Ann Miura-Ko Breaks Down How to Build A Venture Firm in 2023

Welcome back to Turpentine VC, a podcast where we discuss the art and science of building successful venture firms, VC to VC. I'm joined by Ann Miura-Ko today, who's a co-founder of Floodgate along with Mike Maples, who you heard on episode 7 of our show. Ann and I cover a bunch of new ground we didn't get to in our Mike episode, including how Ann would go about starting a new differentiated firm in 2023. How Floodgate developed their ideal fund size and portfolio construction, how Floodgate's partners helped their founders achieve product-market fit, and more. If you like what you hear, please do subscribe and leave us a review. We also started a companion newsletter, which sends the top 3 insights of each episode straight to your inbox. We'll link it below in the show notes. Without further ado, here's my interview with Ann. Ann, welcome to the podcast. Floodgate is the first firm for which I've had 2 partners on.

Awesome.

As a first topic, I find you always so thoughtful on macro. Why don't you— we're having this conversation in early September 2023. Why don't you give a bit of overview of kind of where we are in the cycle right now, or how venture capitalists should be thinking about the macro as it relates to LPs, Maybe as it relates to deploying, or at least how you guys are thinking about at Plug8?

First and foremost, it's really important to, I think, understand that when we talk about the macro around the economic environment, so let's say exits, what are interest rates, you know, whether or not IPOs are happening or not, the government, All of those things are noise on some level because when you get started with a company, those are not the things that make or break the business. And you're so far off from an IPO or your exit that generally those situations will change over time. And so we don't really think about the macro from that perspective. What I do think a lot about, however, is how popular it is to be an entrepreneur. And because what that impacts is the risk profile of founders who walk in our door. And to just give you a little bit of perspective on that, when I first started Floodgate, even the act of Starting Floodgate felt like a rebellious act. I had many of my incredible mentors very worried for me from a professional standpoint because it was unheard of to start a venture capital firm. It was also very controversial if you were starting your own business. And so the act of being an entrepreneur itself was rebellious. I had one founder tell me that his Indian relatives assumed that he wasn't able to find a job, which then I, I reflected on and I said, oh, that's funny, 'cause when I started Floodgate, people assumed I couldn't find a job either. And in that environment where there isn't a hero's welcome to the declaration that you are a founder, You need to actually really believe, and it's easy to see when someone really believes. I think in this macro environment where it is still a hero's journey to be an entrepreneur, there's lots of dangers in assuming that whoever has decided to become an entrepreneur really wants to partake in that journey.

That's a great overview. Is it fair to say that this time period is better for incumbents and incumbent venture firms and harder for new firms? Better in that there's going to be, you know, less capital in the system such that, you know, there's, you know, lower prices, you know, more proven entrepreneurs or more dedicated entrepreneurs, and it's going to be harder for new entrants to raise capital for a while?

Oh, I don't know about that. I mean, it's, it's really interesting. When I, um, before my PhD, I actually worked at CRV, uh, and my second day of work was 9/11. And one of my first tasks was to, uh, analyze the market and understand, you know, what was gonna happen. And, uh, I remember one of the predictions I made was that some, some large percentage of the venture firms that we saw out there were gonna go out of business. And I remember thinking to myself, seeing the capital overhang, how much capital had been raised at the time, how much was happening in terms of exits, you know, how much actual investments were happening at the time. I was thinking it's impossible for any of these firms to really sustain this kind of business. There's gonna be a mass exodus. Well, it turns out, you know, I, I then leave CRV to go back to grad school. And I'm in my PhD program. And when I emerge from that program, you know, 4 or 5 years later to take a look at the market, wouldn't you know it, most of these firms are still in business. And the point is the capital is super sticky. And it is, it's not that it suddenly disappears overnight. It's very rare for actually today for funds to decrease their fund size. You'll see it on occasion, but they will decrease their next fund size. They won't actually decrease their current fund size. And so, so I believe that really the only way in which you will see funds no longer in existence is if they can't fundraise. Now, that might happen to some of the smaller firms. It's hard to know how sticky the dollars are until probably 10 years from now.

Yeah.

And so I'm not one of those people who will bet against the venture market at this point.

In the last 5 years, it was a great time to start a new fund, you know, in terms of it was easier to raise capital. And there were people who were less proven or didn't have track records but had great networks and maybe operating track records, and, or maybe not. But it was just, you know, there were a flood of new funds started. Do you think those times are temporarily on hold, Or it's kind of RIP, good, those times. Like a lot of people, you know, listening or wondering, hey, should I start a fund? And, you know, could I still raise a fund if I could have a few years ago? And if so, when is the right time?

You know, the thing that I would think about if you're considering raising a fund is how is your investment thesis going to give you Incredible alpha. And right now, raising yet another venture capital firm and putting a new one into existence, I think, is a failure of imagination, to be totally honest. Because the way in which you make money is by being non-consensus. And if a bunch of MBA students want to raise their fund, you are no longer not consensus. You are very much consensus. And as every investor in any kind of sector, any asset class knows, that is not the way to make money. The way you do that is to find unproven, assets that maybe people have undervalued. And I think that's the main issue is how do you find undervalued assets where you have unique proprietary information and you can actually value those assets correctly? And I think what's been proven in the last 10 years is that there are myriad of entrepreneurial opportunities where people need investment dollars from a variety of different types of investors. What's also been proven is that not all of these investments have venture capital-like profiles. And so it's a huge mismatch between certain types of businesses and venture capital. But This is where I talk about the failure of imagination. Everyone keeps raising a venture capital firm. What about creating new types of capital available for those different kinds of founders that's appropriate for them? And I'd love to see more innovation on that side.

I love that line of thinking. Let's, let's, let's, um, give an example in the sense of, you know, When you started Floodgate with Mike, you guys had an innovative approach, right? Which is you realized that the, the, the, this focus of risk had gone from technical risk to, to market risk. And that market risk didn't require $5 million seed rounds. It required maybe $500K, you know, pre-seed rounds or seed rounds. And so you guys came in, were offering a relatively differentiated product, a financing product to, to, to meet those changing you know, sort of requirements in the ecosystem for founders. If you were starting a firm in 2023, 2024, and didn't have the benefit of all your, you know, track record, the last decade experience, like if you were yourself a decade ago, but now, what would you do? How might you approach it?

So, I think what you wanna do is you wanna look at the inflections, right? Just the same way we as venture capitalists are asking our entrepreneurs to look at inflections, I would be looking at the inflections. What inflections do I see? I think there's 2 that are fairly obvious. One is there's been sort of this incredible rise of brands, and they're not mega brands. They shouldn't be invested in by venture capital firms, but there are a ton of smaller type of businesses that seem to appear that still need some level of funding. And there's this question of, is it venture? Or is it some other type of financing? And what do you do with that inflection point? How do you create new opportunities, new financing mechanisms for these types of businesses? It may not just be purely dollars. I think there's interesting opportunities there. The second one that's really obvious to me is, is this new, new sort of generative AI, AI component. And I actually have massive questions as to whether or not venture capital is the right way in which you would invest in these types of innovations. And the main reason is I'm not sure value accrues to startup companies selling technology. I think value accrues to companies that actually already have distribution. Some people might call that incumbents. But I think there are new ways in which you might be able to generate value and capture the value as an investor if you weren't beholden just investing into venture scales or technology businesses. So some people have asked, like, if I were to just start from scratch right now, what's something that I might be interested in? That's something I would Totally do. I would probably partner up with a more traditional private equity firm and work on new investment models there. And if there's anyone who's doing that right now that's listening in, I'd love to talk to you.

Hey, we'll continue our interview in a moment after a word from our sponsors. Turpentine VC is proudly sponsored by Synaptic. Are you an investor looking to make better investment decisions? You'll know that the quality of your decisions are determined by the quality of your data. A recent survey shows that 99% of VCs don't have a coherent data strategy. Our friends at Synaptic can provide you the data you need to join the 1% of VCs who do. Synaptic unifies over 100 real-time company performance metrics across alternative datasets like user traffic, employee data, app downloads, product reviews, and more. It's your all-in-one source for all Alternative data that helps you make better investment decisions. Synaptic are trusted by Ribbit, Felicis, Valor, GIC, and more top investors. To learn how Synaptic can improve your sourcing, tracking, and due diligence, visit synaptic.com/turpentine or click the link in the show notes. That's synaptic.com/turpentine. That's really interesting. And so they would basically work with much later stage businesses or work with service businesses or What are the types?

Yeah. I think that it's kind of all of the above, actually. I think the, you know, later-stage businesses, or actually you could even imagine public companies. So what do these, you know, PE firms purchase? They will oftentimes purchase a company that already is public and combine them and take cost structure out. And usually that is by creating more efficiencies across all of their acquisitions. I think there's like new ways of creating both efficiencies, but also knowledge, you know, cross-pollinating knowledge across many different organizations, which I think there's huge opportunity for.

That's really interesting. I, in the spirit of giving people ideas, I have another friend, Jeremy Giffen, who was the first employee at Tiny, uh, that, um, you know, sort of incubates and, and buys, um, companies. And what he's trying to do is create a vehicle to buy some of these companies that are massively overfunded but actually have real revenue and actually have, have potential and, and kind of clear up the—

Yeah, we've seen that with SaaS businesses and combining them and giving them a little bit more scale, reducing their customer acquisition cost. And actually, you know, because some of these SaaS businesses actually do create real value for their end customer, it's just not quite at the scale that they need to. And so they're stuck. And so you see this quite a bit.

Yep. And also I have another friend who's trying to create a vehicle to invest in, I think, companies about either about to go public or who are who are public. I can't remember which one, but just feels that they're massively undervalued in this kind of correction time, but still have long-term potential. Floodgate has stayed pretty consistent, it feels like, to your original mission, or relative to other firms who have gone massive, aggregated AUM, become multi-stage. I'm curious how you've— you guys are famous for saying your fund size is your strategy. How have you sort of settled on the ideal fund size, sort of portfolio construction, you know, portfolio company size, you know, how many companies you need in a fund that, you know, you feel is a right model for you guys?

Yeah, I think for us, we think about what each partner is capable of delivering on. And so we have sort of a checkbook mentality for each partner. We also have to be honest about what is the stage of investing that we love and we're good at, and where will we see not so much economies of scale, but like over time, how will we accumulate knowledge that is really valuable and unique? And the thing that we really settled on early on and we continue to double down on is this notion that even though we're investing very early, we are high conviction investors. And that means a few things. One is, this is the only stage at which we will invest. You can't come back to us at the A or the B and say, hey, we really liked that first meeting. Let's talk again. This is our only shot. The, the second is that it's the only thing we really know. So we know negative 1 to 0 and 0 to 1, and we've, we know it really well because we only each make, each partner's making what, 2 to 5 investments per year. You know, this year I've made 1 investment. Last year I made 2. And so, you know, and my pacing has been roughly 2 to 3 the last few years, and I really understand the decisions that our founders are going through on any, at any, any given moment. And what that affords us is the ability to say, okay, when they moved from this direction to this new direction, we understood why they did it, how they did it, and What were the drivers for making these decisions? And we think that those stories are incred— incredibly important as we guide new founders in their new, new sets of decisions. Um, and so we're experts at that. We're experts at helping founders figure out the company that they want to work on, um, even before they have an idea and helping them refine On that, that idea. We're experts on once you've landed on an idea, how do we help guide you to, do you have product market fit or not? And that's the only thing we focus on. And sometimes because we're so involved at that stage, we'll be asked to stay on the board even longer. So I just finally rolled off the board of Lyft, but it was a 13-year journey for me. Um, and I, I, I'm proud of that because I feel like it's a representative of the fact that as a seed stage investor, we take our job very seriously. And it's not just sort of a series of seeds that we're sprinkling everywhere and hoping that one sort of luckily takes on.

That, that's a fascinating overview. On the product market fit question, how do you know if you have it or you don't? Some people say, oh, you'll just You'll know if you have it, and if you don't, uh, if you don't know, then you don't have it. What is sort of your— is that too simplistic, or what is sort of your, um, litmus test in terms of helping entrepreneurs sort of realize, like, do they have it, do they not, etc.?

So, I, I think, like, do you have it or do you not is less important than what do you do if you don't have it, right? And, and I think that that whole question of, like, the litmus test is Most founders will come back and be like, well, I guess I don't have it. And so how valuable is that test? It's like not valuable at all, especially if your investor's saying, well, you'll know it when you have it.

Yes.

Um, that, it's like the least helpful thing that someone can say. And so to me, it's about how do you help someone figure out signs of life, but also like what's keeping you from getting there?

Right?

Is it this market? This market isn't big enough? Is it that, you know, you need to just add more geographies in order to see? Do you need— do you have a problem in pricing? Is it that the product offering isn't compelling enough? How do you know that? But most people will just spend a ton of time on, well, okay, like if you have This product and this is the value proposition, and you turn this button from green to red, like maybe you'll see more conversion. It's not in, in those kinds of details. It's in sort of more the meta, which is around what is your business model? Who else is in this ecosystem? What role are you playing? How do you figure out whether or not that role is appropriate or not? Who else, you know, might be trying to get into that position? Do you have to educate your customer or do you not have to? There's all of these different dimensions. And the reason there isn't a recipe book for product-market fit is that it is a very individual journey line for every company. And so I think it is extraordinarily frustrating because if there were a recipe book, then like, You know, as a, as a lazy investor, I would be able to come in and just say, oh, okay, like here's a checklist and I know you haven't done A, B, and C. And so therefore, you know, you don't have product market fit and you need to do these steps in order to get there. It doesn't work that way. You actually have to understand the nature of the business. What has been happening? What's the historical, you know, path this company has taken? And only then can you really engage in that dialogue.

You talked about how you think about your firm strategy based on your guys' skills and interests. What would need to be true for you to change or evolve your strategy such that you're doing something like also leading Series A's? Or when you look like, how would the market need, need to change or the sort of venture landscape for you to add something material to or change something material to what you do?

For us to do a Series A as a new investment, I think we would just have to actually be hiring in a completely new skillset into our organization, because I think that that stage of growth is just so fundamentally different from the expertise that we have. It would actually also have to change the way we run our decisions. Because to be honest, like for us, we're so early that the only thing we ask of our partners is that you pound the table. Like you have to personally have conviction. I think if we were writing much larger checks, $10 million into Series As, A, I would believe we need to have much more interest areas where we have some conviction.

Yeah.

significant expertise and coverage. We couldn't be generalists, and we would actually need to track companies in a very fundamentally different way. I think, you know, the mentality that we have is that there are very specific characteristics of people that we're looking for. And when a lot of people will say, oh, we're people investors too, we are people investors to the extent that if you— I will invest in people and tell them I hate their idea, that I don't think their idea has legs. And sometimes I'm wrong, right? I have companies that have done fabulously well after I told them I love them, but I hated their idea and I invested. I believe in people. I think Mike is the same way. Arjun is the same way. Lior is the same way. We, we see something in the person who's starting the company. We see some of their initial insights, and we love not only the founder, but we love their insight. Like, we love the, the place where they're starting, and that may have absolutely nothing to do with the product that they're building at the time. And that's a very different way of thinking about things than Looking at a retention graph, understanding whether or not you have what kind of ARR company has, what their customer acquisition cost is. It's a very different coaching mechanism. It's a very different board meeting. It's a very different way of running the firm. And so, It's partially why we like to go even earlier to a point where the founder doesn't even have an idea, because we feel like we know how to assess people, and we, we are less good at assessing just pure graphs.

And on the people side, is this— would you say your, your capabilities there stem more from just kind of intuition and having done it for so long and done it so well? Or is it— are you guys looking for different data? Or have a kind of different way of evaluating people, or is it both? Or what would you say to that?

It's a little bit of both. I mean, I am always evaluating people based on a few different characteristics that I'm looking for. Some, some around building capabilities, some are their capabilities around navigating IDMAs. For me, there's even a diligence process that I go through to understand how this person thinks, how this person builds. It is partially intuition. It's also partially how we interact. But I have to say, I'm not a universal— I'm not an investor that has universal appeal. For sure. You know, I am probably a little bit of a maniacal truth teller. One of my founders was telling me that he did this reference and this founder said, I mean, she literally said that she hates my idea, but she really wants to work with me. Like, but everyone else is saying they love what I'm working on. They love me. Like, She isn't even paying the highest price. She's not even giving me a lot of dollars. Like, what should I do? And this founder was like, hey, I've, you know, I loved the truth-telling. Like, that's what you need, and you don't need that many dollars right now. Um, and so I don't know if it's an acquired taste or whatever it is, but it's a very specific taste. And I think that's good. I think like, you know, the way I work with a founder is pretty specific. And so, if it doesn't work, it doesn't work, and that's great. But if it works, then I feel like we have a great relationship. It's based on real trust, um, and understanding. And, and so, you know, to me, that, that's the magic. That's why I wake up in the morning and I'm super excited to come to work.

Talk about the The muscles you've had to build as a venture firm on the negative 1 to 0? Because it feels like you guys are, you know, pioneers in the space.

I think there's a certain articulation that is hard to do. So the first is like, most of the time I'll get this thing from a founder who'll say, hey, I thought you were early stage and you're saying this too early. And so really explaining to someone what is actually appropriate for seed or pre-seed firm. And in my mind, for a seed or pre-seed firm to have any kind of advantage at all, you can't be in the mess of competition. And so if you're investing into something that has 20, 30 competitors and they don't have something very specific that they understand, that's almost like a secret. We call it an earned secret. Then you might as well invest in the A, right? You're waiting for the winner to be revealed and you throw yourself into the gauntlet and hope that you are able to give your dollars and have your dollars accepted by the winner. But there's actually like no advantage to picking that person early on unless they show that they have an advantage amongst all of those competitors. And so the seed, what I'm investing in is that insight plus my belief that they're sort of headed in the right direction. Because with the seed, you're setting your timer and you, you are saying go. And within 2 years, you need to raise a successful Series A, and that means you need to have a certain level of traction, and you need to have had success in proving out some level of alleviating some of the risks that you have within your business. And so it's sort of, you're on the track and you better go as fast as you can. There's very little room for saying, you know what, that was a huge mistake. I'm gonna start back from zero again and start over because the clock is already ticking. And now what you have to do is start over, catch up to where you should have been, and then move forward. And so that's, that's seed. To me, pre-seed is we're just in the playground and we're playing. We might have the kernel of an idea, and and what we're looking for is that direction to run in. And the moment that you have signs of life in that direction, we should raise a seed. But it gives you the opportunity to completely say, you know what? I'm just erasing the board again and starting all over again. And I've had companies start. Start all over like 4 or 5 times before they got to, this is the idea. It's sort of this notion of product-market fit. When they get to that idea where like all of us are like, that is an awesome idea, let's go. That's super exciting. And it's, it's not even in the numbers. It's just like you have a sense of this will have legs. Um, and so that, that to me is the difference between pre-seed and seed. And most Series Seed and A investors will look at how much you raised to determine if it was a seed or pre-seed. And so I like to raise less than a million for pre-seed and, you know, $1.5 really to maybe $3 million for the seed.

That, that's helpful context. Sam Lessin came on the podcast and he talked about how What he calls this factory model of venture capital is not going to be as strong going forward. What he calls factory model is, hey, you raise a seed round so that you can then raise a Series A, you hit these metrics, you then raise a Series B, and there's this kind of venture supply chain that is investing kind of on belief that there will, you know, you will package a unicorn at the end of it. And what he says we're learning as an ecosystem is that turns out There's less unicorns than we thought. Perhaps some of these unicorns were, were never actually meant to be unicorns and they were just, um, sort of marked up because everyone had the incentives to be marking it up at every step of the supply chain. That's kind of how it worked. But enough time has passed that we're seeing that there were these fake unicorns. And so, you know, starting from public markets down, people are going to be less, um, trusting or they're going to be more circumspect, you know, certainly the, you know, Tiger Global coming in and SoftBank, et cetera. You know, helped that supply chain and now they're, you know, less active. And so I'm curious if you're sympathetic to that idea or if you agree with it.

I am sympathetic to it. I guess I just haven't seen that impact the seed yet because if that were true, we would've seen seed-level prices drop, but we haven't seen that.

I think the implication is more that, or one of the things he says is sometimes instead of raising a seed round just to raise a Series A, maybe companies will raise maybe a bit more in the seed and maybe have the option to being profitable afterwards. Um, um, just because it maybe it'll be a bit harder to, to raise that, that, that follow-on round just on sort of, you know, promise alone.

I, I would agree with that, except most of the companies that we see at the seed stage Can't possibly predict what they need to get to profitability. And so I, I think that, that ask is hard. And so a good way of mitigating against that is to raise less at lower valuations and then leave yourself the option of raising a little bit more to get to profitability once you know what that business looks like. I think the valuation in the early days really matter, right? If you're raising at a $20 or $30 million valuation, it is not a slam dunk that the seed stage investor will actually generate great returns off of that investment. I think Sam also talked about this off of You know, fairly good, you know, outcomes. And, and, and it's true. And, and so I think, you know, this is where a lot of people will say, well, you know, they raised at this valuation and then they exited at this valuation. And so therefore the investor made whatever final valuation divided by initial valuation. That's the return. That's not the way it works. And, and so really understanding the dynamics of the capital needs of the business alongside what is your going-in valuation, and then what do you actually really believe is a possibility of exit scenarios is actually really important. And right now we're seeing a real bifurcation where companies will get bought for $1 billion. or exit for $10 billion, but you aren't seeing a lot of what we used to see when we first started Floodgate, which was, you know, the $100 million exit or even the $50 million exit within, you know, 24 months of a seed stage investment.

Yeah. How does that change what one invests in? It, it, um, does it mean that certain companies like, um, you know, we had a lot of companies over the past decade, like there's a D2C craze, you know, there were a lot of like kind communities being funded. There were certain kinds of businesses that maybe didn't require venture capital or just weren't the best use case for venture capital. Do you feel like we as an industry are kind of narrowing in on, you know, what's appropriate for venture scale? I don't know if you guys ever did one of these businesses that I'm describing. How do you think about it?

I mean, so we do believe that there are great companies that are built at any point in time. But with the popularization of founding businesses and investing in businesses, I think we've watered down what the real point of venture investing has been. For me, the point of venture investing is to actually invest in something that has the potential to really change the way society interacts with a particular area and really change the workflow of life, to change the way a person lives their life, or change the nature of work itself. It's these huge, grand ambitions. It's not a small optimization on a on edge case scenario. And if you buy into that plotline, there are a lot of different ways in which founders can actually really impact the world. And there are grand challenges that exist everywhere that we can participate in and that have significant value, not only to society, but also Economically. And so, you know, I think a lot about that. I think about where do you have the opportunity to really change the nature of an industry? Where do you have the opportunity to really make a difference? And does this piece of software have the potential of sitting front and center within that. And I think a lot about, you know, the center of gravity of an industry. Will this company sit within that center of gravity? And where I've gone wrong is when the investments that I'm making are great investments, or they're people I really like, and it's not centered on that foundational thought of, is this an insight that I really love because of the potential impact? And when we come back to that, and when we come back to this notion that we talk a lot about, greatness is a decision, and it's a decision you constantly make, then you start to realize there are companies that are really worth investing in, and there are other companies that you just really like. But it's just not our business model.

So you would've been unlikely to do something like a Dollar Shave Club or Casper or Blue Apron or one of these types of businesses.

And they were great investments.

That makes sense. Going back to your AI point, I remember, you know, when ClearBank and Pipe came out, they were talking, it was kind of a, you know, new way of funding, you know, SaaS businesses that was perhaps a better a better fit given the economic environment at the time. And ClearBank also had a mantra of, hey, venture capital should be for funding a different kind of risk or a different kind of profile company. I'm curious if there's some parallel to AI where in fact, even on a startup level, these companies have to raise, they need this capital for the H100s or whatever it is that And what is the best sort of, you know, finance vehicle for that? Is it venture or is there some other instrument?

I think it depends. I think AI is actually turning out to be not just sort of a technology, it's not an industry, it's sort of, it is this inflection point. So the way I read AI right now, and, you know, we've been investing into different versions of AI since 2000, I guess, '10. I, my math modeling background kind of plays into this. We invested into what was called big data previously, and that was sort of machine learning algorithms. And even back in 2010, you could see that that was going to impact healthcare. It was going to impact financial services, defense, all of these other spaces. And then we got super interested in the data behind that. So some of it was like around personal knowledge management or knowledge management. And, you know, today what we're seeing is this real interesting inflection point from a technology standpoint, which now has given compute an API to humans. And And that what that means is now it's made compute a collaborative function. So now I'm not just giving instructions to computation, but rather we're collaborating together to discover things. And that's just a very new model of working with these resources that we've had. And I don't think we've scratched the surface of what that means. And that's just in language so far. And you could see that it's coming in all sorts of other domains as well. Certainly visual, audio, all these other formats. And so, you know, saying that, you know, what's the right financial mechanism for AI? Is a little bit difficult because I think there are definitely places where we can actually make real investments. But what it requires is for companies and founders in particular to have a real insight of this is how it can be used. And this is something that I know that most people may underappreciate or not understand. And, and I think, you know, that's where the rubber hits the road because otherwise you're just investing in, you know, one of 100 companies that are trying to do marketing or sales software using generative AI and sort of welcome to the world of generative AI. There's thousands of companies doing that now, and it's super easy to create a company Because all of this software is now available. It's a technology that's accessible by API, which means it's literally technology accessible by anyone who can code. And so there's no moat in that. And so how do you actually create something that is defensible for you to get seed investing? I think, and now I think a lot of people are thinking about this too. So You know, to get even Series A investing is going to be tricky. And so you have to, you have to think through that piece.

Someone was asking Elad Gil this, the same question about moats, and he said something like, yeah, but also like Notion or Retool, when they started, they, they didn't have a moat either. They just, they built a, a great product and then they got locked in via the sort of contracts that they were able to, you know, create and then the brand that they built on top of it. And so, Like, are you excited about either vertical applications or, yeah, so where are you excited within AI?

So, number one on that front, again, if you're a Series A investor, you have a huge advantage there, right? 'Cause you can wait until they've kind of gone on their walkabout through the wilderness and then they've come out on the other side and they figured it out. So, that's sort of one thing. But as a seed investor, you know, we are looking at places where, founders will have some advantage. So one is a natural place where I, I see a lot of other investors saying this too, is just sort of verticals. And so if there is a way in which you can describe why a new foundation model needs to exist in this space, so maybe the tokenization is a little bit different given the data that you have. I think that's really interesting. And we've seen a bunch of companies in that space. And there's reasons why, like, the workflow becomes really important. And how do you own that workflow? How do you own the data? There are a lot of examples where, you know, founders are working in very, like, unsung verticals. And I think that's super exciting. The second is, I think we haven't really figured out UI either. So going back to this Notion example, I You know, everyone is so stuck on ChatGPT that everything looks like a messaging app. And I don't think that's necessarily the only way to unlock the power of LLMs. It's definitely a way of doing it. And everyone's still stuck on prompts. And so, you know, I'm always curious about new mechanisms by which the end user interacts with the model. And I also think that, you know, the ability to create new ways of leveraging LLMs will be really interesting. So, you know, whether it is running simulations Leveraging these LLMs, I think there's new ways of understanding what language means. And so therefore, you know, how do you actually, you know, attack the language or create new ways interacting with language? I think that's really interesting as well.

I want to be mindful of time. This is a great place to wrap. Ann, thanks so much for coming on Turpentine VC. This has been a great conversation.

Thank you.

Turpentine VC is a podcast from Turpentine, the network behind Moment of Zen and Econ 102. If you liked the episode, please leave a review in the Apple Store or rate us on Spotify.