complex.

12:19How were you able to do that while still having a startup team? I'm guessing you don't have an enormous team, so you've had to be able to do a lot with a leaner team, as every startup does. So how were you able to build trust with the government, stay compliant with new regulations? What were some of your strategies? It's a huge challenge. We were lucky because we had to comply with things one after the other, not all together, because at that time regulation was raising. So we were lucky in the sense that we went almost one by one with the challenges. Now it's more challenging because you need to comply with more things at the stop.

12:59So I think more challenging times, depending on what you do, you will need to comply with more or less regulation. That's also depending on how critical is the solution in terms of safety that you will offer. But yes, it's more challenging. But again, now, yes, I think all the LLMs will help you in preparing documentation, in assessing what you need to comply. That's an advantage because before you needed an expert at least to understand what pieces you need to put together. And now any startup team can chat with cloud and it will picture pretty well at least what you need to do. But you will need an expert.

13:45Okay. You mentioned the idea that your team can use Claude or another LLM to do this research that used to require a specialist. What are some of the other tools that you and your team are experimenting with and using and having success with? Since you are yourselves an AI company, internally, what are you using to be more efficient? Well, internally, we are working a lot with Claude, though. A lot. And we are linking that to our HubSpot, to Redmine, that is our repository of all the development of the company. And also building up projects inside cloud where we have all our documents. So preparing a new offering and tailoring the solution to my client and even preparing a return on investment for them, it's easy.

14:39And that for me, it's a game changer because the easiness to prepare a really high level offering for a client has really been enormously helpful. Whereas before it took me maybe one, two days to put it together. Now it's half an hour. Has AI, have these AI tools been a bit of an equalizer in terms of the global markets? established this hub for digital health in Barcelona and in Spain. How do you see these tools being deployed at the Barcelona Health Hub and being that equalizer? I don't think that's, I mean, because this is available for everybody. I don't think there's a clear impact in different regions because of that.

15:28The opposite. I think that, yes, well, or maybe it's how you frame it. Everybody has more access to the same information if they look for it. But again, when we are talking about the hub, we put into the equation another thing that it's personal interactions still matter a lot. And I travel the places that I'm working with governments to see the people, to meet them in person. Yeah. And this is still important. It's like flawed could make you any number of lists of contacts at that government. And until you fly there and you meet them and you have coffee with them, nothing's going to happen. It's impossible.

16:17I think it's totally impossible that ever happen if you don't fly them and meet them. Yeah. And in many ways, the access to that data is making that in-person even more valuable because everyone has access to the data. So it becomes almost meaningless to have someone's number or their email address. It just doesn't matter anymore. No, absolutely. Any tips that you have just for building trust? Obviously, sharing clinical papers is always important to have the right clinical information. And yet there's also an intangible factor. There's also like X factors in terms of actually building business relationships that last.

16:56Any strategies that you can share? Well, there are many. I mean, of course, the obvious one are seals of excellence, like ESOs or other compliance that you can get the seal. That's a trust element. Of course, clinical validation, any public scientific study that you can publish will provide trust. That for the medical side of the equation is what really matters. whether you have published scientific accuracy or whatever. And then, of course, the use cases with reference clients. I mean, if you can publish a use case together with a big company that really signs the return of investment, the value, the KPIs, that's a huge asset for the next client.

17:45So it's to keep on building this as you can, because it's not easy. It takes time. It takes years actually. Okay. Last question. What should we be looking forward to over the next like 12 months from Medicdoor? Any big announcements? Well, yeah, we really hope to celebrate a government-wide agreement. That would be, that's our target. That's our target. Okay. Working on that. Awesome. On the works. Well, this is important work that it was important when you started. And, you know, sort of the way healthcare goes, it's every bit as important today as it was when you got started. So it's a slow moving train.

18:23And yet you seem to be in the right place at this critical juncture. So, Christian, thank you for giving me the update. And we wish you all the best. Thank you, Logan. A pleasure. Take care. Thank you. deals. And finally, an invitation to all of the extraordinary health transformers, whether you are a startup, someone building the future of health, or whether you are an investor, funder, or buyer of the solutions out there in the marketplace, this is an invitation to you to join the Startup Health member community. Learn more at startuphhealth.com. Look forward to hopefully seeing you soon. Thank you.