Unveiling the AI Boom: Insights from Marina Cortes, Venture Capitalist

0:00All right, welcome to the podcast today. Today we have the honor of interviewing Marina Cortez. She is a venture capitalist based out of Toronto. She went to school in South Korea. So thank you so much for coming on the show today and being here with us. And would you mind telling the audience a little bit about your background and what kind of got you into venture capital to begin with? Yeah, absolutely. Well, thank you for having me on the show. My name is Marina Cortez. I'm an alumni from Seoul National University in South Korea. I studied in Venice over there. And what got me into venture capital was simply my law for startups.

0:36I started with pre-commercialized startups and eventually I made my way on the investment side. I worked on both consulting on the buyer and sales side and also stayed in pure venture capital. I worked for a different firm throughout my career. I worked as an AI investor as well. And currently I'm an investor with intrinsic venture capital in Calgary, Western Canada. Very cool. Cool. So you said like, you know, your love for startups kind of brought you into this. I'm super curious, though, right? Because if you studied medicine, like what did that look like? Did you have experience with a startup at the beginning or like how did you kind of make that pivot or find that?

1:15I don't know, that interest in startups. So what really fascinated me was the way that, you know, research can turn into a larger project and how do you make impact into human health? And I really wanted to have my biotech startup when I was younger. And eventually I ended up on the investor side. And I worked with pre-commercialized startups with professors that are trying to spin out their research out of the lab. And how do I capitalize this high-way scale up? And so for me, that really draws me into investment. And then I realized that when you're on the investor side, you also tend to make a bigger impact.

1:51Because then you oversee a portfolio of startups. So I thought this is quite interesting to have a greater impact on a greater scale. Very cool. Yeah. And like you were mentioning, right now, there's a lot of different changes happening in venture capital. And of course, one of the biggest ships I think I'm seeing today is just the interest in AI in general. I guess the last couple of years, venture capital had a bit of a rough spot in funding new companies and whatnot. What are you seeing as far as, I guess, a general landscape on venture capital and AI? And what does that look like today? The venture capital landscape currently is being affected by the economy and the AI landscape, which is affecting founders eventually down the road.

2:41Currently, there are a lot of deals that are really, really interesting on the tech side. You have tech that can help the elderly. You have tech that can be helpful for fintech payments. All of that is going to revolutionize the way that we function and the way that you work on your day-to-day life. But it's also going to help on the professional side. Whenever you go to your doctors, the doctors will spend less time working on the paperwork because they are going to have some help with AI tools, for example, and spend more time with the patient. Now, the problem with venture capital is that we do face an economic downturn at the moment.

3:18And so it's very difficult to deploy capital. And there are many levels that are upstream from us, from our limited partners, as well with the economic outlook right now that it's not really bright. So this is making it very difficult for founders to raise that. Right. Yeah, for sure. Sure. And, you know, I've talked to a bunch of founders that, you know, I'm friends with in my within my circle and whatnot. And I am hearing, I guess, a bit of a silver lining on the fact that, you know, some of my friends who recently raised just for regular startups, of course, really big struggle. And then they said, you know, as they are other ones that have started shifting and starting AI startups, there's a lot more perhaps interest.

4:00I'm not sure what the rate of closing is necessarily, but there is a lot of interest there. From your end, when you're kind of looking at some of these AI startups, there seems to be a lot right now that are coming across your table and those kind of deals. What factors are you looking for or are you considering to tell how viable one of these AI startups is? Yeah, that's a very good question. So when you invest in AI, you tend to be on the responsible investor side. I mean, I know that a lot of investors are being responsible, but what I look at is how responsible is the tech? And this is something that's very dear to me.

4:35I'm very much into investing in founders that don't brand their company as AI where they are boring AI, for example. I want to see really AI that's part of the core technology. And I also want to see AI that is explainable AI. Do they understand fully the model that they have built? Do they have a thorough understanding of the results that is being outfitted by the model? Also on the data set side, this is something that's very, very important on the health side because I do invest in health quite extensively. And so when I look at AI in health, I want to make sure that the data set will train across a number of patients that is representing the whole population, not just into one gender or not just across one ethnicity.

5:19I want to make sure that the whole data is representing the full scope of the technology. If this is targeting, for example, diagnostic and such, because it's well known that you will have different reactions based on certain genetic background. And this is very, very complicated on the health side because we do have a lot of factors that are coming up. But, you know, generally the optic here is that I want to make sure that they thought about everything when they give the data to their AI models and that they understand what they're building. Yeah, I think that's so important. I love that because when you look at drugs getting approval in the US and they go through that rigorous process, for example, they have to do their case studies and they have to do their clinical trials and whatnot on a very diverse group of people.

6:13And so I think, like you're saying, with medical in particular, but then with a lot of other areas, if you have these AI models specifically trained on one data set that isn't inclusive of all of that, like, look, like the regulators are going to want it eventually anyway. So I think that's really a forward thinking that you're already thinking about that and making sure that people are incorporating that. And it goes for medical, but it goes for a lot of other areas, too, where you really want a very diverse data set. Otherwise, your AI is going to be, you know, really one sided, and it's not going to be a very flushed out solution.

6:46So that's very cool. What are you seeing like, as far as the areas that you think AI is going to impact the most in the future? Obviously, healthcare, very big, and you're very dialed into that. Are you seeing, what are you seeing there or perhaps any other areas? I think that AI will impact healthcare a lot because currently there, you know, there's definitely a lot of work to do in health and it's very much paper-based. It has, you know, a lot of things to catch up on. But space technology is also very much in line to be impacted a lot by AI. Really? you have a lot of, yeah, there's a lot of data analytics that comes from the satellites and there's a lot of data that is being in touch right now.

7:26And so you need to make sense of this data. And so AI is very helpful for that. I'm very much a fan of space technology. And I think that climate as well as being impacted, you know, the way that we look at climate change, like carbon capture and all that, all those models for carbon capture, it's being powered by AI. And so you want to see a little bit more of what's going to happen there. This is quite interesting because I think this is a very genuine approach to look at carbon capture. And then what do I do with the carbon? Do I transform this? Like, how do I store it? And they're very, very interesting models now where they capture the carbon and then they have a specific model to capture the carbon at very highly efficient rate.

8:11Then they recycle this into something else like byproduct for the cosmetic industry and all that so this is quite cool and we talk about a little bit about planetary health i think this is a buzzword right now that's coming into venture capital and i think that ai will be very helpful in planetary health in general uh you know alongside healthcare and space tech that's super interesting yeah to be honest um you know i'm pretty dialed in with all the new stores and everything that's coming out in ai and i have not i don't believe I've seen a lot of articles or a lot of coverage on the topic of climate AI and also space, anything related to there, right?

8:49Like I see fintech, obviously, or like legal or even medical. Those are really interesting. Have you seen any specific interesting startup concepts or use cases in those areas or perhaps any others that have recently, you know, crossed your desk or, you know, or that have really stood out to you? In fintech specifically? Or, I mean, I'd be really curious about space, but, you know, I guess any others either. Yeah, there's a very cool startup right now. I don't know if I should speak about the name of this startup, but because I don't know if we can promote it. Yeah, yeah, for sure. There's a very cool startup right now in Hawaii.

9:28I think it's a very popular name that's looking at surgical precision detection of objects in this listener. And you probably know about space junk, right? And there are a lot of space junk in Earth. And there can be a potential threat to commercial air travel and to even some of the lunch and all that. And so right now there is the space detection. There is also some model to capture that junk that is outside in outer space. and you can see a model of AI that will kind of detect, you know, where is the space junk and then there's another startup that will come and it's going to be like a giant starfish to capture that space junk and recycle it.

10:14That is cool. That is really cool. I'm a big fan of SpaceX and it's been, you know, currently it's being used for more practical application. I mean, going to Mars and all that is quite interesting, but this is really on the long run and in the future month, what you can see is, you know, how can I provide a model to help, you know, protection, defense, for a wildfire tsunami, for everything that is management of asset as well. So if we can use AI to kind of calculate the amount of asset that we have into a certain geographic location. So whether this is, you know, the amount of crop or the amount of natural resources like copper and all the things that could influence our exchange rates and the value of the money.

11:00So great, quite cool stuff that is happening in space. And then FinTech as well. I really like FinTech. I've invested into a company in FinTech actually that uses AI. Anything for authentication, lack of password, anything that can make your life a bit easier whenever you pay because companies want you to pay quick and fast and you don't want to change your mind. The less friction, the higher likelihood that transaction will probably go through. Exactly. And they want you to pay everywhere. Maybe you're going to be able to buy something from your car you know, Excel. Those are quite cool technology as well that are coming out.

11:35That's very interesting. So recently, it's been in the news a lot. In fact, NVIDIA, I believe, it just kind of blipped past a trillion dollar market cap. I believe it might be a little bit lower than that. When you're kind of looking at the AI space right now, do you feel like, who do you think the big winners are? Is it going to be like the NVIDIA selling the shovels? Or is it like, do you think all of these software startups that are coming out right now, are they really going to be able to capitalize on this? I guess, what's your perspective on kind of the winners in that space? That's a good question.

12:05So, you know, when we had the economic downturn, they really took a very bad return with the Silicon Valley Bank, for example. I thought that Brexit would come up and say Brexit was a startup, you know, a few years ago. And I thought they would come up and take over these loans and, you know, grab a piece of the market and then eventually expand to consumer. That seemed to be like a logical sort of event for me. But then I realized that they didn't think I thought they would. And then they had other problems on the background as well for Silicon Valley Bank to help with the issue. But you never know what's going to happen.

12:45Because if you have a next event, what's going to happen next? What is going to be the next drama? So this is where you see during a period of very difficult time, this is where you see the best startups that are showing up. WhatsApp was created during one of the recession period that we had. So I think that there could be other events that would help the startups and become very popular. Now, the power of NVIDIA, it's the chip that they're selling. I think that they do definitely have a lot of ability to scale. and it really depends on their strategy. But right now, it's a good time for a startup to think a little bit about the next 10 years and see where the opportunity in the market.

13:34Right. Yeah, I think that's so important. I've seen many, many startups that are literally just building on an API to open AI or something, and without much of a strong moat, it's definitely something you want to look at the 10 years down the line versus just like what's the thing you can spin up today. Do you think, so a lot of people have accused, you know, what's happening right now in AI as being sort of like a bubble, right? Where do you see AI compared to perhaps the last round, which was kind of like the Web3 or other kind of phases we've gone through in venture capital and startups? Where do I see this?

14:13Speaking of Web3, I think it's quite interesting because there are a lot of development right now in the metaverse and um it's it's very volatile you know we don't really know if this is going to pick up or not beyond what it is right now but i see that anything that is related to web3 and i think that is related to um nfc and metaverse is going to be powered by the newest generation by gen z is you know the most active buyer right now but then you're going to have gen alpha that's going to be spending money within the next 10 years. And my daughter is a gen alpha. So I wonder when she's going to be 10 years old.

14:53What is she going to buy? What is she going to like? So it's really driven by the consumer at the end. So we need to be open-minded and see what we like. I think that space is probably going to open a lot and healthcare needs to be democratized a little bit more to author, you know, where it's for startups to provide help to your medical practitioner. So we'll definitely, you know, see where this is leading. FinTech is never going to die. I think it's going to be, you know, probably even more powered by Web3 right now. So probably a lot of things happening as well in FinTech and insurance tech as well.

15:33Okay, very cool. So in your experience and from what you're kind of seeing right now, Now, what would you say for some of these, you know, fintech companies or any companies, any startup companies that are coming out right now, particularly in AI, what are some pitfalls that you're seeing some of these startups are making? You know, perhaps from the last run where we saw valuations tank a little bit, what are some pitfalls you're seeing and what are some ways that founders can overcome those? The main problem that I see right now in the deal flow is that founders had overinflated valuation at very small revenue.

16:07and the only advice that I can give to a founder right now is focus on generating cash and focus on generating revenue and this is going to help your valuation grow the way that you want it to because the pandemic inflated the valuation quite a lot and so these founders that came to us with like 35x of the revenue, 45x of the revenue you had in Silicon Valley you had deals with like$1 million revenue and the valuation was like$45 million. How are you supposed to fund this right now knowing that the economy is not here to support you? And from a venture capital perspective, you also have your LPs that are holding on capital calls and you have that dormant capital, but you're not sure if you want to spend it because you don't really know if you're going to have more capital injecting it to your firm.

16:58So how can you fund a startup and take a lot of risk? You don't want to take risk. So that's why a lot of venture capitalists right now are evaluating startups at somewhere between the range of 6 to 9x of the revenue for the valuation. It really depends on the field and the valuation is across the board. Whether this is like medical device or a space tech, if it's capital intensive or not, there are a lot of variables that come into the calculation. But the general rule of thumb right now is 6 to 9x. If you don't have that revenue to show, that calculation, then it's going to be difficult to raise at a higher valuation.

17:35And if you need to have a down route to finance your startup, then consider it. When it comes to AI, right now there is a call by a lot of leaders. I remember that there was a letter that was signed by Elon Musk and a lot of folks as well in AI. And so it's very important to show that you are different. And it's very important to show that you are responsible to what you're making. Seeing the greater picture is great, but it's showing that the market is responding to it and showing that you understand thoroughly the technology that you're building, I think is going to help this decision a little bit faster.

18:17Yeah, no, that makes sense. When you're looking at these startups and particularly some of these AI startups that are coming up right now, What are the factors that are the most important to you in like your decision making process on saying, you know, like this looks like a very viable business moving forward? Is it the team? Is it the tech? Is it what are what are like the key things you're looking at? Well, when it's early stage, the team is always going to be the deal breaker, right? That's going to be very important for us to understand who the CEO is and who the CTO is, especially for an AI startup.

18:47we want to know if the CTO has proper training and proper experience in working with AI model and what is the vision of the CTO, where does he want to take the company with the tech? We're talking about models that are being trained on a certain amount of data set. And the fear right now is that if you leave that model open in the wild, AI will become smarter and smarter. So it will train itself, you know, eventually to, if you give it access to unlimited data sets. So you need to understand thoroughly what that model is doing. And I think that when you talk to the team, you want to know exactly what they think.

19:26And you want to know exactly how they build their team, how they build their tech and their strategy and their vision on the long term. Yeah. You want to see also if they have an understanding of responsible AI and explainable AI. Now the product, the team is probably the biggest part of the division for early stage. Now the product is also, you know, the, where do you want to take the product next? Like how do you, where do you want to take this company in the next five years? How are you going to exit? We want to understand that our goal is to generate return for our own investors. So we want to make sure that on the tech side, the team side, everything is locked, but also on the financial side, are you generating revenue at the moment?

20:05If not, do you have an idea on how are you going to commercialize this? and how will you generate revenue in the future. Now, on the later stage, I think it's a little bit of a different game because it tends to be very capital intensive. Server cost is very, very expensive when you have an AI company. And so the more you grow and the more you are in need of capital. So focusing on the revenue, I think it's going to be very important to show profitability of your company and depend again, you know, your sector, But I think these are the problems. Okay. Yeah, no, that makes a lot of sense. When you're like looking at these, because you talk about the importance, especially with the economy, what it is today, the importance of these companies and making revenue and whatnot.

20:53But at the same time, we kind of have this huge AI wave right now. There's a ton of AI companies that are being built. A lot of these things are new and very revolutionary. When you're looking at some of these AI startups, how do you kind of, I guess, what's the balance you find between like the actual revenue it's producing or perhaps just the disruptive potential that a startup has, right? Like maybe there's this crazy area they're trying to disrupt. There's a lot of potential, but they're just starting and the revenue is not there yet. How do you kind of strike the balance right now, particularly as we're seeing like all of these disruptive AI companies being built?

21:29That's a good question. So actually a lot of that question resides in the fact that your fund has a thesis. And so if your fund does not invest in pre-revenue, pre-commercialized startup, you're locked by that thesis, you know, specifically. But if you do have the ability to invest into both, if you've established your fund as a generalist fund and that you can talk to very early stage founders, pre-since stage or a since stage or either in the early traction pre-revenue or generating some revenue, then definitely you can look at what is the bigger picture you know where do they want to take the company and eventually this is the ceo that we bet on we really put it all the time on getting to know the ceo and his understanding of the strategy does he understand the market does he understand where the need reside you know how do you solve the problem that you're facing and we always tell startup, you know, explain to me what is the problem and how are you going to solve it with your technology?

22:33And we want to make sure that you can commercialize that solution. So you need to have a very good understanding of that equation. And this is going to be probably 80 % of discussion. Then after that, we'll talk about, yeah, the tech and potentially some traction. If you have great traction with corporation, like NVIDIA signed a contract and all that, this is great. but you know how much money is that generating we're also looking at you know how much profitability you can make from this early traction and how much you can keep the light on okay yeah yeah okay that makes sense um and also and just a follow-up on that uh you know as we're seeing like a lot of these disruptive technologies come out you you've spoken a lot to the fact that uh it's important to make sure that these are you know responsible ai and we're really focusing on that.

23:20I think that's super critical. When you look at, and of course, I think, you know, self-governing in a way, whether that's from the startup or from the investors, incredibly important. But there's a third player in this. And I would be curious to hear what your thoughts are on where you see regulation in AI going. If you see, what do you see that relationship looking like with companies and investors in the future? This is a very interesting question. So So regulation will come from the government. And my question is, how much does the government understand about AI? And I think this is not an easy question to answer because even for us who are investing in AI, even for us who have like a technical background and a specialization, a scientific specialization, such as health and AI or a space tech and AI, it's pretty difficult for us to understand where is this going to go?

24:15how far is AI going to go? And what would regulation do? You know, that's also another question that the government would answer. We had, you know, that type of issues in the past with privacy when the privacy law changed in Europe where they installed, you know, all this regulation for privacy and the data has to stay in the country and then you cannot share privacy data, etc. Which is fine. But for AI, it's a much complex, much, much more complex issue. You need to understand how far can you go with that and what can the government do? Is the government going to stop AI on the privacy level?

24:56Is the government going to stop AI on the amount of data that is being fed through training? Are they going to put limitations as to where AI can go? The problem of AI is that the more data you give it and the smarter it gets. And so I think this is going to be very difficult to government to answer, you know, on how do you prevent AI from going nuts? And, you know, you throw a bucket of water on it. You probably saw that OpenAI had a jump opening and they were looking for somebody to stop the server in case and go nuts. Right. It's a physical issue. And again, right now, it's not a software developer issue, but this is going to be very difficult, I think, for a governing body to understand what AI does and to put limitation and regulation on it.

25:53I think the regulation would be very, very high level, if any, such as the data set must stay in the country and some server geographic location, probably some trade as well. The same way that they've implemented the trade programs and the trade agreements for the exchange of goods, probably they might be able to do the same thing for data to protect the asset because data has a value. They can probably think of this type of creative solution. But enforcing responsible AI and explainable AI is going to be a very gray zone that they need to define, for example. So I think investors play a lot of role into showing a leadership here?

26:37How do you build a model that is responsible and that is explainable? Yeah, that makes sense. And I think when we talk about being explainable and having that transparency, I think that's going to be a big thing for consumers and regulators as well. And also corporations incorporating it, right? Like you don't want to use an AI on your backend or for one of your products if it's not transparent and there's biases in the data and you're unable to identify where that's at. So I think that's really critical. Kate, my one other question for you that I'm just super curious your thoughts on. So we're seeing like OpenAI and we're seeing Google right now.

27:12And there's a couple other big ones, right? We have these people with these huge data sets that are training and they've grabbed this data. Now they're going to Congress. Like we saw Sam Altman, CEO of OpenAI, say, hey, we need to regulate this more. And I'm wondering what your perspective is on will we see any more of these big AI models like OpenAI and Google? Or are we seeing the end of that with, you know, them going and saying we should regulate this? And in addition, you know, previously OpenAI, they were able to like, for example, they grabbed all of Twitter and all of Reddit and incorporate into the model.