E413: How AI Will Reinvent Venture Capital

0:00What does it mean to be an AI-native venture capital firm? What it means to be AI-native is that you're not just sticking AI into the firm's existing processes, hoping that something's going to change. It's actually thinking about the whole firm from the ground up with AI at the core. And so there's one example of this that's quite basic, right? A non-AI native venture firm, in my opinion, is a firm that still has humans who are actually looking at the CRM every day to figure out what's in the deal pipeline. An AI native firm is where the humans in the firm don't even look at the CRM because that's updated agentically.

0:43It's happening in the background and where things are being done for you proactively through the use of AI. And so I think that's the amazing opportunity that firms like us have right now, which is to rethink our entire processes with AI from the ground up. We're recording right now. It's 1.30 on a Tuesday. How have you used AI today? Let's take the prep doc that you sent over before this podcast. I think I got it from you late last night. I took a quick look at it. when I woke up today to take a look at the doc again, I had a number of bullet point suggestions based on, to talk through your talking points, based on our agents having gone and looked at all the context in our firm, the conversations we've had about being an AI native firm, the logs of what our agents have been doing over the past couple weeks.

1:45And so I had bullet points prepped for me with very little input from myself. based on our use of AI. And so that's just one way I used it today. Another way is I had a board meeting this morning that I just came from to this podcast and we got a board deck last night for the company. Wasn't that much ahead of the meeting, but I got notes on what has changed in the business between this board meeting and the last one. What are some of the conversations that we've had back and forth over email since the last board meeting and how did they relate to what we're about to go in and discuss in this board meeting.

2:23And so I feel like I showed up as a board member with just a lot more background understanding of what had transpired between sessions than I would have otherwise. And it wasn't like I actually stared at the prep during the meeting to figure out what questions to ask, but I did feel better prepared for the conversation that we were about to have based on that came from our AI agents. Let's go into the plumbing a footwork, an AI native VC firm. How have you constructed the firm to integrate AI? Give me some specific examples. We've gone through different evolutions on this, right? So initially, we were, of course, just using ChatGPT and Claude, sticking data into it, figuring out sort of what can the models do.

3:11That was sort of V1. V2 was us starting to build agents. And initially, actually agents on our local machines. So you've probably had this experience too, David, where if you fire up Cloud Code and you start building an agent, it can do things for you on your desktop. If you have a Mac mini in the office or at home, you can run these agents there as well. But we had some funny experiences where we wanted to use our agents, for example, to do a brief on a company or to go find out a piece of information about our portfolio companies. And the agents wouldn't be able to do that because the laptop of the person running the agent was closed.

3:48And so V3 for us was migrating these agents to the cloud, was having a central database, Postgres database, where all of the context from the firm lives that enables us to build agents on top of it. And so that's the phase that we're in right now. And that's a phase where agents are always on. They're doing things constantly in the background for us. we can run them at any time and they're also picking out context based on the query from us versus leveraging apis to tap into each individual system if that makes sense which is a lot more costly from a token standpoint it takes a lot longer to get back results i'm sure we'll have a v4 and a v5 to talk about over the next year because things are just moving so quickly on the capabilities and models, the stacks that make the most sense as well.

4:42How much does AI cost you on an organization-wide basis? The cost for us is actually still pretty small. It's in the thousands of dollars monthly. It's not yet in the tens of thousands of dollars monthly, and certainly not more than that at this point. Again, we are a very small firm. It's only five people at Footwork who are full-time employees. We're running sort of three or four agents really aggressively at the moment. So the cost is actually still pretty minimal for us. I want to get into the weeds in a bit, but at a high level, what is AI specifically doing for I wrote a blog post earlier this year talking about the jobs to be done within a firm like ours and where AI can add value.

5:31And so when you think about the venture capital flow chart, it is find, decide, win, help, exit. Those are the five functions that we have as investors. And then there's also a firm building function where those are internal processes, finance at the firm, relationships that we have with our LPs, fundraising. AI is adding value across all of those dimensions. So find, decide, win, help, exit, and firm building today. Let me try to give you a few specific examples. Our agents are named after, right now, football managers. Both my partner Mike and I are very into football or soccer, as we call it in America.

6:11And so we have an agent named Pep, an agent named Foggy, an agent named Mo, after three of football's best and most famous managers. pep is the agent that briefs us on companies that we're looking at or companies that we perhaps should be looking at so it's the main agent on the find dimension right now and so every morning i get briefs from pep on the companies that i'm seeing that day how they perhaps relate to companies i've seen in the past what i thought about those companies in the past just to context load me as I go into those sessions with new companies. And it's becoming more and more proactive about what it can do.

7:01So imagine things like suggesting other companies you should chat with that are related to this particular company if you find it interesting in the meeting. Fogey is our agent that has full knowledge of everything at Footwork. And so yesterday I asked Fergie, can you just take a look at the closing docs of this particular portfolio company and remind me who's on the board of the company officially? That was an answer that it was able to give back. And so it's kind of our internal company brain agent. Those are two examples for you that are hopefully fun to think about that we're getting leverage from today with AI.

7:44And you said something interesting earlier. You said you're not using CRMs like a typical VC would be inputting data. You're using AI to populate those CRMs. Talk to me about the interaction between AI and human and your processes. When we started working on this, we were wondering, what will our stack look like in the future? Will we even need a CRM? And we actually have found that there's real benefit to still having a CRM. So we still have Affinity as our CRM, but I don't look at Affinity much at all anymore because the way I interact with Affinity and the data within is through our agents.

8:24And so as an example of what we do, when I have a meeting with a new company, I take notes on that company. If we're using Granola in the meeting, Granola has notes as well. But our agents automatically take those notes, put it into the CRM, and then they track how we're conversing about the company over time to update where that company is in our pipeline. So our agents will know if I've passed on the company because it will see that via an email to the company. It will see that via an internal conversation we have, such as a team meeting conversation where we talk about the pipeline and we make decisions on which ones we're excited about and diving into and which ones we're less excited about and are moving on from.

9:09Before all this, I would actually manually take my notes, stick them into Affinity. I don't do that anymore. I would also manually go and take a look at the pipeline and update the status of each company. I don't do that anymore. But we do actually still have the CRM. It is a single source of truth still for us, or one of the sources of truth, I should say, for us. important because over time what it's captured is how we've thought about every single company, which decision we made, and that context is very important to have these agents be able to improve and learn from themselves. There's this management consulting term today called change management, which is just a fancy word for talking about how everybody now needs to integrate AI and what are the best practices for that.

10:00What have been the best practices for you for integrating it within Footwork? The first step for us was just making sure that everyone was familiar with the tools. Everyone felt empowered to take a little bit of time to set things up such that processes could be automated over time. And we started talking about every week in our team meeting how every single person is using AI. so you kind of had to show up to that meeting with something to talk about and then of course there is there's a distribution in people's usage of AI over time but our our philosophy was like the floor for everyone can be raised through automating processes to creating agents and then some people are going to find kind of the really high ceilings of possibilities and hopefully those can flow back into the average being a higher floor of people's usage.

11:04At our portfolio companies, we've seen a bunch of different types of efforts to get people to use AI. We've seen companies mandate AI usage or else you won't be at the company. We've seen companies closely track people's token usage and have leaderboards that sort of enable people to see how different people are using AI, companies that have systems for folks who are more AI native to mentor folks that are a little bit behind in that learning and to sort of to graduate in their learning as a result and for those mentors to be incentivized and rewarded for that work alongside their individual work at the company.

11:48I think that all of us are in this phase of just trying to figure it out. Like the interesting thing is that no one is really an expert yet. Even the people at the very bleeding edge are discovering every day, it can suddenly do this for me that I never would have thought possible. So the key is just to be in that learning mode and mindset. And so I think that's what everyone has been trying to figure out how to promote internally. Maybe you could double click on that. what are some ways to get the organization in this learning mindset? It's making sure that people who are more at the frontier are sharing those learnings with everybody else.

12:31In our firm has had an aha moment and they've been different for every individual, like for when they realized, wow, AI can do this. And it's meaningfully making my day-to-day job better as a result. And so just trying to find ways to get people to the ahas to then realize, wow, I should go deeper on this. I should, I've become obsessed by it myself. That's the key. It's kind of like what teachers have to do in education, right? Everyone learns a little bit differently. Everyone has different skills. And the key is to get someone to a magic moment where they feel so passionate and curious that they go figure it out for themselves.

13:19I found the same both in our firm as well as in portfolio companies is that the processes that seem to permeate across the organization starts with somebody getting an idea, but most importantly, executing it, essentially de-risking it and trying this out, figuring out all the kinks, locking it in, using it every day. and then teaching other people in the organization exactly what they did. Kind of this end-to-end solution on one individual. It works. Everybody could see that works for this person. And then he or she goes out and teaches everybody else on the team. Is that how it's worked for footwork?

13:55Or do you have an AI person that figures out the best practices, works out the kinks and then integrates it into the other partners? Initially, we were doing all this work ourselves. And then what we realized is we think we'll really benefit from having somebody who's fully focused here. And so we went out to make that hire and we found Andrea, who joined us a couple months ago, who is fully focused on everything AI internally. So he's our AI lead. Think of him as kind of the operations person of the future that every venture firm will have, but perhaps even almost every company will have. Because what he does is manage the agents internally.

14:43And so I think an operations person of the past would have only managed humans, but I think an operations person of the future manages AI agents as well as perhaps humans as part of that day-to-day job. So Andrea owns all of our AI work now. He's building agents. He's sort of updating our context layer every week. He's taking learnings back to our team. on capabilities that he's seen unlock for himself or for others on the team. A side benefit of this is having someone who's just playing with the frontier capabilities of the models on a daily basis, I think is very helpful in informing our overall thesis and worldview for investing.

15:31That is another thing that we are having more and more brainstorming sessions about and that is getting updated in the background based on all the conversations that we're having as a firm. Sort of what does footwork think about the world in 2026? What do we think are the most investable areas and opportunities for us? What are less interesting for us and why? And so having Andrea kind of be the DRI on this, I think is already shown to be very beneficial. and we'll see what unfolds in the years ahead. Most investors don't lose because they lack information. They lose because they can't separate signal from noise.

16:15Every day, thousands of earnings calls, SEC filings, expert interviews, and market updates compete for your attention. The challenge isn't finding more research, it's finding the few insights that actually change your investment decisions. That's exactly what I unpack in my conversation with Ryan Fenerty. Instead of talking about AI and theory, we discuss how leading investors are using it to surface differentiated research, move faster than competitors, and make more informed investment decisions. If you're an investor looking to gain an edge through AI-enabled insights, I think you'll find this conversation valuable.

16:43You can access this limited release episode by going to alpha-sense.com slash how I invest. That's alpha-sense.com slash how I invest. I just recorded with Thomas Scriven, who's head of private equity at University of Pennsylvania Endowment, and he invested in the manager of three former Palantir engineers, and they will deploy themselves, they're bio managers now and they'll deploy themselves into portfolio companies and figure out their biggest AI bottlenecks for five to six weeks, unlock these bottlenecks and that's kind of how they're using that to get into the top companies. That's super cool.

17:20When this moment in time where every organization is trying to figure out how to best leverage AI, some have a mandate to do it, others just know that they should be doing it. and we're all trying to figure this out. And so I think that is a really interesting wedge to get to know companies really well and figure out what to invest in or to gain access to something that you're excited about. You mentioned that you're using one of your agents, I believe it's Mo, focuses internally within Footwork on the firm itself. Do you see that evolving venture capital into new business models or more scalable ways to deploy capital?

18:04That's a good question. There's that possibility. I think we still feel like our overall business hasn't changed and doesn't need to change. We love partnering with just a handful of companies every year, working really closely with those founders, leading rounds, getting our ownership in the initial check, yet felt that innovation is required on the model itself. But again, I think we're in this phase where the world is changing rapidly and we'd be silly not to think about how will our overall industry change and what's our place in it that perhaps needs to be adjusted as a result. We'll talk about looking for AI firms that provide value via AI, but there's also a bunch of firms that are utilizing AI internally.

18:53What are you looking for that tells you that somebody is going to win in this new world? The gene, the trait in founders that we look for more than ever is their slope of learning. How quickly are they able to learn and adjust themselves? Because when the world is moving as quickly as it is, it feels like that is maybe the most important trait to have. And so how does that show up in actually evaluating companies? We ask questions such as, what's changing your thinking about the business this week? Take different functional areas or topics they're thinking about and seeing even in a couple different days of meetings, has their thinking adjusted in a certain way?

19:40Are they learning something on the field that has them thinking differently about an element of their business? And so that is a really important characteristic for us to look for in founders and one that we were always interested in, perhaps less fixated on than we are in this moment in time. Footwork is one of the first venture capital firms that at least I'm aware of that's using AI as an investment partner to help make decisions on investments. What's the story of how that came to be? how it came to be is that we were seeing all these companies leverage ai and have completely new businesses as a result of the advances in the llms and we felt like the only way for us to invest in that world to have a different perspective is to try to become an ai native firm ourselves and we felt we had the unique capability to do it because we're only a handful of people versus a firm that is much bigger and has many more legacy processes as a result.

20:50We started on this journey several years ago. We've gone more and more all in on it in 2026 as the models themselves have become better as we've realized we can really get leverage out of this. And so how it shows up to us, thinking partner as an investment partner is because our agents have context in all the decisions that we've made historically and also on companies that we wish we were a part of along the way, but we didn't get to meet or we didn't prioritize. What we're trying to do is leverage these agents to help us make better decisions. At the end of the day, that's what could drive a tremendous amount of value for us.

21:42Every time we now go in and try to make a call on a company, it feels like we're better prepared to make that decision than ever before. It feels like our judgment is captured and therefore should improve over time. We'll see if we feel like we really got leverage out of it only in many years from now when we see the full results of these decisions. One of the really hard parts about venture capital is just how long these cycles take for us. But we're optimistic that we'll realize it was an amazing idea to have invested in this in 2026. Does the AI get its own vote? Is it helping improve your thinking?

22:26How are you actually leveraging it on the investment side? It doesn't get its own vote today, but it has perspective on what are the key questions that we should ask. It tries to power it back to us. Hey, here's where you're seeing it spike. And here's how you've seen similar companies spike in the past and what the outcomes were of those decisions. And so it's not yet an end-to-end decision-maker thinking partner, but it is providing us leverage to, I think, have better judgment ourselves. And I think the potential over time is for it to really become a decision-maker and judge or companies. The models are just not quite there yet.

23:18the context layer we have internally, I think still needs to be bolstered for that to actually be the case. Look, we're only three and a half years into this LLM era. And so the models are only going to get better over time. I'm optimistic about what we could see three and a half years from now as a result. And you have support for today's episode comes from Square, the all-in-one way for business owners to take payments, book appointments, man staff, and keep everything running in one place. Whether you're selling lattes, cutting hair, running a boutique, or managing a service business, Square helps you run your business without running yourself into the ground.

23:56I was actually thinking about this the other day when I stopped by a local cafe here. They use Square and everything just works. Check out is fast, receipts are instant. Sometimes I even get loyalty rewards automatically. There's something about businesses that use Square. They just feel more put together. The experience is smoother for them and it's smoother for me as a customer. Square makes it easy to sell wherever your customers are in store, online, on your phone, or even at pop-ups and everything stays synced in real time. You could track sales, manage inventory, book appointments, and see reports instantly, whether you're in your shop or on the go.

24:26And when you make a sale, you don't have to wait days to get paid. Square gives you fast access to your earnings through Square checking. They also have built-in tools like loyalty and marketing. So your best customers keep coming back. And right now you can get up to$200 off Square hardware. When you sign up at square.com slash Go slash how I invest. That's S-Q-U-A-R-E dot com slash go slash how I invest. With Square, you get all the tools to run your business with none of the contracts or complexity. Run your business smarter with Square. Get started today. Support for today's episode comes from Square, the all-in-one way for business owners to take payments, book appointments, man staff, and keep everything running in one place.

24:59Whether you're selling lattes, cutting hair, running a boutique, or managing a service business, Square helps you run your business without running yourself into the ground. I was actually thinking about this the other day when I stopped by a local cafe here. They use Square and everything just works. Check out as fast, receipts are instant. Sometimes I even get loyalty rewards automatically. There's something about businesses that use Square. They just feel more put together. The experience is smoother for them and it's smoother for me as a customer. Square makes it easy to sell wherever your customers are.

25:26In-store, online, on your phone, or even at pop-ups and everything stays synced in real time. You could track sales, manage inventory, book appointments, and see reports instantly, whether you're in your shop or on the go. And when you make a sale, you don't have to wait days to get paid. Square gives you fast access to your earnings through Square checking. They also have built-in tools like loyalty and marketing. So your best customers keep coming back. And right now you can get up to$200 off Square hardware. When you sign up at square.com slash go slash how I invest that's S Q U A R E.com slash go slash how I invest with Square.

25:56You get all the tools to run your business with none of the contracts or complexity, run your business smarter with Square. Get started today. Support for today's episode comes from Square, the all-in-one way for business owners to take payments, book appointments, man staff, and keep everything running in one place. Whether you're selling lattes, cutting hair, running a boutique, or managing a service business, Square helps you run your business without running yourself into the ground. I was actually thinking about this the other day when I stopped by a local cafe here. They use Square and everything just works.

26:22Check out as fast, receipts are instant. Sometimes I even get loyalty rewards automatically. There's something about businesses that use Square. They just feel more put together. The experience is smoother for them and it's smoother for me as a customer. Square makes it easy to sell wherever your customers are in store, online, on your phone, or even at pop-ups and everything stays synced in real time. You could track sales, manage inventory, book appointments, and see reports instantly, whether you're in your shop or on the go. And when you make a sale, you don't have to wait days to get paid.

26:49Square gives you fast access to your earnings through Square checking. They also have built-in tools like loyalty and marketing. So your best customers keep coming back. And right now you can get up to$200 off Square hardware. When you sign up at square.com slash go slash how I invest. That's S-Q-U-A-R-E dot com slash go slash how I invest. With Square, you get all the tools to run your business with none of the contracts or complexity. Run your business smarter with Square. Get started today. Front row seat to the evolution of AI when it comes to venture capital. How do you see that changing venture capital in five years or 10 years from now?

27:23It's already changed in a few ways, which is there's work that I did when I started in the industry, such as analyzing the data room and rebuilding the financial model. I've been doing reference calls, all of which can be automated today. Unfortunately, the entry-level analyst role, it doesn't feel like one that we need to have at Footwork. And I think other firms will come to the same conclusion. Now, will we still potentially hire analysts as an industry? Yes. But I think those analysts will be different in their jobs, will be wielders of the AI products. and so that's just one area that i think will really evolve the back office of firms should also look much leaner over time i think a lot of those processes can be automated and then i think the key for us as humans building and running these firms and investing is going to be back to the work of judging companies where again i think that the last place that ai will will really be able to make end-to-end actions lots of pieces of find decide win help exit and firm building will get automated there'll be fewer people needed to run a venture firm and the people who are in the firm as humans will have more of their time freed up to just spend time with entrepreneurs and others in the ecosystem to think and to make decisions versus to do tasks and a bunch of the other things that come with company building for anybody.

29:25Are those the last two things that get automated, decision-making and relationships? Is there anything else you'd put in that bucket? I think that's right. relationships manifest themselves across sourcing across winning, across helping companies of course across exiting companies too so I think it is those two things, I think it's relationships across all those dimensions and it is fundamentally about human judgment and decision making that will be the last things to be AI'd away at some point There was about 3 ,000 firms, 3 ,000 venture firms. Many of those are on their last fund. How do you see AI accelerate the extinction of emerging managers?