In short
Podcast Summary: How I Invest with David Weisburd - Episode E309: Why Most VC Firms Will Die by 2030
Episode Overview In this episode, David Weisburd interviews Camilo Acosta, an experienced AI engineer and founder of Perceptive Ventures, focusing on the implications of agentic AI and its impact on venture capital, labor markets, and investment strategies. The discussion centers on how AI is transitioning from a tool that assists humans to a decision-making entity that could potentially disrupt various industries.
Key Topics Discussed
- Background of Camilo Acosta
- Formerly a top AI engineer at Meta.
- Founder and managing partner of Perceptive Ventures.
- Combines experiences from both startup and large tech environments to guide investment strategies.
- Understanding Agentic AI
- Definition: Agentic AI refers to AI that can predict, judge, and take action, effectively replacing human decision-making.
- Examples: Waymo as a notable illustration of agentic AI capabilities, showcasing the potential to replace human workflows.
- Impact on Industries and Labor
- Disruption of Knowledge Work: AI is expected to disrupt many sectors, particularly those that are less regulated.
- Creation of Opportunities: Although many jobs will be replaced, new jobs will emerge, especially in service sectors that AI cannot easily replicate.
- Investment Strategies in AI
- Key principles for investing include:
- Investing Ahead of Regulations: Targeting less regulated industries for quicker disruption.
- Focusing on Founders: Emphasizing founders' vision and capability over the initial idea.
- Thesis-Driven Investments: Shaping investments based on a defined vision of the future, particularly in agentic applications.
- Navigating the VC Landscape
- Emerging managers face potential extinction due to a saturated market.
- Successful venture management relies heavily on branding, marketing, and maintaining strong relationships with both founders and limited partners (LPs).
- The Future of Venture Capital
- The episode emphasizes that all asset classes and industries will be influenced by AI, necessitating a shift in how VCs assess and invest in companies.
- Discusses the shifting role of allocators and the potential for automating investment assessments.
- Key Philosophies and Learnings
- The importance of recognizing the ruthlessness often required in high-stakes entrepreneurship.
- Acknowledgment of the different profiles of engineers versus founders, with a focus on risk tolerance and the mindset necessary for disruptive innovation.
- Advice for Future Founders and Investors
- Founders should be conscious of their leverage dynamics with investors.
- Investors should be willing to embrace the complexity of working with transformative but potentially difficult personalities.
Key Takeaways
- AI's Evolution: The transition to agentic AI will reshape decision-making across industries, creating both disruption and new opportunities.
- Investment Focus: A clear vision for the future is essential for successful investments in the rapidly evolving AI landscape.
- Building Relationships: Maintaining strong networks with LPs and founders is critical for venture success.
- Ruthlessness in Leadership: The best founders may not be the most agreeable, but their relentless drive can lead to significant innovations and outcomes.
Conclusion David Weisburd's conversation with Camilo Acosta provides valuable insights into the evolving landscape of venture capital in light of agentic AI. As the industry prepares for significant changes by 2030, understanding the dynamics of investment, leadership, and market disruption becomes essential for future success.
Written by AI. May contain mistakes. Listen to the episode to check what was said.
Chapters
Tap a time to open that second in VOTransitioning from AI Engineer to VC
0:45 to 1:40
Camilo discusses how his experience at Meta influences his venture capital approach.
“I ran a$10 billion product suite at Meta, which sounds large, but in the grand scheme of Meta was actually not that big.”
Understanding Big Tech's Interest
1:40 to 2:40
Exploration of how big tech’s interests shape investment opportunities.
“A lot of AI outsiders, I think, were convinced that artificial intelligence was about building tools to help knowledge workers.”
The Future of Agentic AI
2:40 to 4:10
Discussion on the concept of agentic AI and its implications for industries.
“going to make things so efficient that more people will transact and there'll be more opportunities for business?”
Disruption vs. Job Creation
4:10 to 5:40
Camilo weighs the impact of AI on jobs and industry disruption.
“That's actually what led me to leave and start Perceptive.”
Disruption Frameworks and Regulation
5:40 to 7:20
Framework for assessing AI's near-term and long-term disruptive potential.
“They're becoming table stakes for the industry.”
Identifying Market Leaders
7:20 to 8:55
Insights on identifying firms that are ahead in the agentic AI space.
“A lot of founders, let's say most founders are optimizing for the next three to five years in whatever industry they're in.”
Investing in an AI Future
9:49 to 11:25
Camilo shares strategies for investing in a rapidly changing AI landscape.
“Do I just go in, grab my coffee, press play, and then leave and come back at the end of the day?”
Building a Successful VC Firm
11:25 to 13:05
Discussion on the essential elements of running a successful venture firm.
“Assessing different industries, assessing different managers in different sectors can be automated.”
The Importance of Marketing in Venture Capital
14:01 to 14:39
Learn why marketing and fundraising are crucial for venture managers to succeed.
“They've continued marketing of that success.”
Building a Strong Network as a VC
16:46 to 19:18
Understand the significance of having a strong network of LPs and founders.
“I think the key to being a successful venture manager is having a very strong network.”
Show all 16 chapters
The Role of Founders and Investing Philosophy
19:18 to 23:05
Explore how founder characteristics influence investment decisions and strategies.
“founders fund is certainly one, and Dries and Horowitz is another, but there's actually very few when you think about it.”
The Dual Nature of Successful Founders
23:05 to 26:58
Discover the complexities of working with highly effective, yet challenging founders.
“And that was a wake up call for me and for our team in terms of what it takes to build a generational business and how we look for those people and back them and the compromises you have to make.”
Future of Computing Interfaces
26:58 to 28:00
Discuss the evolution of computing interfaces and the shift toward audio-based interaction.
“Psychopathy is a clinical definition of not caring, having low empathy and all these things.”
The Shift Towards Audio Interfaces
28:00 to 28:48
Discover how audio input is becoming the preferred method of communication and its implications for technology.
“And part of the reason for this is that we developed the spoken word long before the written word.”
Career Reflections and Lessons Learned
28:49 to 29:44
Reflect on valuable career insights and the importance of learning from industry leaders.
“What is one piece of advice you could go back to 2006 when you had just graduated Princeton?”
Entrepreneurship vs. Corporate Culture
29:45 to 31:29
Explore the differences between entrepreneurial and corporate mindsets and their impact on management style.
“I worked at Jeffries for a summer, so I'm similar to you, very entrepreneurial.”
Transcript
Automatic transcript. May contain errors.0:00David Weisburd:So Camilo, you started at Meta where you were a top AI engineer, and now you run Perceptive Ventures, one of the top agentic AI seed funds in the world. Tell me about how your experience from Meta informs your day-to-day as a venture capitalist.
0:15Camilo Acosta:It's pretty critical. And before joining Meta, my company was acquired, so I was a startup founder and CEO. And both of those experiences are really critical to how I invest today. Having been a founder, I know how fast one can move and how nimble one can be. But having also been a big tech operator, I understand which large spaces big tech finds attractive and wants to move into. And the key here as an investor is making sure that we're not investing in spaces that big tech is going to dominate or be really interested in. And by the way, when I say big tech, I also mean open AI and Anthropic because at this point they have sort of like a neo-incumbent power in AI.
0:44David Weisburd:How do you know whether big tech, whether Meta, Facebook or Anthropic, open AI is going to go after space or not?
0:51Camilo Acosta:Understanding scale. I ran a$10 billion product suite at Meta, which sounds large, but in the grand scheme of Meta was actually not that big. And so there are$10 billion businesses that these large companies don't really have an interest in entering or pursuing because it doesn't really move the needle when you're a multi-trillion dollar company. That's too small for them. But a$10 billion startup is a very interesting outcome for a venture investor like myself.
1:12David Weisburd:If there's not top engineers on meta that are going to go after a problem, you think that's a safer place to build a startup and to build something that could be dominant there versus having to compete every day against these large LLMs?
1:23Camilo Acosta:The big problem with incumbents is that they have distribution power. So no matter how innovative your product may be as a startup and how forward thinking it is or well executed is, if you don't have the distribution power of these behemoths, it's quite dangerous to compete against them.
1:35David Weisburd:So when you started Perceptive early on in your Genesis, you made a big bet that the future of AI would be agentic. Why is that?
1:42Camilo Acosta:A lot of AI outsiders, I think, were convinced that artificial intelligence was about building tools to help knowledge workers. But the reality is that artificial intelligence is really centrally concerned with replacing human decision making. And that has three components at its core, prediction, judgment, and action. And these three components combined replace human decision making, which is the entire point of AI. So when you put them all together, it creates a completely agentic system. And the clearest example of that today for laymen is Waymo. When you get into a Waymo, the system is performing all three functions.
2:14Camilo Acosta:It predicts where the cars are going, where it is going, how objects in the environment such as pedestrians and bikes are behaving, and that applies judgment to those predictions. It says we should do X or Y. And then it takes action by actually moving the car in the necessary direction at the required speed based on those predictions and judgments. That's a fully agentic system. And that's exactly what's happening with software. Software is becoming agentic, which means it's not only eating software, but it's actually able to eat entire human workflows.
2:39David Weisburd:So do you fall in the camp where you believe AI will destruct human labor, or do you think it's going to make things so efficient that more people will transact and there'll be more opportunities for business?
2:49Camilo Acosta:I think it will disrupt and destroy a lot of industries for sure. I think it's also going to create a lot of new opportunities. And there's a lot of jobs that are also very difficult for AI to replace. We're always going to need restaurants, we're always going to need hotels, we're always going to need plumbers and electricians, we're going to need people are serviced, the millions and millions of robots that are going to exist in our world. So it's going to create new jobs, but it will certainly disrupt and replace a lot of human knowledge work, human labor.
3:17David Weisburd:Being on the inside and seeing all the disruption that's down the pipeline, what's your framework for figuring out what AI disrupts in the near-term future and over long-term?
3:27Camilo Acosta:In the near term, it is easier to disrupt things that are less regulated. And so the areas that will be slower to change are the ones that are more regulated. Spaces like the law or health care, where there are merchant guilds that protect those industries and accredit and license the individuals that work in those industries, working in concert with government to license those individuals. Those things are going to be a lot slower to change, but they will eventually, as safety is proven not to be better with AI, as efficacy is proven not to be better with AI. But it's just going to be a lot slower.
3:58David Weisburd:Last time we chatted, you said that there's a couple of players in the market that saw this agentic future years ago and they have a head start. Tell me about these companies.
4:07Camilo Acosta:Working on AI internally at Meta, we knew where things were going. That's actually what led me to leave and start Perceptive. We saw that future coming because we were part of the group building. Our vision at the time, even if we were to pull people internally at the company though, would have seemed crazy. People broadly in the company might have thought that was pretty futuristic. But those of us working on these topics realized it wasn't that crazy because the models were getting there. So there is a head start that some of these big incumbents have, but it doesn't mean that this is the end all and be all.
4:33Camilo Acosta:We are in the early innings of LLMs, and we don't fully know if LLMs are the right architecture for developing AGI or superintelligence. So it's still a question mark if even this is the right path long term. It could be some other new company comes about and develops a new architecture that it's actually more efficient and better for developing AGI, TBD. But certainly when it comes to the LLM landscape and the the big platform models, there is certainly a head start that these big companies have.
5:01David Weisburd:Seed stage venture capitalists in general have a difficult job of predicting the future 10, 12, 14 years from now. Today, that's even more difficult with the pace of AI. How do you create a framework investing 2026 for 2036 and beyond? One of the hardest things of investing is seeing what's shifting before everyone else does. For decades, only the largest hedge funds could afford extensive channel research programs to spot inflection points before earnings and to stay ahead of consensus. Meanwhile, smaller funds have been forced to cobble together ad hoc channel intelligence or rely on stale reports from sell-side shops.
5:38David Weisburd:But channel checks are no longer a luxury. They're becoming table stakes for the industry. The challenges have always been scale, speed, and consistency. That's where AlphaSense comes in. AlphaSense is redefining channel research instead of static point-in-time reports, AlphaSense channel checks delivers a continuously refreshed view of demand, pricing, and competitive dynamics powered by interviews with real operators, suppliers, distributors, and channel partners across the value chain. Thousands of consistent channel conversations every month deliver clean, comparable signals, helping investors spot inflection points weeks before they show up in earnings or consensus estimates.
6:15David Weisburd:The best part, These proprietary channel checks integrate directly into AlphaSense's research platform, trusted by 75 % of the world's top hedge funds, with access to over 500 million premium sources. From company filings and brokerage research, to news, trade journals, and more than 240 ,000 expert call transcripts. That context turns raw signal into conviction. The first to see wins, the rest follow. Check it out for yourself at alpha-sense.com slash how I invest.
6:43Camilo Acosta:We try to think a lot about what the world is going to look like and specific industries are going to look like and invest towards that vision of the future. We have a very thesis driven. And so we steer away from what seemed like near term improvements and innovation. We've historically never invested in copilots and things like that because we didn't believe that copilots were the end goal. And so we try to think about, well, what is an agentic future for health care look like? What does agentic future for legal look like? What does agentic future look like for construction, et cetera, et cetera, et cetera?
7:12Camilo Acosta:and work our way backwards from that in order to say like this, this company fits that mold, but more specifically this founder fits that mold because we have to see that that founder gets that future. A lot of founders, let's say most founders are optimizing for the next three to five years in whatever industry they're in. And that's how far out they think. And they, you know, they pitch you like, Hey, I'm going to get to series A. I don't care how you get to series A. I'm concerned about how you're going to build a hundred billion dollar company.
7:35David Weisburd:How do you marry those two things, which is I need to be pragmatic enough to attract the next round of funding while also So not being leapfrogged by the next generation of technology.
7:44Camilo Acosta:I would say it's a little controversial to say out loud, but I would say that the tier one investors in venture across stages, whether it's Seed, Series A, B, C, are very forward thinking. They are investing in very disruptive innovation, whether it's Kozla or Founders Fund or Sequoia, that's how they think. And so there is an element of just being de-risked because we're aligned in how we think with some of those bigger players.
8:06David Weisburd:And these investors are investing at the Series A pre-revenue. basically they realize that gap between the future and the current state, and they're willing to subsidize even from a revenue basis. They're looking for something fundamentally different in terms of milestones.
8:19Camilo Acosta:Yes, it does. It depends on the industry, right? So in B2B, we've historically seen in the last couple of years that it is very easy and quick for AI companies to get to revenue quickly. So by the time they get to series A, they're doing substantial amounts of revenue. So from that perspective, they're de-risked. But also you can look at Kozla being the only venture investor at OpenAI back in like 2017, 2018. and something like that. When OpenAI, it was pre-revenue then, and it certainly still bleeds cash today. So there is their nomin of both, being that visionary willing to back pre-revenue, but also there's a lot of companies that are getting revenue very quickly these days based on just AI.
8:52David Weisburd:You're on the bleeding edge of Agentec AI as an investor. How much do you build a thesis around different parts of the market, and how much do you let your founders draw you into what they see as the future?
9:03Camilo Acosta:A mix of both. Sometimes there are industries that we haven't sat down and thought about and fleshed out in terms of what we think that future looks like. And so we can chat with the founder about what they think is gonna happen. And then we try to sit down and say like, all right, does that make sense to us based on what we see happening in other industries as well? But it is often that we have already come to a place of understanding or having a general rough sketch of what we think the future is gonna be for that space. It's interesting as a former founder, like you're just stuck in the weeds pretty often.
9:33Camilo Acosta:And so it's often hard to see the bigger picture. So rare is the founder that really understands what the bigger picture looks like and how they're going to get there.
9:42David Weisburd:A lot of people have trouble conceptualizing an agentic AI future. How does that work? Let's say I'm a business owner. Do I just go in, grab my coffee, press play, and then leave and come back at the end of the day?
9:55Camilo Acosta:You know, what's crazy is there's a world where AI is the business owner. You know, I've talked to really visionary founders who are working on this. You know, what is a world where AI owns property, where AI has legal rights. And if so, then they are the business owner. They can spin up businesses, which is pretty wild. Until we get to that sort of regulatory regime. Yes, I do think there is a world where, and I think it's even here now already where you see small business owners that can press a few buttons and run a large operation because they've automated so much of it. It's pretty awesome to see.
10:28Camilo Acosta:The question again becomes like, at what point do they become replaced by an AI that owns the business?
10:34David Weisburd:And I subscribe to this agentic AI future. I think it's more or less obvious if you think from first principles. What's not clear to me is a second order effects. So if I'm an investor, I'm running an endowment, how should I think about investing into other asset classes, assuming that an AI agentic future is imminent?
10:53Camilo Acosta:Every asset class is going to be touched by AI private equity. There were there's been some large titans of private equity who were anti-AI a year ago now, but they're publicly trying to defend their portfolios because they're scared that the companies they've invested in are not going to survive the AI revolution. Every asset class is being touched by this. Trying to say like think of what might not be maybe commodities, but but even then commodities, there's a lot of the AI boom has driven a need for certain kinds of commodities. So it's even impacting that space in different ways. And I think the job of allocators of LPs is going to change because the Yale model, the Yale Endowment model of investing, even that could be automated.
11:28Camilo Acosta:Assessing different industries, assessing different managers in different sectors can be automated. So it's, it's really going to touch absolutely everything. Even when we think it's not.
11:38David Weisburd:How do you think about AI enabled businesses versus AI first businesses in the non-venture space?
11:44Camilo Acosta:It's hard to say, uh, then anything is not going to be touched by AI. So even think like you mentioned widgets. So I like to think about industrial, like legacy industry kind of businesses, petrochemicals, you name it. like really old school stuff that requires machinery to operate. That's also getting automated. We see we get pitched ideas in those really obscure industrial areas that you would think of. You just buy that, makes the widgets go faster, go, you know, run with less people, the whole operation with less people. All of that is also getting automated. So even if you're a lower middle market PE firm that does that, and you've actually started to see this, is some of those PE firms are starting to hire AI engineers or AI technologists to look at their portfolio and say, like, all right, how is AI going to disrupt these companies?
12:29Camilo Acosta:Like, that's the thing that people forget is like, you think that just because you're not doing it, it's not going to happen. But the reality is that someone else is going to try it. And if they're successful, they're going to beat you. So that's how we see it. And we see what's happening.
12:39David Weisburd:And by some accounts, emerging managers in VC, there's an extinction level event happening where anywhere from 50 to 75, some predict up to 90 % of emerging managers are on their last fund and just won't be able to raise any more funds. How are you surviving this extension level event and how are you preparing yourself for the next era of VC?
13:02Camilo Acosta:One of the reasons it's occurring is because too many venture managers came out of the pandemic high when anyone with a pulse in a deck could raise a fund, frankly. And it wasn't necessarily people who should have, to be honest. There's a lot of skills required in order to raise a fund, build a firm.
13:21David Weisburd:And you made that distinction. A lot of funds don't know the difference between running a fund and running a firm. How are you building your firm? And what lessons have you learned from building a firm?
Read the full transcript
13:32Camilo Acosta:A firm is a brand. And most people don't realize that to be a successful venture manager, you have to be consistently marketing that brand to founders and to LPs, founders and LPs constantly. They think that it's just a function of having deal flow and making the investment. And it's not. The only reason your firm exists is because you have been able to bring LPs into your capital base and stewarded that capital well and continue to market your performance and your ability to continue doing that over and over and over again. Venture, if you look at the studies about successful venture firms starting all the way back to Sequoia, etc., they are successful today because of the success they had in the past.
14:09Camilo Acosta:They've continued marketing of that success. and a lot of venture managers don't like marketing and they don't like fundraising and they don't like investor relations, but that's the key to building a successful firm.
14:22David Weisburd:You have some enviable LPs, especially earlier on in your firm building. How are you able to secure those LPs and what are some best practices for other emerging managers that want to bring in blue chip LPs? When you want more, you start your business with Northwest Registered Agent. They give you access to thousands of free guides, tools, and legal forms to help you launch and protect your business all in one place. With Northwest, you're not just forming an LLC, you're building your complete business identity. From what customers see to what they don't see, like operating agreements, meeting minutes, and compliance paperwork, you get more privacy, more guidance, and more free resources to grow the right way.
14:59David Weisburd:Northwest has been helping founders and entrepreneurs for nearly 30 years. They're the largest registered agent LLC service in the US with over 1500 corporate guides, real people who know your local laws and help you every step of the way. What I love is how fast you can build your business identity with their free resources. You can access thousands of forms, step-by-step guides, and even lawyer-drafted operating agreements and bylaws without even creating an account. Northwest makes life easy for business owners. They don't just help you from your company. They give you the tools you need after you form it.
15:29David Weisburd:And with Northwest, privacy is automatic. They never sell your data because privacy by default is their pledge. Don't pay hundreds or thousands of dollars for what you could get from Northwest for free. Visit northwestregisteredagent.com slash invest free and start using free resources to build something amazing. Get more with Northwest Registered Agent at northwestregisteredagent.com slash invest free. Let's be honest. Subscriptions out of fast. Streaming services, apps, memberships you forgot you even signed up for, and canceling them is usually a pain. That's where Experian subscription cancellation comes in.
16:07David Weisburd:Experian can take the pain out of canceling subscriptions by handling it for you. You just keep the ones you want and put money back in your pocket. Over 200 subscriptions are cancelable. You can also save money by letting Experian negotiate the rates on your bills. They'll keep an eye out for new deals and saving opportunities and negotiate directly with your provider on your behalf. And the best part, you keep 100 % of your savings. Get started with Experian app today. Results will vary. Not all bills or subscriptions are eligible. Savings not guaranteed. Paid memberships with a connected payment account required.
16:44David Weisburd:See Experian.com for details.
16:46Camilo Acosta:I think the key to being a successful venture manager is having a very strong network. And people think that that just means a founder network. I think that's half of it. And the other half is having a strong LP network. Again, if you don't have the money, you're not an investor. So you have to have a strong LP network to begin with. It just happened in terms of how my life came together, my career, that I did know a lot of LPs and I had raised money as a founder. So I already knew how to do that. So I already had the network built in to go out, tap it and start my firm. And from there, word spreads.
17:14Camilo Acosta:LPs are a very insular, talkative group of people. They like to share their deal flow in terms of venture managers that they're finding. And that's how my LP base was able to grow is through that word of mouth with these investors.
17:27David Weisburd:Talk to me about these AI tools. What AI are you using internally that gives you an edge over your competitors?
17:32Camilo Acosta:I believe that you can encode and build models of what a successful founder looks like in B2B startups and then also in consumer startups, because I think they're actually different. And so that psychographic model, I think, is key. And I think it's something that's repeatable and scalable. If you look at some of the character traits that really define the best founders of the history of technology from Bill Gates, Steve Jobs, Mark Zuckerberg, et cetera, et cetera, they'll share common things. And so I think that you can actually model this and use that model just like any other AI model to say this is the right founder to bet on versus using just gut instinct, which is what a lot of precedency venture investing is a lot of gut instinct.
18:09Camilo Acosta:But gut instinct is just data. It's just accumulated data in your mind and body that I think can be actually modeled.
18:18David Weisburd:It goes without saying AI is a hyper-competitive seed market. How do you compete against other firms? And also, how do you compete? You mentioned the Kostela, Sequoia Founders Fund. How do you compete against these multi-stage platforms as well?
18:30Camilo Acosta:We are of the size where we don't have to compete with the big boys. We can collaborate with them, given our check size. And so we're in deals with a lot of the big guys, Founders Fund, Kostela, et cetera.
18:38David Weisburd:So this is due to size. Where we do compete is the other pre-seed and seed firms.
18:43Camilo Acosta:And where we win is because we are exited AI founders. We've been down the road as founders before, and in particular in AI. And so it's a very, I don't want to say it's easy, but it's a pretty straightforward argument to make to any founder. Like, we've just been there before. What can help you with the challenges you're facing today and will face tomorrow from inception all the way through exit? And so that's what really has been the key for winning for us, because most VCs are run by career investors who've never built anything before. And then the firms that do have some level of operators, they tend to be operators.
19:15Camilo Acosta:They're not founders. Very few firms have actually been built by former founders. founders fund is certainly one, and Dries and Horowitz is another, but there's actually very few when you think about it.
19:25David Weisburd:Tell me about some of the biggest misses you made as a VC and what were the learnings from those mistakes? The most consistent logical fallacy or mistake that I see from smart VCs, and this is like a very predictable at this point, is that they really like the founder and they really hate the business and they don't invest. And they're right about both, but their decision not to invest is wrong. Right.
19:46Camilo Acosta:And what happens when you do make that decision or to invest or not invest is you're saying like there has to be a pivot and I'm investing with the hopes that this team will pivot. Sometimes you feel so strongly about a team that you say I'm going to invest and they're going to figure it out and I don't believe in this thing that they're doing right now that they're going to figure it out. That's how we got our first check. When I was a founder and CEO, we got our first check and our investor told us, they're like, we have no idea what you're doing or where this is going, but we really like you as a team.
20:12Camilo Acosta:So here's a check. And that's the kind of bold investing you need to do. And then we've done, We've invested pre-idea. I met a founder who had amazing two or three exits up to his belt. And he was like, I don't know what I'm gonna do next. And I was like, take my money. I don't know what you're gonna do either, but you're smart enough to figure it out. So sometimes you just get to get better.
20:30David Weisburd:Would you rather invest pre-idea or in the wrong business model? Both are fine.
20:34Camilo Acosta:I don't think there's an either or. Again, if it's the right team, it's the right team. You have to believe that they're willing to move quickly. Because sometimes you can have a really smart team, but they don't iterate fast enough. if they don't, they get caught up in sunk costs and say like, oh, we've already built all of this. Like, do we really want to pivot to something else? You know, you need to invest in someone who's willing to let all that go and be ruthless with their time.
20:55David Weisburd:Talked about AGI earlier. Do you lose sleep on this doomsday AI situation where there's an extinction level event, not in the venture ecosystem, but humanity in general?
21:07Camilo Acosta:I don't lose sleep over, but I do think it's real. I think we have to think about it. And And it's unfortunately true that we have to balance it with national security concerns. So there's often this tension in the U.S. at least of building for safety and building for speed. You can, you know, dumb it down and say some throw papers open AI. But the reality is, is that we don't build it. China will.
21:29David Weisburd:So why even think about it? The decision is we have to move forward.
21:33Camilo Acosta:There is that tension. Right.
21:38David Weisburd:What's a key philosophy that you've changed over the last 12 months that you now, that leads you to make different actions today?
21:43Camilo Acosta:One of the key things I've come to accept is that some of the best founders in the world are not necessarily always the best people to work for or to work with. There is an element of ruthlessness that's required in building generational companies to transform humanity. It's a sacrifice they are making personally, but it's a sacrifice they ask their employees to make as well and a sacrifice they ask investors to make as well. And so there there's often a desire to do business with people you like, right? Like it's a common maximum in the business. And the reality in venture is that sometimes you want to do business with people you don't like because they are the best, the most extraordinary, impactful people in history.
22:21David Weisburd:Reminds me of a story you told me when we were having dinner a couple of months ago. Tell me about that.
22:25Camilo Acosta:Yes. So we, it's an interesting story. So we were early in a company and the founder did behave in a certain way that was He was not very agreeable. And I'm trying to be careful how much I say here. He was not very agreeable. And when we talked about it in IC, one of my partners said, hey, do we want to work with this guy? He seems like an asshole. He's doing kind of assholy things. And I said, actually, I want to work with him even more now because he can be an asshole to me, despite all the things I've done for him. That makes me realize this guy is going to do whatever it takes to build this company.
23:05Camilo Acosta:And that was a wake up call for me and for our team in terms of what it takes to build a generational business and how we look for those people and back them and the compromises you have to make.
23:15David Weisburd:How would you explain that? Because there's obviously the opposite is also an axiom, compounding relationships, compounding reputation. But here specifically, it's almost a zero sum mentality to relationships that was a superpower. How would you explain the mechanics around that?
23:29Camilo Acosta:there's at the core a level of humility you have to have because when someone's being disagreeable or an asshole there's an element of like i'm offended it's my ego that's saying like i i'm hurt i don't want to be hurt i don't want to continue interacting with someone who's hurting me so it takes a stepping out of that and saying that doesn't matter what matters here is as a venture manager i'm here to make money for my lps and is this the right person to make money for my So it's taking yourself out of the framework that you often are in, in terms of being led by ego and about pain and whatever and hurt and saying like, what actually matters?
24:05David Weisburd:Is that a specific case where you had to just burn through all your relationships in two, three years because there's a short window? Or is there some general wisdom there for highly disruptive startup founders? At the end of the day, the best founders in the world have all the leverage.
24:18Camilo Acosta:They can choose their investors. And so they will have leverage in that dynamic always. is once you come in, they still will have leverage over you, even if no matter how big your position is in the company, because they can always raise more from someone else. Right. So I don't think from a founder perspective, like you just have to become conscious of that. That's the dynamic now and that's okay. I'm not here to be in the spotlight. I'm here to support founders. I like how Vinod Kozla says he's not a VC, he's a Venture Assistant. He helps founders, he assists them in a service provider. That is the role.
24:49Camilo Acosta:And so you have to own that and let this ego of the VC, like push that aside. And it doesn't mean that your relationships are fraught. It just means that you have to be careful with those people and people who worked. I'm fortunate enough to know people who worked with Steve Jobs or were on board with Steve Jobs and they all knew like, you gotta be careful with him. Doesn't mean that you can't have a relationship with him. It doesn't mean that he's not extraordinary. It just means you just have to be careful. Yeah.
25:12David Weisburd:Would that be the case if you were a customer of that entrepreneur? Clearly that would be counterproductive to scaling, but it's really that you're a commodity to that entrepreneur. so he didn't find the need to use niceties and waste time with you as a commodity versus if he was trying to secure a large product or a large, large contract, that would be a necessary condition for him to succeed. Yeah.
25:33Camilo Acosta:And, and again, these founders, it's not that they're perma assholes. Like they, they, they are, they're quite charismatic, very friendly people when the moment is right. And so they know how to turn on the charm when it comes to sales, when it comes to customers, clients, et cetera. And that's, that's their magic is that they can be both. Now, when I look at my portfolio and we look at our top companies, top founders, unicorn founders we've backed, they'll have that ability to from one second go ruthless to the next second be like the nicest guy in the world. And so they know that they, in order to build a massive company, they're going to have to be the nicest guy in the world
26:03David Weisburd:where they're outpitching.
26:06Camilo Acosta:But when push comes to shove and there is a moment that requires decisiveness, they're not afraid to pull out the ruthless side of them.
26:12David Weisburd:Yeah. If you think about your energy and being nice as being highly energy consuming, they have an efficient use of their energy when it comes to different parts of the value chain. Of course, that sounds psychopathic, but when you take away the moral frame on it and you look at what's going to lead to, if you're trying to create an entrepreneur from scratch, that's going to execute at the high speed with the fewest amount of resources, that's almost like the perfect equation.
26:35Camilo Acosta:Yep, exactly. And I think we like to judge, even using the word psychopathic, it's like, oh man, do we want psychopathic founders? But what if that's just simply what it takes to build transformational global companies? It's the level of psychopathy.
26:52David Weisburd:You could be 100 % psychopathic and provide massive value to society. I'm not going to name any founders. Psychopathy is a clinical definition of not caring, having low empathy and all these things. You could literally be clinically psychopathic. And if you have the right incentives, they could end up actually carrying society forward. And you could say that's wrong, that's bad, but on a net basis, they're actually very positive for society.
27:12Camilo Acosta:We need them. Exactly. Yep.
27:16David Weisburd:Last time we chatted, you said that whoever controls the computing platform controls the future. What form factor do you expect the future of compute to take form?
27:26Camilo Acosta:Let's talk about what today's interface model looks like in these computing platforms. So we primarily use two interfaces. There's the computer, which we're chatting on a computer right now. Most knowledge workers sit at a desk looking at a monitor or a laptop. And this is interface model number one, let's call it. The second interface model is the phone. And I actually argue this is probably the most important one because most people on the planet have a smartphone, but most people on the planet do not have a desktop. And the GUI, the graphical user interface, that we use on laptops and monitors and phones, it's been around for 40 years and it's entirely text-based.
27:53Camilo Acosta:Even though it's visual, I still have to read the written word on my calendar, on my emails, websites. Reading is actually not that fast. Like it's not necessarily the most efficient way to process information. And part of the reason for this is that we developed the spoken word long before the written word. We're wired for speaking, which is why we're having this conversation by talking instead of writing back and forth to one another. So that's a fundamental principle behind these interaction models. And again, the phone follows that logic, but I think the phone is going to be the first one that breaks because it's an even more inefficient interface.
28:18Camilo Acosta:People often check their email on their phone and then wait to respond on their computer because typing on a phone is so cumbersome. And we're already seeing early signs of that, that audio is faster and preferable. You see people using the dictation function to write emails and SMS on their phone. That's audio. That's a spoken word. So the input is becoming more and more audio based. So I think that's what we're going to see right first. And that's going to be the platform shift that matters the most. And we've already started to see it with Humane, which came and went. But now OpenAI is going to release its own audio first device.
28:44Camilo Acosta:And while it may have a visual component, again, it's going to be primarily an audio interface.
28:49David Weisburd:What is one piece of advice you could go back to 2006 when you had just graduated Princeton? One timeless piece of advice that would have either helped you accelerate your career or helped you avoid causing mistakes?
28:59Camilo Acosta:One thing that I wish I had done was actually work at a big tech company earlier in my career, before I was a founder. When you're a founder and you've never worked anywhere before in tech, you're really flying blind. But when you work at high talent density places like Google or Meta, you understand what the bar is across a variety of functions. So it wasn't until I got to Meta that I really understood what world-class design looks like, world-class product, world-class engineering. I thought I knew as a founder, but it really wasn't until I got there that I understood the processes and systems that the best engineers in the world, the best designers in the world, the best product people in the world use to get the job done.
29:33Camilo Acosta:And so if I was going back and talking to the younger version of myself, I would have said, go learn there, go learn from them, and then go out and build a company so you're not making a lot of mistakes.
29:44David Weisburd:And I've only worked for three months in my life for somebody else. I worked at Jeffries for a summer, so I'm similar to you, very entrepreneurial. And when I think about these meta engineers and these Google engineers, are they the same level of talent at the high end as the top founders, and they just have a different risk appetite? or are founders just a different class of engineers, designers, et cetera?
30:05Camilo Acosta:I think it's just a risk appetite question. And going back to the psychopathy conversation, there's a different profile there. But in terms of -
30:14David Weisburd:They care too much about being liked?
30:16Camilo Acosta:They care too much about being liked. There's, yeah, they're not the extreme of that psychographic profile that our founder -
30:23David Weisburd:Not enough trauma. Not enough trauma necessarily.
30:27Camilo Acosta:So I think those, which I think are the, frankly, the key factors because there are tons and tons of intelligent, brilliant people in the world. But how many of them have the right, let's call it the right trauma, the right wiring, the right psychopathy, the right, et cetera, et cetera, et cetera, to be a generational founder that transforms the world?
30:40David Weisburd:To play devil's advocate on you wish you would have went to Meta or Google earlier, you never know how that plays out. You could have been a lifer there. And I also, I started my career as an entrepreneur, and it's embarrassing how ignorant I was at the time. But yet I developed the skill set of being an entrepreneur in an unknown environment that if I had to pick, I'd go with that skill set versus the hard skill set.
31:04Camilo Acosta:The skill set of a founder doesn't translate very well to corporate America is what I found. When I got to Meta, one of the first things I told my manager was, here's a list of a dozen people I think we should fire. And she just laughed at me. Was that day two? This was in like the first week. Day one. I quickly could tell who was pulling their weight and whatnot. And she just laughed at me and she said, hey, this is not how things work in corporate land. And we're going to have to work with these people, even if they're not great to work with in terms of their efficacy. And so I have a strong bias for fashion as a founder.
31:36Camilo Acosta:And that doesn't always translate very well to corporate land where there's systems and processes that slow you down. And you can't just fire and hire the team that you want. So I think it really depends on what you're optimizing for.
31:47David Weisburd:Kamilo, this has been an absolute masterclass. Thanks so much for jumping on the podcast. Thanks for having me. That's it for today's episode of How to Invest. If this conversation gave you new insights or ideas, do me a quick favor. Share with one person in your network who'd find it valuable or leave a short review wherever you listen. This helps more investors discover the show and keeps us bringing you these conversations week after week. Thank you for your continued support.
From the publisher
What happens when AI stops assisting humans and starts replacing decision-making itself?
David Weisburd speaks with Camilo Acosta about the rise of agentic AI, why incumbents still leave massive openings for startups, and how AI will reshape labor, venture capital, and entire asset classes. Camilo explains how investing ahead of regulation, betting on founders over ideas, and building for a fully agentic future define the next era of venture.




