E72: Managing Healthcare AI with Mara Lederman

This podcast is brought to you by Ginsler Wealth. The information discussed today is meant to be general in nature and not tailored to your specific investing circumstances. This means that you should talk to your regular Ginsler Wealth advisor or your own advisor before acting on any information we discuss today. Nothing in this podcast is intended to be a solicitation. And past performance does not guarantee future returns. Some Canadians get excited about the prospect of changes in healthcare delivery, while some get upset. This episode is for the former group, but should be most listened to by the latter group. Today on The Unlimited Podcast, we speak with Mara Lederman, former U of T professor and now co-founder of Signal One, revolutionizing the management of AI in healthcare. We talk about her journey from being my grade 8 classmate to MIT, professorship, the leap into entrepreneurship, and what the future of healthcare delivery should look like. Let's go. Before we get to my conversation with Mara Lederman, there's a famous entrepreneurial story worth knowing. Shopify didn't actually start out as Shopify. Its founders were trying to build an online store to sell snowboards, but the e-commerce software available at the time wasn't good enough, so they built their own. And somewhere along the way, they realized that the technology they had built to solve their own problem was actually a much bigger opportunity than the snowboard business itself. So they pivoted. They stopped focusing on selling snowboards and started selling the platform they'd created to help other businesses sell online. That platform, of course, became Shopify. It's a classic business school example of listening to what the market is telling you and recognizing that sometimes the most valuable thing you've created isn't the thing you originally set out to build. Mara Lederman used to teach that very case at Rotman. Years later, as co-founder of Signal One, she found herself living a remarkably similar story. Well, hey everybody, welcome back to the Unlimited Podcast. A little bit of a summer break, maybe, uh, 5 or 6 weeks since the last episode. But I'm very happy to be back now, uh, for you with a little bit of a, I think what's going to be a health and technology theme for a couple episodes, which is great. As my listeners know, I often start off by saying, oh, I'm very excited today. We have a special guest or a guest that I'm very excited about. And today is no exception. So I'm going to say that again. And that's because today's guest is one of my oldest and dearest friends back from Associated Hebrew Day School in Toronto. This is Mara Lederman, who is with us today. So Mara, thank you so much for being here.

Hi, Brian. Thanks for having me.

Nice to see you. I see you. The people don't see you, but I see you. So as I said, Mara and I went to day school together. Mara was without debate the smartest person in our class. Sorry if I'm embarrassing you, but you know, that was just the case. Mara has gone on since that time to earn a PhD in economics from MIT. No surprise. Spent more than 2 decades at the University of Toronto's Rotman School of Management, where she taught strategy. and ultimately helped lead Creative Destructive Lab in Toronto, Destruction Lab, sorry, in Toronto. And we're going to talk about that, where she worked with some of the most promising early-stage technology and AI companies. And then she did something amazing. After many years as a professor, she became an entrepreneur. And with Canadian AI leader Tomi Putonen, who you may have heard me mention in the past on this podcast, because we've interviewed his partner, Jordan Jacobs. twice. She founded a company called Signal One, which is helping health systems figure out how to use AI at scale. And we're going to talk about that as well. So there's a little brief intro to Mara and our history. And again, Mara, just really happy and excited to chat with you.

Thank you. And thanks for the generous introduction. Uh, I'm excited also.

Okay. Amazing. So again, not to beat the dead horse, but yes, you know, 14 years old, we're in class together. Uh, we're graduating maybe in grade 8. I didn't know what the heck on earth I was going to do at that time. Did you have any idea what you were going to do in grade 8? Be a professor, be an entrepreneur, like, or maybe the better question is, give us your path or tell me the path. Catch me up, I guess.

Yeah.

Yeah.

So what's happened since we last saw each other in grade 8? And to be fair, we've seen each other a bunch of times.

That's true.

But we haven't really caught up like this. No, I had no idea in grade 8 what I wanted to do when I grow up. And sometimes I feel like I still don't know, you know, what I want to do. I'm still waiting.

I'm still trying to grow up. My wife tells me so.

We're still trying to figure it out. And, you know, we've both just, you know, raised kids through high school. They're in university. I watched them, as I'm sure you did, start thinking in high school, what am I going to do? What courses do I take? Everyone else is figuring out their courses. There's so much pressure now to figure things out early. I think for me, it wasn't until high school when I started to figure out what I was interested in. I loved economics even in high school. I loved it even more when I did it as an undergrad, but it was certainly one of my favorites. I liked math. I was interested in business. I remember I went to Boston when I was in high school and my brother was applying for med school or residencies, and we went to look at Harvard and I got to go with, and I bought a Harvard Business School t-shirt. And at the time I was like, maybe I'll go to business school. And then, then I went to Western. I fell in love, truly fell in love with economics. Uh, I don't know that there's that many people who say they love what they study as an undergrad, but to me it was and still is such a sensible way to make sense of the world. And then, you know, I thought I might go to law school, I might go to business school. And then one of my professors said, why don't you go get a PhD in economics? And I hadn't even fathomed that path. I wasn't sure I wanted to be an academic, but I knew that I loved economics. I knew I wasn't ready at 21 to go have quote unquote a real job. Now all our kids want real jobs. I didn't want a real job at 21. So I started looking into these PhD programs in economics, uh, and I visited some of them and I was super interested. Right. In what they were teaching, but also in where people could take jobs afterwards. And this was, you know, I guess 25 years ago now, you know, the door for what PhDs in economics are doing is open to, you know, opportunities we didn't even fathom. Maybe we talk about that. But I took a bit of a leap at 22 years old and said, I'm going to move to Boston. I'm going to go do a PhD. I'm going to do it in an amazing place and we'll see what happens. And then the rest, you know, we'll get to, but that took me, of course, to academia for the first part of my career.

Right. So speaking of Harvard Business School, I went there and I loved it, but I often say I don't really want to do this because, you know, too much work, like way too much work. But I often say I really should actually go back and do the MBA now because what I learned, you know, and actually interestingly, just for folks to know, it's September 11th today as we're recording this podcast.

Yeah.

I was in my first week of school on September 11th, and I will never forget Walking out of class that morning and seeing my classmates and everyone distraught. So sorry, sorry to bring it down for a minute, but I think it's, it's, you know, sort of important to say. But, you know, I often say I would like to go back and do an MBA again now because the world has changed like crazy. There was no Facebook. There was— forget about AI. There was no Facebook. Zuckerberg was still maybe an undergrad at the time. Anyways, So I'm curious, like as a professor, what did you see for the last 20-some-odd years? What did you see as sort of evolved and how did you teach, you know, as the world was changing? Maybe a weird question, but you had a front row seat.

Yeah, no, it's an interesting question. Um, and I think to be honest, the pace of change probably has been fastest in the 4 years since I've left academia to run the company. So I can only imagine what's going on now. I would say I had 2 front row seats to this. So when I was at MIT, which was '98 to 2003, I was there during the first dot-com run-up. This was the early days of internet businesses. This is when we first heard of the term e-commerce, when people were trying to figure out like, I can buy a book on the internet, can I buy a piano? And like, where is gonna be the dividing line? And I would say what was so fascinating to me at the time, I was doing my PhD and a professor in the MIT Economics Department created what was at the time the very first course on the economics of the internet for MIT econ undergrads, and I got to be his TA. I mean, I wasn't yet thinking about what it would be like to be a professor, but a little bit I was. And I was like, how do you teach a course on the economics of the internet? Because nobody's written anything on the economics of the internet because we're figuring it all out. And what he did and what I continue to see academia, at least in economics, do with successive technology waves is basically figure out where fundamental economics helps us understand these phenomenon and where it doesn't. And I'll just again, give you an example. The big thing with the internet and, you know, when Amazon came out and all of these like price comparison sites was market transparency on prices. Like that was one of the first things that changed. Like if you wanted to get 10 price quotes for a book before, you had to go to 10 stores. There was literally no other way to do it. And now with the push of a button, you could get, you know, for a book or a CD or some very well-defined product.

CD. What's a CD?

Remember those? Remember a CD?

Of course.

And so he taught a whole segment to these undergrads, 19, 20-year-olds, on what does economics tell us happens to markets when the cost of search and transparency falls dramatically? And so that's basically when we started teaching about AI and my colleagues at Rotman have some of, I think, the most profound and compelling books on the economics of AI. We framed it the same way, a drop in the cost of prediction. When we built CDL around different technology areas, blockchain and quantum, we would often do the same thing. Like fundamentally, what is this technology changing? Where can we draw on traditional economics to help us? And where do those economics fall apart? And so it's those 2 things, like leveraging what continues to hold and then redefining the playbook that I think is most interesting. I think the other places we're starting to see it is on, on the management side. And how do we start teaching MBAs more than undergrads now, people who are gonna go into a very different workforce, you know, the management skills they need given the, the growing role of technology. And, you know, I think it's, it's a hard thing to do for the business schools be just because of the nature of curriculum design. They can't move as quickly as, as everybody would like them to. And so they do sometimes follow with a lag.

I can't imagine how quickly they'd have to keep up to be on, you know.

And then never mind with the innovation and learning technologies, right? And, and how, how they deal with that.

I guess, uh, that, that's for them to figure out now that you're not there. So you mentioned, uh, CDL or Creative Destruction Lab, and I know you were extremely involved there. What is CDL? Uh, what were you doing there with these founders? Just to give some color, 'cause I think it's an amazing, I don't know if you call it an incubator or a mentorship program or something, but it's like, this is what we need in Canada is, you know, experts to help launch, you know, the next wave of businesses and entrepreneurs.

Yeah. So I think CDA, I think we'd call it an accelerator or a mentorship program. It's not really an incubator cuz it doesn't quote unquote like incubate companies the way we often think of incubators as doing, but it is a program that is targeted at early stage deep technology companies. many of which have or are in the process of spinning out of university. So truly novel technology and helping them in that early phase, really, I would call it sort of seed stage, right? Transitioning from what we would affectionately say was like a science project into a, you know, an investable business. And the program was started by my colleague Ajay Agarwal back in 2012, initially sort of focused around Canadian technology companies, trying to kind of close the gap between great research and limited commercialization. It was right around the time of some of the biggest breakthroughs in deep learning that were coming out of the computer science department at the University of Toronto. So it was natural to have a program focused on technology and then increasingly focused on AI at U of T, though by no means was it exclusive to, to U of T-based companies. Over time, the program has scaled massively, and I think, you know, it should be one of our amazing Canadian success stories when we think of how hard it is for Canadian organizations to scale, it has one, it is one that has scaled. So it initially scaled across universities in Canada, scaled to the US, scaled to Europe. I think it's now in Australia. I'm probably missing, you know, a couple of the sites, but it is really meant to, you know, create what we used to call a market almost for mentorship where we'd bring together mentors who were experienced operators who had built, scaled, then usually exited technology companies. With really promising young science-based companies and run them through a 9-month highly structured program that was designed to give them sort of focus and guidance to get them through those like early days where you have to do everything and yet you don't have the resources to do everything. And if you don't prioritize right, you run outta money before you kind of accomplish what you need to accomplish. Toronto in, in its fairly early days, freeing up time for others to go scale it outside. The Toronto site.

And what was your role there specifically?

So I was the site lead. So every CDL lives in a university and will typically have a faculty member or someone else who oversees it and provides really this sort of, I would say, like the academic and the intellectual leadership. There's also, you know, an operating team who operates the program. But as we think about, you know, what are the new areas we want to build programs in, how do we continue to kind of keep it current, keep it forward-looking. Where do we bring program innovation? I did all of that. And then also kind of build deep ties to the rest of the business school and university. It was great fun. I loved it.

Sounds amazing. Um, I guess that's where, I mean, I'm trying to put the pieces together. Is that where you and Tomi, your now partner, is that where you met?

That's where Tomi and I met. Yeah. And that was, you know, Brian, quite frankly, even my research before wasn't around AI or innovation or technology, right? I studied branch of economics that we call industrial organization. Like, I mean, you know, how do companies compete and how do markets work and when do they not? And I did a lot of antitrust economics and antitrust consulting. And then I got into Creative Destruction Lab and that's where I really had my eyes open to this sort of exciting area of, you know, tech-based entrepreneurship. And that's where I met Tomi. So Tomi was, when it started, it had 6 mentors. Like now there's thousands, I don't know the number. And Tomi was one of the 6, having been one of the few people in Canada who had built multiple AI companies and exited them. Also, you know, as many of your listeners will know, you know, he and some of his former, his partners in other endeavors really have been kind of like center points of building out the innovation ecosystem in Toronto with the Vector Institute and, you know, venture and the connecting to the university. And he was a mentor and we kind of became friends well before we ever became, you know, co-founders and, you know, the people each other talks to more than they talk to anybody else anymore. So.

Amazing. And was the, because, you know, your company now, Signal One, is obviously focused on, uh, well, obvious, will be obvious when listeners learn more if they don't know already, on the health side. Was there a big deep health focus at CDL that you and Tomi were both working on? Or how did the connection that the opportunity in health was, was there?

That's actually exactly what happened. So CDL successively kind of specialized, and the first thing it specialized in was, you know, programming around AI. And after a few years of that, we started to see what were all the application areas that these AI companies were trying to apply AI to. And health, of course, and not surprisingly, kept coming up. And so what we did, you know, as this grew, is we introduced specialized AI streams in particular verticals, health being one of the very first and probably now the largest. And so Tomi and I worked together on that. We introduced those health programs, I think, in my first year leading the site. I think at the time he had actually brought some funding from TD Bank to kick it off the ground. He was one of the anchor mentors. And he just had, you know, I'd known this all along, just sort of a bit of a passion for AI in healthcare with the belief that that's where the greatest, like many, that that's where the greatest opportunity would be. I was sort of just, at the time, kind of curious about it. My own interest in applying AI to healthcare, I think, emerged a little bit later. And partly, I think both of us, we were just impacted by what we saw during the pandemic when we saw our health system really crumble. and recognize that the sort of sustainability problems we face in healthcare, like, we're not gonna build or buy or hire our way out of them. And I think for both of us, that crystallized, you know, the need and the timing being right to try to do something in that space.

Right.

So we work quite closely with these health streams. We probably saw a few hundred startups over our years there in the health AI space.

Okay, so amazing. So you've led right into the, okay, all of a sudden you and Tomi, or you in particular, I guess, 'cause we're focused on you, Realize there might be an opportunity to move from a very, uh, I don't, I wouldn't say comfortable, 'cause I know you, you're probably working your butt off and all that stuff, but a comfortable, secure role, you know, professorship. And now you're trying to decide, oh my God, should I just leave it all and start a business? Can you walk us through that?

Yeah. Let me, let me—

Those days, weeks, or months. I don't know.

Yeah. So let me give you a little color because again, it all happened during the pandemic. So anything that happened in the pandemic didn't play out the way you will imagine it played out. So we are, I don't know, a year and a half in, you know, it all blends together. Everybody is working from home. You probably like me, you're working twice as much as you ever worked before because there's nothing else to do. The kids can't go to any programs. Everyone's locked up with their children and losing their minds, right? At some point. And out of the blue, I get an email from Tommy. I hadn't seen him since before the pandemic, I guess. You know, how are you? Let's catch up. I was like, that's weird. People always want something, right? When they want to catch up out of the blue, you know, so jump on a Zoom, um, and we catch up a bunch. How's everybody? What are you doing? You know, we do, you know, regular catch up. And then he said, he's like, you know, it's time to really start a health AI company. And he's like, you know, I've always been interested in this. He had been finishing up his his time at TD Bank as Chief AI Officer. He goes, I wanna talk to you about it. I think it's something you should think about doing with me. And I was like, you mean on the side of my desk, like all the other things I do in addition to my day job? And he's like, no, Mari, you don't build companies on the side of your desk.

Right, correct.

This is, this will be the only, if you wanna do this, you know, this is the only thing you'll do. And so we talked a little. I said, I gotta think about it. Life is crazy. I said, let's get together. And he's like, well, we can't get together. We're not allowed. We're not allowed to, no one's allowed to see anyone. And the funny part is his wife is an infectious disease doctor. He's like, I'm really not allowed to see anyone. Like we take COVID really seriously in our house.

Right.

And I said, well, let's meet outside. It turned out I was taking my daughter to a skateboard park, um, 'cause she wanted to skateboard. He goes, there's a skateboard park right near my house. I'll meet you there. We'll sit on a bench 6 feet apart and we'll talk about it.

We won't look, we won't look at each other.

Well, you know, there is, you know, and we met at the skateboard park in the beaches and listeners who know that can picture it exactly. And my daughter, who was probably like 12 at the time, is like trying not to kill herself skateboarding.

Yeah.

Um, and that's where Tommy and I really started talking seriously about, you know, should we do this together? What would the company look like? I'll be clear, the company today looks totally different from what we talked about.

Yeah.

On the park bench. And anyone who's built a startup You know, wouldn't believe me if I said it looked like what it, you know, what we first thought. Um, and, and that's when he put it to me. He goes, but this is a big decision, obviously. And I said, I need to think about it. I mean, already the wheels were turning though. I was so excited about the possibility of doing it. I think many people came outta COVID rethinking careers, right? Because you, when all you do is sit at home and do your job and you don't have the joy of being with your colleagues in person and the joy of kind of spontaneous interactions in your workplace, it makes you think like, is this the thing I, I, I wanna do? And, and I loved my academic career. I mean, I loved it. No one who met me would've said, oh, Mara's gonna leave. But I was ready for more. I knew I'd started super young. I was 27 when I started as a professor, younger than my MBA students. I'd done it for 20 years, and I knew that there was something else I needed to do.

Yeah.

In the second part of my career. And so, It didn't take long to actually tell Tomy I'm in.

Right. Good for you. Yeah. Yeah. I, uh, my firm was incorporated on February 18th, 2021. So, uh, similar situation, but what you're doing is maybe much more exciting. That's why you're being interviewed. I wanted to learn and hear about Signal One, but maybe just on that topic of, you know, becoming an entrepreneur, like you were mentoring entrepreneurs for 10 years or so, CDL with Tomy, with others. What was like the biggest, I don't know, surprise or something that jumped out at you when you now went on that side of the table and was an entrepreneur as well? Anything that like, uh, interesting?

Yeah, people ask me that. People ask me that all the time. Like, what, you know, from mentoring to doing or from teaching to doing, right?

Yeah, yeah.

'Cause I have to plot the courses around the CDL and, and I think about this a lot as a former academic who's now, you know, operating a company. And I think I'll give you a couple comments. The biggest one is almost everything we thought about at the university, um, and told these companies and taught these founders and these students is right. And yet it is 10x harder to do it than to say it.

Yeah.

Right? And so it's not like we're saying the wrong, like all the things we say, like startups are all about running experiments and the faster you can test things, the fast— yeah, that's true. But when we try to do that, there's a 1,000 real-world practicalities where it's like, I could run this experiment, but I already told the customer I was gonna give them this other thing, you know?

Yes.

And there's 100 things like that where conceptually we really do understand the things that make building a startup hard. And yet the fixes we offer and the suggestions we offer scrape away so much of the complexity And busyness is the reality of running a startup. So that's probably my biggest one is just the doing is so much harder than you realize it is until you've done it. Every day is hard, right? Like every day is hard. Even the good days are hard.

Correct. I was saying to my team, we do, you know, strategy sessions multiple times a year, but you know, just in general, we meet every week. And I said, you know, it's very easy. People will come to me and they say, yeah, I'm doing X or Y and I'm struggling with, something, this and that. And I find it reasonably easy to give advice to somebody else. And I think I know the answer. It's pretty crystal clear and you should do this and that. And then I say to my team, but yet all the key stuff that we have to make decisions on every day, very, very hard, you know? And I think it's basically what you've just said. It's like, there's, when you know the whole complexity, it's much harder to make decisions than when you just get that surface level and can— make recommendations.

That's exactly what it is, right? The real-world compli— the, the theory is great at getting at the heart of the issue, yet, and, and pushing you in the right direction, yet the implementation of the right answer doesn't get to abstract from all the messiness the way the theory does. And that's what makes it so hard, right?

Right. But fun, I guess.

But fun. No, no, no, but fun.

So now maybe just give us, uh, maybe, I don't know if it's As you said, the business has evolved, uh, quite a bit from what you and Tomi, you know, sketched on a napkin or on your discussions. But what is SignalOne? What were you founded to do? What are you doing now? And what was the opportunity, by the way? Because you, I, I could tell the excitement in your voice when you're retelling the story of being on that bench. Like, what was the opportunity that made you go, oh my God, like, we gotta do this?

Okay, so let me tell it that way, 'cause the opportunity is the same. It is just that the part of the opportunity that Signal One is trying to address has evolved over time, right? So the opportunity is the same one many people have talked about. There is probably no other industry that has greater potential to be impacted by AI and greater need, right? And so healthcare systems really are, you know, crumbling to some degree. And it's not just in Canada. We see the same problems in the US. And at the heart of it, I think, not to sound too much like the economist that I am, is just like a fundamental mismatch in demand and supply. People are living longer, which is fantastic. They are living longer with more chronic diseases, more complex conditions. At the same time, a lot of the like sort of social and economic determinants of health are deteriorating for some groups of the population. And so we have overwhelming demand for healthcare, and yet our ability to meet it with the existing sort of supply and supply model, physical hospitals, Inpatient care, in-person care as opposed to virtual, it can't keep up. And so the way that manifests is either people don't get care because people don't get care, people get bad care, people wait too long for care. We see it in Canada as really long, you know, emergency department wait times, really long waits for specialists or for procedures, people not having family doctors. But at the end of the day, it is demand exceeding supply. And in most other industries, one of two things happen. You either expand supply, right? But we can't do that, you know, in a cost-effective way. Or you raise prices to ration demand. And so sometimes I see like, it is the biggest, I think, imbalance in demand and supply since like tickets for a Taylor Swift concert, right? But we know what happens there. Price just ratchets up to make it all work.

I couldn't get tickets. I tried.

Exactly. And so no matter what you do—

My kids still like me a little bit, but not—

You know, not as much as their friends, like their dads who got tickets. But that is the opportunity. The opportunity is, is how do we take healthcare and let it follow the path that many other consumer-facing industries have had, where technology in general, and now AI in particular, is leading to kind of dramatic productivity improvements, cost reductions, and transformation in how the service is delivered? Okay, that's the opportunity, right? Yep. And initially, Signal One, we were gonna go after it by, you know, building AI applications, right? There are so many, you can walk around a hospital and basically point, AI can help there, and AI can help there, and people do that. We're not gonna do that, but you could do that. And we thought we'd be one of the companies who picks off opportunities. My partner, Tom, he's a computer scientist. He had built machine learning companies before. We have an incredible technical team. We had some partnerships with, you know, hospitals here in Toronto. And we thought we would do that. And those applications are having some impact, but not as much, I think, in general across the industry as people had hoped. And here's what happened. We paid attention to the market like a good startup, and we learned a couple things. Number one, there were many sources of applications, right? Not just startups like us, but hospitals could build their own applications. The big technology players like the electronic medical record companies were easily spinning up applications, and they were already embedded within the workflow. which is a huge differentiator in healthcare. And there were a lot of specialized software companies that were also targeting AI in the emergency department, in the surgical suite, in the back office. And so we would be one more application company.

Yeah.

More importantly, what we learned, especially when we started going to the US and we started talking to these hospitals, is they're like, we have tons of applications. We can build them. We can buy them. We've got researchers. We're a massive academic medical center. You know, there's no shortage of that. They're saying we can't pull it all together kind of in an integrated way, in a way that lets us create value from it. And that was actually the same problem that Tomi solved when TD first bought his company. They also had a bunch of AI projects across lines of the business, but they didn't have kind of the underlying technology platform to bring them together, to monitor, to validate, to measure value. And so that's when we pivoted the company, right? And people have, you know, made an analogy to Shopify, which is obviously a generous one and a nice one to have in Canada. And a story I used to teach, you know, at Rotman.

Yeah.

But saying, you know, the applications actually are in big supply. What is scarce is the management infrastructure that allows them to be managed. And I think we made that pivot. And then I think what's happened that's reinforced that pivot Of course, ChatGPT came out. GenAI has exploded, particularly into healthcare. Why? Because healthcare is so language-rich, right? So it's, you know, it's so ripe for, you know, technology that can both read and create, you know, language-based output. Agentic AI has come out. Agentic AI is particularly valuable in labor-intensive industries. Healthcare, you know, labor is like 60 to 70% of operating costs for most large health systems. And so the applications are exploding and the need for the management layer is even greater. And that's what Signal One provides. So that was probably a longer answer now that I think when you're like, just tell us what you do. But I think it's important for everybody to understand the opportunity is massive. The part of it that we're trying to solve is now this enabling layer, uh, which we think is first order.

And yeah, is anyone going to solve in, in healthcare this fax situation still where everything is, many things are still faxed. Because that is a trillion-dollar, that's a trillion-dollar market right there.

I know. Yet the technology to replace fax machines exists. So I think this is one where they'd say it's just an adoption problem, right? Or an integration problem, right? We don't need to invent a new piece of technology to replace faxes. I know.

Oh my God. Unbelievable. So can you give us maybe, um, I mean, I understand now what Signal One is doing. What are you seeing in hospitals that they are using AI tools? I mean, I guess AI tools to do, like, what's an example of how healthcare is being potentially or can be transformed with AI? Just to give our listeners a sense.

Sure.

Because there's a subset of folks in the world who think AI, I was talking to someone this morning, is going to take over the world and kill us all, or who knows what. But the potential benefit to society, I think, is It's humongous. So like, what are you seeing from your side? Because you've got the front row seat. There's probably no one that has a closer seat than you.

I mean, there's so many examples, but let me put them in a few categories. So let's think of some truly like that are in the area of clinical care, right? Probably the area that has taken off the most with AI within clinical care is radiology-based AI applications, right? And this is because computer vision is is well proven out. Computer vision tools are really good at looking at radiology imaging, have been shown for years to actually be better than humans at finding things, categorizing things, and being consistent, right? Every health system we talk to, at least in the US, has at least one and typically multiple AI applications that either read or do a second pass or triage, whether it is X-rays or CT scan results or endoscopies, either to prioritize it for the physician to look at to find things that maybe are secondary to the thing that they're looking for, the reason it was ordered, that enable faster coordination of care when things are time sensitive. Maybe you're on brain bleeds or something. If you're waiting for a surgeon to read it, someone could die waiting in the pile. Do a first pass with AI, pull those to the top, make sure. So there's a whole sort of set of applications that I think have clear ROI and and benefits in that domain. That's one. Another is, you know, let's think of some that are more emerging. Just all of the patient interfacing, the scheduling, reminders, phone calls, pre-procedure instructions. Instead of that being done by humans, a lot of that can be done with voice agents, right?

Right.

Whether outbound from a doctor's office, inbound to a doctor's office or to a call center. Again, not Distinct from what we see in lots of other industries, but it is now coming to healthcare. And it's important. This isn't just a let's replace workers, right? This is saying we can provide a level of touch with patients faster, earlier, more frequently than we couldn't do otherwise. One of my favorite examples I heard at a conference is phone calls to patients who are scheduled for colonoscopy. So a week before, they get a phone call. And it reminds them, don't forget to do your prep, right? Here are all the instructions. Don't forget to stop your medications. These are all the medications. And then what they're looking at is how many fewer people show up for an appointment not ready such that their colonoscopy has to be canceled. And you can't reuse that appointment because nobody else is sitting around just prepping for a colonoscopy in case something, right? And so when, when someone shows up the day of and they didn't stop their blood thinner or they didn't stop their iron or whatever it is, right? Or they didn't prepare properly, that's part Right. And so you have this long queue and yet you have open spots. And so they're just building AI voice agents that can do it. Then they listen, they have another AI that listens and will escalate any call that feels like it needs a nurse to call the next day, will call back. So tons and tons of applications like that. And then tons of applications just kind of in the back office, the same way every company is around finance and billing and, and HR. And then I think there's some bigger areas that we're just starting You know, to think about how do we really improve care delivery? And I can give you an example of one we're doing, we've done, we built with North York General Hospital. We've talked about this a lot recently. It's been in the media. We put a tool into their emergency department, co-developed, really close coordination with the team on the ground there that helps rezone patients when the ED is busy. So very simple principle. EDs work in zones. You're super sick, you go to, you know, the highest acuity zone. You've got a cut on your finger, you go to the lowest, and maybe there's 3 or 4 in the middle, right? And each zone is tailored for like the level of acuity it sees.

Yeah.

But sometimes some zones are super full and other zones aren't, and there's a wait for the full zone. And there's always going to be a set of patients who are safe enough to be seen in multiple zones, right? But it's too much work cognitively for the nurses on the ground to go find them in the waiting room because they're busy doing other things. So we have an algorithm, an AI algorithm that effectively evaluates during peak time all of these patients and will recommend up to 3 zones that they can safely be seen in and sort of pull to the top the zones that have capacity. And it's having improvements on, you know, being seen faster, fewer patients left without being seen, and less stress kind of in the waiting room. So that's the kind of thing.

Amazing. So you mentioned earlier, we were talking about what's the opportunity in healthcare, and a big issue is, you know, the supply chain. and demand mismatch, going back to Economics 101. I can only go Economics 101. I'm not gonna—

But you can go so far with Economics 101. That's the point.

I don't want to ask you any more economics questions because I'm going to look really dumb. So Economics 101, supply and demand. And obviously you mentioned some of the humongous challenges at solving that. What is, you know, as you think about, especially because you're living and breathing this all day, every day, like, what do you think if In a perfect world, if governments and people would just do what's in their best interest or what, you know, you might think is in their best interest, what does healthcare delivery look like 8 years from now with the help of your tool and your peers that are building applications and all that? Like, what's the future look like if we just get this right?

That's a great question. And it's definitely not just a technology problem. Like, we should be clear and we can distinguish between technology taking like existing care delivery models And making them a little bit more efficient, or we can think about truly transforming how we deliver care. That second one, which is the bigger opportunity, requires redesigning how people work, potentially, you know, changing payment models, changing legal and regulatory frameworks. So I want to caveat that. But I've been asked this question before, especially in Canada. And so it's something I've thought about. And if I could put it in one sentence and then we could elaborate, I mean, I don't have the whole plan for 8 years.

Mm-hmm.

But I think if I could give like one piece of advice, we should be thinking about, we should push all care into the lowest cost but safest place to deliver it, right? And, you know, early in starting this company, we spoke to someone at a large US health system, a CMIO who was very thoughtful about innovation. He said, in the future, hospitals should be nothing more than ICUs. That's it. Anything that isn't ICU-level care should be done in something other than an acute care hospital. Because the cost structure of an acute care hospital sort of doesn't justify most of the people who are there for anything other than ICU level. So just, we never say it, but say it over and over again, like, how do we deliver every unit of care in the lowest cost setting that is safe to do it? And that means more virtual care, more urgent care, but not emergency departments, It means more outpatient surgical centers. That doesn't mean do surgeries in a hospital and just release someone right away. You know, outpatient surgery centers, or they call them ambulatory surgical centers in the US, are growing there. We could do it here. We've had some resistance, but it is, you know, more recovery at home. We work with a hospital in Montreal called Jewish General on a hospital-at-home program. They literally discharge people. You know, they're not actually discharged. They're still under the hospital's care, but the last few days are done at home with iPads and traveling nurses. So now think about that. People are at home, they're freeing up a bed. You have nurses who don't wanna do inpatient care, but wanna come back into the labor force. You give them an area around their house. So that is the mantra that I think we should— and then—

That's awesome.

And then you think about where technology enables that from sorting, from communication, from remote patient monitoring, right? You could have all sorts of sensors on people and you could have nurses working outta their homes monitoring the data. And the signals that you're getting from them. So that's what it could look like with less and less care delivered as inpatients in an acute care setting.

I love it. Sounds amazing. Okay, so let's pivot for the last few questions here. As you said, we both dropped off kids this weekend at Queen's Commerce. They're entering their university and eventually the working world. And you've been teaching students for years and now you're an entrepreneur, et cetera. What do we tell these children now? Like what should What should they be learning now? You know, I told my son Ozzy he had one elective because, you know, commerce doesn't give you a lot of electives. I said, just find the one that's AI, that has anything to do with AI and do that because— That's funny.

I gave different advice.

Oh, did you? Okay.

I said, just find the thing that interests you. I said, just pick a course that is different than business that's interesting. And she chose a healthcare, I think a social determinants of healthcare course.

Okay, good.

But I might like your advice more. Maybe if she's sick of hearing about AI.

I don't know. I don't know. I don't know what's better.

We'll see.

So yeah, so what do we tell these students, kids, young adults to focus on, to learn? What would you recommend young students do today to set themselves up for success in this, this world?

So it's a hard question. I, I don't have a lot of the other things we talked about. I've talked about before and I have strong points of view on.

Yeah.

This one I'm trying to figure out like everybody else, and I'm trying to figure it out, you know, as the parent of 2 university-age kids, as someone who hires a lot of young people. And so I'll offer some thoughts and then maybe you and I can, you know, we can play it back and forth and maybe by the end we'll arrive at something, right?

Yeah.

So I still think it's important to learn how to learn, right? And so how you learn is different, but you need to know how do you go from not knowing about a topic or how to solve a problem to know how to do it?

Right?

And it feels like, oh, it's easy to do. I just ask Claude. But the reality is the number of people who tell me, man, I, I was just with friends last night who had to fill in, he was doing an RFP for his business. He's like, I worked on it with Claude and then I read it. I thought it was good, so I left it to the night before and it didn't look good at all. Right?

Yeah.

And so I think number one, there was no, you need to know how to go from almost zero on something to let's say 80% across a wide range of topics. And that can be aided by AI, but it is not gonna be, I think, replaced by AI, particularly 'cause it's very hard if it's a new topic to know if what AI is telling you is right or wrong, right? And you and I were saying this at the beginning, when you run a business, it's all problem solving, right?

Yeah.

Most of those problems are new. And so, we still need people that you can hand a problem to and say, go, you know, FIO, as our kids say, figure it out, right? So, you have to FIO it. and you have to file it to a high level of quality. And so the quality control, like the QA on the learning through AI is becoming a really important skill. And I think they're all gonna get their wrist slapped when they produce things in, let's say, their first job using AI, you know, that is low quality. So that's one thing I think is learn how to learn. I think the biggest thing we still need, we do at work all the time is choose what problems to solve. Right? It may feel like we can solve everything because AI helps us, but we still have to choose which problems do we solve in what order with how much of the company's energy, attention, resources against it. And so that comes down to understanding fundamentals of whatever you're studying, whether you're working in science or in business or, you know, it's still kind of like, what's an important problem to solve and what's an unimportant problem? And then I think you have to learn, like you said to your son, like, go learn AI. And I said this when my older one came home last spring. And he was supposed to travel. The trip got canceled. He's like, what should I do? I said, you need to go learn Claude.

Yeah.

'Cause it could be the way you put on your CV was, I can do Excel and PowerPoint and Adobe Acrobat. No one cares. And he did a whole project with Claude and built a whole tool for law school applications and actually law. So you have to learn how to use it. But what I find is, at least among my team, it's a complement to people's like intelligence. It's not a substitute. So like Claude isn't letting me hire less talented people who use Claude. It's that my most talented people are blowing my mind on how productive they can be when they put AI to work for them.

Yeah.

And that I think is, that's like the complementarity we're gonna see in the market, which means it may be the same, like a, like a lot of the skillset we would've hired for before, we might still be hiring for, 'cause those people are gonna be supercharged by this technology. So I don't know, that's a hypothesis. I don't know how you're seeing it, but I am blown away by the people who work for me, all of whom I hired before Claude came out, let's say, and we bought our enterprise license.

Yeah.

Claude doesn't make me not wanna have them. I was like, finally someone has cloned them for me. They've basically become clones of themselves.

Yeah. And, and we did the exact same thing, Claude Enterprise, just obviously security and compliance and whatnot. And, but I agree with you. I say to the team, you know, Use AI for absolutely everything you could possibly use it for, but just remember, in the best case, AI is going to take it 80% there. The last 20% is the most important, and that's where your brains are really needed to, as you say, not deliver something subpar or help it if it's making mistakes or errors or hallucinating or something like that. Yeah. And I think the first thing you said, and we talked about it before, as you said, problem solving. Every day there's 20 problems to solve. And so, and we all, we can call them problems. We can also call them opportunities, by the way, because—

Sure. Challenges, you know, whatever we want to call them, but new things, right?

New things. And you have to be able to work through it and learn to solve it. So we were, we were leaving our son, uh, just about to drive home to Toronto. And he said to us, you know, uh, yeah, you know, it's my friends are doing this over there and other groups over there. And, you know, we just moved all day and, you know, what, what should I, you know, what should I do? And I said to him, Ozzy, welcome to university. Make the decision. You know, like figure FIO, figure it out. I didn't say FIO, but I'm going to use that because you got to start, you know, well, we've been trying to train this with the kids for a while, but you got to figure out how to solve problems.

Yeah. And what, you know what's still true, Brian? People still trust people. I, I, I, that to me. Like, you know, we've started this company, we're a small Canadian company. We sell to the largest, most prestigious US health systems.

Yeah.

Really, like when we tell people who our customers are, they're like, how did a Canadian company do that? You know, at the end of the day, a lot of it is they say, we just trust you and Tomy. We like you.

Yeah.

We trust you.

Yeah.

We like how you talk to us. You don't lie to us about AI. And so I think continuing to teach our young people to be trustworthy in all the right ways and for all the right reasons, I actually think that that kind of human-to-human trust is, is gonna become more important because everything that is written, nobody knows what's behind it. But when I sit in a room with an executive and I tell them, and I'm gonna be there once a month and this is what it's gonna, they're like, okay, this isn't AI promising. You know, this is one of the founders of this company. Yeah.

Yeah. Good. That's a good, that's a great lesson. Very, very key. Okay. Last question. We're almost outta time here. This is one that we'll ask all our, all our guests. So our theme over here, we have a tagline, it's called Your Wealth Unlimited because we're a wealth management firm. We help families with literally everything that touches them financially. We can look at anything, we can do anything for the families to help them on the financial side. So I always ask the guests, okay, the world is unlimited. You know, we knew each other when we were both in grade 7 and 8 and we, you know, could do anything. You've done your path, you know, you're going through your path, but if you weren't doing what you're doing now, If you could do anything on earth, unlimited opportunities, unlimited options, is there something else you would have done or would be doing?

That's such a hard question, right? People have asked me a similar question, like, do you think about the path not taken? I don't think about that a ton. I've said that. I said, I'm actually incredibly content. I'm content. I'm incredibly blessed with the career I've had and the opportunities that I've had. There's a million other things I would like to do. I don't know that I would trade what I do for any of them. For me, on my career, I've gone from thing to thing 'cause I'm motivated to learn, right? I was a professor, but truly I'm a student and I love learning. And so when I do something and get good at it, I get bored, unfortunately.

Mm-hmm.

And so, you know, for me, the question is like, what would I do differently? What would be the next big learning opportunity? And I'm still pretty early in the learning curve of, I feel like, the current opportunity. But I don't look back and say, Man, if only I had become a neurosurgeon, though I think I could have been a happy neurosurgeon too.

Right.

But there isn't one thing out there that I was like, oh, I wish I was directing on Broadway or being a neurosurgeon. I wish I lived in more places before I settled down with my kid. That is one thing I look back, and if I could give young people, like youngish people, it's hard to move once you have roots and a family. And when you are young and working, there is so much to be gained from living and working in different places and cultures. So I'd be doing that. I'd be living somewhere else, I think, right now. That's the only difference.

Awesome. Well, and on the learning side, you know, I suspect given what you're doing now with Signal One and the world as it's moving and all of the innovations that will come in healthcare that you will, I'm sure, be partly responsible for, I think you'll get your fill, more than your fill of learning there. So I think you've set it up well. So listen, thank you so much. It really was special to do this with you. We spent untold hundreds of hours together growing up. And I always did and continue to hold you in the highest regard. So thanks so much and wish you the greatest success ever, because your success will mean amazing things for the world on the healthcare side. So thanks.

And thank you for having me. This was fun. And I feel honored to be on the list of guests you've had on this show. It's a pretty impressive group. So thanks so much.

Okay. Thanks so much. The Unlimited Podcast is hosted and produced by Brian Ginsler and is edited and mixed by myself, David McMillan. This podcast is not to be taken as investment advice.

If you like what you heard, don't forget to rate the podcast and follow us on social media @GinslerWealth.

Thanks for listening and see you next time.