Michael Eisenberg: How China Could Overtake the US in the AI Race | E1167

Michael Eisenberg 0:00 Foundation models are the fastest appreciating asset in history. Why invest in a business where the assets walk out at night? If I'm in Europe and I'm starting a company, I would get out. Lina Khanos, I think, openly is, I think, a threat to American capitalism. Whether Biden or Trump gets elected, she needs to go. I don't buy that the IPO window is closed. The IPO window is wide open. The question is, what's the price you're willing to take?

Harry Stebbings 0:20 Ready to go? Michael, I'm so excited. Dude, we've been friends for years, and this is the first time that we get to do this in person. So, thank you for joining me.

Michael Eisenberg 0:39 Thank you for having me. The first time we get to do the podcast in person, but not the first time we meet in person.

Harry Stebbings 0:43 Uh, no, no, no, no, no. The first time podcast in person. But, you know, it's so funny. I'm sure you find this actually, but I was saying the other day on Twitter that actually it makes such a difference doing, um, in-person shows.

Michael Eisenberg 0:53 I totally agree with you.

Harry Stebbings 0:54 Really, for me, it's a game-changer. N-

Michael Eisenberg 0:55 Thank

Michael Eisenberg 0:56 you for having me.

Harry Stebbings 0:56 Not at all. But I want to dive straight in, because it seems like we're in this kind of bizarre AI bubble now, and we have one side of the table that says, "No, this is the most transformational technology that we've seen in 20 years," and we have another side of the table that says, "This is completely nuts. The valuations are crazy. The amount of cash that's being burnt is crazy. This is peak." How do you just think about that from a starting point, before we do a comparison?

Michael Eisenberg 1:19 One of the hardest things to do in the venture business is hold ... And in life, by the way, is to hold two truths in your head at the same time. I think both of those things are true. This is the most transformational technology I think I will ever see in my lifetime. Um, and at the same time, uh, there is, like happens with many transformational technologies, just an incredible gold rush that brings in a lot of people. Both these things are true. You know, when the internet came on, uh, and I was starting my venture career, this is in 1995, it was, it was incredible. I mentioned to you before we came on air, the first deal I did was called Picture Vision, and it was an online photo sharing, when there were still cameras with film and negatives. You had to actually scan the negatives, uh, in order to send the pictures. You'd upload them. It looked like paint drying, uh, to put- upload, uh, upload the pictures. And then there was pets.com and pets.com.com and pets.com.com.com, because it was an incredible gold rush. But the internet, eBay, Amazon, et cetera, turned out to be incredible companies. And I think the same will be true about AI, which is, by the way, a giant bucket that we need to kind of break apart. And then, you know, right after the dot-com thing, people have forgotten, but there was like a fiber optic craze. Everyone's forgotten it, 'cause it's called the dot-com crash. But there was a fiber ac- uh, optic equipment and infrastructure craze. And we laid a ton of fiber optic cables, and it was a bubble, and the stocks went crazy. But we live off that infrastructure today. And that's what's made the current moment possible. Same thing happened, you know, uh, Cisco bought, um, blank on the name of the company for $6 billion in the fiber optic routing space in, I think, 2001. Then that bob- bubble burst, and, you know, it took a bunch of years to kind of find our footing again. So yes, AI is the most transformational technology I think I'll ever see in my lifetime. True. It's also true that there's a massive financial gold rush going on in what turns into a bubble. True. What's also true is those bubbles create the infrastructure that we need going forward, and that's also true. So, a lot of people are gonna lose a lot of money during this, like a lot of money. Um, and some foundational companies and technologies will be built.

Harry Stebbings 3:15 Who are those people who lose a lot of money? I'm not asking for the names of those people, but what sort of styles, what behaviors, what categories is the ones that will lead to that loss versus value creation?

Michael Eisenberg 3:24 Let's start with a non-obvious point, which is that there's a lot of LPs with a lot of exposure to a lot of the same kind of companies through multiple different funds, and they think they have p- portfolio diversification across a significant number of venture capital funds, but everybody's piled into the same trend, and in many cases, piled into the same company. C-

Harry Stebbings 3:42 Do you think LPs are aware of that?

Michael Eisenberg 3:44 I don't know. I don't know. I think the best ones probably are, and others probably aren't. There's a second kind of category, which I call logo chasing. This happens in every one of these cycles, where I got to be in the hot one. And there could be five or six or seven hot ones, and I need to have that logo. Why? Because I want to tell people I'm in OpenAI, I'm in Character.AI, I'm in Diffu- uh, I'm in all these things. So, uh, I need to be in them. Uh, then I think there's a third category o- of people who, uh, think you can replicate a model that exists in one geography and make it in another geography and it's gonna work to the same scale. And they pay up a high price, but it doesn't work in those geographies to the same scale. I think that's kind of a third category of people. And then I think it's important to say, you know, uh, our mutual friend, Gavin Baker of Etridus has this line, which I completely agreement, agree with, that foundation models are the fastest appreciating asset in history. And I think that is absolutely true. I was talking to our founder of, uh, Seeking Alpha, which is a company by the way I did 19 years ago on Benchmark, and I'm still on the board of. Uh, and he made a super interesting observation, which is that financials or stock markets, which is a minute-to-minute update business, or, or, or set of data, doesn't do well with a lot of these training models. So, you can predict like a lot of things and even write an analyst report, but it doesn't matter 'cause the numbers are off 'cause they changed two minutes ago. And I think how we deal with also the refreshing of information, which is getting, uh, to a higher velocity, specifically in areas like financial and stocks, but, you know, that's gonna be another area that's super interesting.

Harry Stebbings 5:12 S- so we actually just released a show today with Aravind, the founder of Perplexity.

Michael Eisenberg 5:15 Yeah.

Harry Stebbings 5:16 And I said to him, you know, about the depreciating asset that is these models and, and what that means for their, you know, aligned price. And he said, "You don't get it. The price is actually aligned to the teams that are able to bring these models together." That is where the value in these companies are. That is why OpenAI is worth $90 billion. It's not the model itself. It's the collection of specific people with specific skills that are able to bring them together that is worth the value.

Michael Eisenberg 5:41 Perplexity is my favorite app. It's on the home screen of my phone. And still, last I checked, indentured servitude was outlawed in the United States.

Harry Stebbings 5:49 (laughs)

Michael Eisenberg 5:49 And, uh, a little bit, it's like what Bill Gates' father said to him, that everyone's forgotten. Why invest in a business where the assets walk out at night? So, it's true that this is a unique talent business.... but if you don't have a lasting, uh, kind of asset, I do believe that software in that case, of Microsoft was a lasting asset. The data is not. And the models, as we're starting to see, uh, in various ways, people are walking out the door and some of them didn't work, right? There was Mustapha's company that was picked apart for its, for its bones and flesh by Microsoft, and a very, very elegant exit. Super elegant. And I think we'll see more of those. But I don't buy that five, six, seven, eight, nine, 10 of these are gonna work. It's just not true.

Michael Eisenberg 6:31 Financially it's not true.

Harry Stebbings 6:32 I, I think that you're gonna see all the large cloud providers, Google, Amazon, you name it, essentially say, "Hey, we need to acquire the foundation models, your co-hack, your Coheres, the smaller ones who can be acquired." Uh, and then you're gonna see a battle for the crown between Anthropic and OpenAI. And so large cloud providers acquire kind of smaller providers, and you have a battle there 'cause they're too big to be acquired. Do you agree with that summary?

Michael Eisenberg 6:55 Could be. Um, I think about it a little differently, which is where do people who use, uh, AI start their day? It's a little bit like the Google question, because I think as you kind of play this forward, you're gonna see a whole series of agents. Like, I'm not gonna go to the NewYorkTimes.com to read it. I'm gonna go to my AI screen and say, "Fetch me the articles that interest me from The New York Times today," and something that, you know, entertains me or is different or something I gotta know. Uh, I think where people start their day is what's important, and then there'll be kind of an underlying layer that gets accessed by APIs, uh, which feels like Anthropic is, is ahead on right now. Um, which will service a lot of these companies. And a lot of this will be built into, we talked about the hyperscaler, uh, stacks, um-

Harry Stebbings 7:40 Do you think these are even businesses, though, for venture? They are so cash consumptive. We also have, um, irrational financial buyers and investors who are obviously large corporates, who have different motives to venture investors. Do you think foundation models are even an asset class that venture investors can make money in?

Michael Eisenberg 7:57 Yes. I don't think we should think about anything in the venture or startup business as an asset class. There are asset classes out there, like commercial real estate, you can make money in a class. The venture business is completely sui generis. It's not a class. There is a small number of people in this craft who make money, and there's a small number of startups. You know, the S&P 500 has kind of kept going up on the back of four companies who are venture-backed. And these are incredible generational companies that have pushed forward. The rest of them are kind of, you know, trailing behind. And we just need to focus on the question of, are all of these just incredible companies that will land their, themselves into some index and some fortune, uh, some S&P 500 that works? I think the answer is no. And so, this is not a class of company's foundation models. There is, there are one or two companies in there that will make their venture investors a lot of money, and the rest of them will cost their LPs and GPs a lot of money. It's just life.

Harry Stebbings 9:00 (laughs)

Michael Eisenberg 9:00 It's what it is in this business.

Harry Stebbings 9:02 No, I love having you on the show. It's always my favorite chat. Uh, so I get you and I, I agree in many respects. I do just want to break down then, when you look at AI, you said like, "We put it in a bucket," I'm like, what does that actually even mean? How should we break down that bucket? And how do you break down that bucket when you think about investing in it today?

Michael Eisenberg 9:19 So look, we invest only in Israel, and we've had a theory. You know, we should back up one step, which is AI, AI, AI. The real moment is the LLM moment, which is, AI has been around for a while. Like in 2013, we invested in a company called Windward, which is now traded in London, which does maritime AI. That's just since 2013. 2015, I invested in Lemonade, which was AI for insurance. We said bots not brokers, before bots were fancy. Okay? What's happened in the last two years or so is the emergence of the LLM, which is yet another leg up, uh, or different way to kind of prosecute, uh, AI. AI itself is not new. When you think about that for a second, the bucket we look at, we invest only in Israel. We ask, "Can Israel do foundation models?" We said no, and we've said that for a very long time. Why? 'Cause a lot of this comes out of academia, comes out of a lot of these very large companies. We didn't think there was a talent density in Israel specifically for foundation models. Uh, because a lot of the Israeli technology comes out of the military, not academia. You know, why is France so good all of a sudden at AI? 'Cause of academia.

Harry Stebbings 10:22 Right.

Michael Eisenberg 10:23 And what we said was, what we want to do is applied AI. And what we want to do is kind of full stack string through the system of AI.

Harry Stebbings 10:32 How do you think about industries which are ripe for AI to be fully efficient in terms of every process from start to finish, or as much as possible, versus those that are not actually and cannot be verticalized in that way?

Michael Eisenberg 10:45 I don't know the answer to the question.

Harry Stebbings 10:46 Mm-hmm. That's the first time anyone's said that on 20VZ. (laughs)

Michael Eisenberg 10:49 (laughs)

Harry Stebbings 10:50 Well, let me tell you, Harry. (laughs)

Michael Eisenberg 10:52 What's an area that I think there can be efficiencies in? Places where there's a huge amount of paper processes, where people are overburdened, but it's kind of compressible into a bunch of short workflows that you can kind of use AI to do right now. But I think kind of more broadly, uh, there used to be nuclear countries and non-nuclear countries. Now there are going to be AI countries and non-AI countries, and in the same way, there's going to be AI companies and non-AI companies. And I think what's going to be shocking about this that's not like the internet is, it's going to be harder for legacy companies to catch up to AI companies, which is different from the internet. You can kind of keep your retail business and still do e-commerce. You could kind of, in the early days, in the '90s, we had this thing called the intranet, which is how, uh, Fortune 500 or Fortune 5,000 companies embraced the internet into their companies. And, uh, they used the internet and they got better at customer service and they digitized. AI is different. If your data is not set up right, you're just not gonna get there. And I, you know, I listened to Warren Buffett say about his insurance businesses, that they thought they had 60 different databases. They found out, I can't remember what the number was, they had 600 or 6,000 different databases, they didn't talk to each other. It's like, good luck.And, um, when I heard that, you know, I said, "We should hold our lemonade for a long time."

Harry Stebbings 12:07 Can I ask you, when you think about, like, value accruing in the stack, I think what I worry about with a lot of them is customer service AI, SDR for sales teams AI. There is so many competitors in very specific niches, neeches, niches, whatever.

Michael Eisenberg 12:22 That's not a specific niche, by the way. Those are classic areas where software has done work. There was Siebel, there was Salesforce.com, there was Zendesk, there was... Those are classic areas where people run to first. And I think the more interesting places to invest, not to overuse Howard Marks' four quadrants of non-consensus right, but is, is to look for kind of out there things that it's hard for people to compete in, because you have specific domain knowledge that other people don't have.

Harry Stebbings 12:49 So, I really don't like competitive markets when I invest. I find that you have really cha- challenging times in terms of product marketing. You have higher CACs, you have lower retention rates, higher churns, you're all competing on the same channels. Do you agree with me or do you take the view that, no, competitive markets are competitive because they are where big markets are?

Michael Eisenberg 13:08 I'm in the Harris Demings camp.

Harry Stebbings 13:10 Thank God. We would edit it out if you weren't. (laughs)

Michael Eisenberg 13:14 (laughs) You know, I- I- on one hand, I never think about competition, 'cause I don't think that's what undoes companies, competition. I think bad execution undoes companies. And at the same token, I prefer, actually, market education. I prefer these completely kind of unchartered spaces. Um, I'll g- I'll give you, you know, a few examples. But like, we just backed a AI to synthetic biology and chemistry rocket fuel company. And you go, "Rocket fuel?" So, one, it's very competitive. A lot of people provide rocket fuel, but there's nobody who provides synthetically engineered rocket fuel from the bottom up, you know, that- that's at one-twelfth of the cost. And I just, you know, we might- we may succeed, we may fail. I don't know. Uh, there's a lot of market education that will come to saying, "Hey, we're not gonna take oil and refine it. Instead, we're gonna build it from the bottom up." Uh, it's gonna take a lot of market education to do that. But I think that's okay. I much prefer that.

Harry Stebbings 14:08 How do you size that in portfolio? So, it's like that is a- a inherently more risky play than a SaaS business or a FinTech business, I would argue. Um, how do you size that in a portfolio?

Michael Eisenberg 14:19 Oh, so I don't think it's a more risky play than a SaaS thing. I'll- I'll argue that SaaS is way more risky. One, because of the competition and price erosion. It's just, we had a false sense of security from models. Like, I can pull the model forward. I gave this presentation at our annual meeting a few years ago, which I said, you know, "We've hit peak SaaS." Anytime you can kind of stamp out MBAs from universities, you can pull the model forward, uh, that's not gonna be a venture business. Um, when I hear people talking about some of the growth funds, "Oh, software's like a bond, you know, with a higher yield." I go, "Just no." That's not the- that's not the venture business. I gave a cringe from that?

Harry Stebbings 14:58 Yeah, I just got goosebumps. I'm like, "Software's like a bond?" It's like, "Oh."

Michael Eisenberg 15:01 So, you know- you know, that- you have tons of growth funds and even PE funds thinking that they can predict the future revenue of software. In some cases you can. It's not, it's not that you can't, but doing that as a strategy I think is both inherently risky and non-remunerative because it's really, really, really tough, uh, to be unique. You know, in two- when I first joined Benchmark in 2005, I had some time on my hands because I didn't have a portfolio. So I try- I did this little piece of research to try to figure out why do companies get premium multiples?

Harry Stebbings 15:27 Mm.

Michael Eisenberg 15:28 So, uh, the obvious answer is high growth, right?

Harry Stebbings 15:31 Right.

Michael Eisenberg 15:32 But it wasn't always the case. And the thing I- I discovered, I- I kind of called and interviewed at the time sell-side analysts on Wall Street, uh, who had a- a more prominent place in the world in those days. And in all my research, I came away with the following, uh, insight, which is companies get premium multiples when it's the only way to play a future trend in the public markets.

Harry Stebbings 15:53 Hm.

Michael Eisenberg 15:53 And so, if I think the world is going towards X, AI, whatever it is, and I'm the first AI company, the only AI in a given market, I'm gonna get a premium multiple because it's the only way public investors can play that trend that they want to play.

Harry Stebbings 16:05 This is NVIDIA, no?

Michael Eisenberg 16:06 I think this is big time NVIDIA. And so being unique actually has real value. Again, provided that you have the growth and, you know, the- the gross profits, uh, et cetera. You can't ignore that stuff. Um, but all things being equal, if I have multiple ways to play a market, uh, you'll get a lower multiple. If I have a single way to play a market evolution, I'll get a premium multiple.

Harry Stebbings 16:29 I- I- I actually think about this a lot too, and I'm just like looking at, you know, I've got the founder of Klaviyo on the show tonight, which is a $7 billion company, you know, doing $750 million a year in revenue, growing 60% year on year.

Michael Eisenberg 16:40 It's quite remarkable, 'cause most public companies today in the software space are growing 10% to 20%, which is part of my peak SaaS theory.

Harry Stebbings 16:45 It's absolutely remarkable. But it, you know, the multiple on it is not great. I think the mul- multiple's like a 6X or a 7X. Um, and then I look at an Atlassian, which is trading at a 12X, and it's not an inherently better business. And I'm just trying to figure out the difference between the two and to your point there, like why one is premium and one is not.

Michael Eisenberg 17:06 I don't know enough about either business to have a view as to why Atlassian is trading at 12X. But I would assume that if you wanna play kind of developer growth-

Harry Stebbings 17:16 Mm.

Michael Eisenberg 17:17 ... which I think is a real trend, that's why you want to own Atlassian. There- there are more developers in the world today. It'll be interesting to see what happens with AI, but I think in general, horizontal SaaS processes are going to be disrupted by AI. We just actually did a portfolio review where we looked at like where the cat- we categorized our investments. Of the last 17 investments we made at Aleph, one was a SaaS

Michael Eisenberg 17:39 company.

Harry Stebbings 17:39 Sorry, last 17 investments, all but one was a SaaS company?

Michael Eisenberg 17:42 No, only one was a SaaS company.

Harry Stebbings 17:47 Only one? Vertical or horizontal?

Michael Eisenberg 17:49 Uh, horizontal, only one was a horizontal SaaS company.

Harry Stebbings 17:51 When we look at a lot of the big companies that we see today, and I'm- I'm not naming any or denigrating quality or anything like that, but like your Airtables, your Notions of the world, these are all horizontal SaaS companies.

Michael Eisenberg 18:02 Yeah.

Harry Stebbings 18:02 You know, bluntly, a lot of people say, "Well, AI actually kills vertical SaaS 'cause it means that the cost of creating that same vertical SaaS, but customizing it for your organization is way easier." So vertical SaaS is the one that dies.

Michael Eisenberg 18:12 I heard a rumor, I don't know if it's true, but I've heard it from three people now, which either means I'm in an echo chamber or, uh, it might be true, is that Amazon has said, "Don't buy any more software."... uh, AI can take care of all the kind of business process needs we need in software. Uh, I think that's going to be a trend. And so, I think you'll see-

Harry Stebbings 18:29 What does that mean, in your opinion?

Michael Eisenberg 18:30 ... I, I think you'll see a lot of companies build their own software, um, or you'll have tools that create kind of unique business process software for a lot of these companies. I don't know if it's tomorrow or in five years, but I think there'll be more of that. Um-

Harry Stebbings 18:45 You're skeptical?

Michael Eisenberg 18:46 Well, no, I'm just, like, I think we drastically overestimate the internal education or knowledge of company employees. And I don't mean that disparagingly. But you know, when we look at Accenture, they just posted 2.4 billion in generative AI revenue.

Harry Stebbings 18:58 Yeah. Why is that?

Michael Eisenberg 18:59 Because companies have no freaking idea how to integrate AI into their business units. W- w- what kind of business is Accenture?

Harry Stebbings 19:06 It's a consulting business.

Michael Eisenberg 19:08 Right. It's not a software business, it's a consulting business.

Harry Stebbings 19:10 Mm-hmm.

Michael Eisenberg 19:10 So, you have unique knowledge. By the way, McKinsey is also doing extremely well today in these at-risk fee businesses, right? They're not taking, uh, a lot of their fees today in, in kind of consulting fees, but they're saying, "We can improve your bottom line. We can improve your top line." And they go in there, and they use a ton of software and a ton of data that they've bought and created themselves, right? The fastest-growing business inside of McKinsey is software. And they go in and say, "I can bring, like, a lot of really, really smart people and a lot of technological capabilities and upgrade your business into the next era." That either does or doesn't include off-the-shelf software. I don't know, but I think there's a lot less of it-

Harry Stebbings 19:50 Huh.

Michael Eisenberg 19:50 ... going forward. Again, you have to ask a very simple question. I like... I'm more of a big-picture guy than a, than a micro guy, but why are so few of these companies growing better than 15, 20% in the public markets? You can barely find any. I think the market for purchasing software is slowing down.

Harry Stebbings 20:05 100%. Well, so, I mean, there's a couple of different options. There's one, you know, you could have, like, a box, which is just, like, the saturation rate of the market is-

Michael Eisenberg 20:12 Yep.

Harry Stebbings 20:13 ... so high already. I mean, they already have pretty much everyone as their customer, so they need to upsell a new product which they've been... Aaron will admit this, so I'm not shitting on him, but too slow to do. Uh, same with, uh, Dropbox as well, which you would admit too. Um, or, to your point, they're slowing down, or not buying as much.

Michael Eisenberg 20:31 Whatever it is, I think it's slowing down.

Harry Stebbings 20:32 Yeah.

Michael Eisenberg 20:32 (laughs) So, you know, th- is that a trend? I don't know. I think we've squeezed a lot of efficiency out of businesses using kind of basic software already. And so, to get to another level of efficiency or revenue generation is a big deal. I think also corporate buyers in whatever business they're in, what they're interested in is not the software to squeeze the incremental dollar of efficiency out of my business. It's, "Where's my new revenue opportunity from this?" And that's gonna require kind of a big leap, and you don't get that in traditional SaaS software.

Harry Stebbings 21:01 I guess the question that's interesting there is, like, are you able to actually extract value efficiently? Because if you are making millions and millions for your customers in that way, how do you actually extract the value that you are providing in an efficient way that's not like a, a SaaS model where you only get 50K and you're actually saving them 10 million?

Michael Eisenberg 21:20 I think buyers today are willing to pay for value. I think there's a bunch of reasons for this. Um, it's harder for companies who are used to price-per-seat models or used to large enterprise to kind of adapt to this. There's been a bunch of pieces written, I think Sarah Tabler wrote one, that people will pay for work.

Harry Stebbings 21:42 Yeah.

Michael Eisenberg 21:42 I think that's a real thing. I think people will pay for improvements to their business. And I think the consultants like Accenture and McKinsey are at the leading edge of this. Um, but I think it's gonna come into software

Michael Eisenberg 21:52 very, very quickly.

Harry Stebbings 21:53 Yeah. No, I, I, I totally get... Can I ask you? I, I remember speaking to, uh, our mutual friend Mark Evans before, and he said, "Hey, the thing you just always have to remember with adoption cycles is we always overestimate in a year and underestimate in 10, so just breathe." And I just always remember that, and it, it really actually affects how I think about investing and, you know, adoption cycles. Do you think that's the case here, or do you actually think, no, no, no, we are seeing the fastest technological breakthroughs and development within LLMs every single week? It is a velocity that we haven't seen before.

Michael Eisenberg 22:26 I think both of those things are true again.

Harry Stebbings 22:28 A wise man once told me that it's about holding two opposing opinions. (laughs)

Michael Eisenberg 22:32 Exactly. I think, I think the adoption is, is incredible. I think the pace that these things are learning at and, uh, and adapting at is incredible. I mean, you just... By the way, you see the efficiency in customer service at a lot of these fintech companies, whether it's Lemonade or Adyen or-

Harry Stebbings 22:43 Mm-hmm.

Michael Eisenberg 22:43 ... or, or Klarna or any of these people. It, it's, it's stunning. And at the same time, when you kind of look across a broad economy, it's gonna take a lot longer than people think. And there'll be some disruptors and some people who get this, and you know, and regulation will have something to say about it. Not in the way everyone's talking about, you know, like, uh, you know, uh, uh, Sam Altman trying to do regulatory lock-in. But there's just a lot of businesses that are regulated. Like, pharma is regulated and financial services are regulated and, you know, it's hard for regulators to keep up with AI. And that's gonna be a big deal for how much it can be deployed.

Harry Stebbings 23:14 So, my biggest concern actually is that you have regulators who fear that it's getting away from them, place these very punitive policies or, you know, regulations on data, on data access, and actually, we have a real plateauing in the development of AI systems because of regulatory challenges.

Michael Eisenberg 23:31 Super high level, I think that's more likely to happen in Europe than in the US.

Harry Stebbings 23:34 Agree. (laughs)

Michael Eisenberg 23:35 (laughs) I think that is gonna set Europe back in competitiveness more than it's already been set back in competitiveness over the last 20 years. If I'm in Europe and I'm starting a company, I would get out.

Harry Stebbings 23:47 (laughs)

Michael Eisenberg 23:49 Sorry.

Harry Stebbings 23:50 Oh, thanks, Michael. (laughs)

Michael Eisenberg 23:51 Uh-uh.

Harry Stebbings 23:52 Welcome to London.

Michael Eisenberg 23:53 I think, I think the US can ill afford to slap too many shackles of regulation on AI companies because they're in, uh, com- competition with China.

Harry Stebbings 24:03 Yeah.

Michael Eisenberg 24:03 And I think Israel won't do it. And so, when you look at kind of the key markets for AI development, where the US is first and China is second and Israel is third, um, I think those are likely to be less regulated than most, and certainly than Europe. And so, these, those will be advantaged economies, uh, going

Michael Eisenberg 24:22 forward.

Harry Stebbings 24:22 So, we had Alex from Scale.AI on the show the other day.

Michael Eisenberg 24:25 Mm-hmm.

Harry Stebbings 24:25 And he said, "The thing that we're not talking about enough is that, you know, really..."... AI is the most demonstrable power that any country or nation can hold and leverage in a, in a case of war, and we should have closed systems as a result. Open systems would mean that our, our biggest enemies, Russia, China, are able to access the same levels of technology that we all. We have to close the systems, and we have to leverage AI to build AGI to win wars. Do y- do you agree with that perspective?

Michael Eisenberg 24:57 I agree with it in part. I think it is true that AI is likely to be a key competitive advantage in wars. Uh, even significant. Look, we're living through a war in Israel right now, and so, uh, kinda closing the kill chain using AI is a thing, not just in Israel. It's all over the world. And there's, there's a big European company now that's doing a few hundred million dollars of revenue that uses AI to close the kill chain for European armies. And-

Harry Stebbings 25:27 Helsing?

Michael Eisenberg 25:28 Yeah.

Harry Stebbings 25:29 Hm.

Michael Eisenberg 25:29 And so, um, this is, this is a real deal. I said before th- i- if the 20th century was about nuclear countries versus non-nuclear countries, the 21st century is AI countries versus non-AI countries. We shouldn't lose sight of the nuclear ones 'cause there's these terrible people in Iran out there, you know, who are looking to acquire a nuclear weapon. Uh, by the way, not just threaten Israel, they threaten Europe. That's the key thing people don't understand. They have missiles long enough to hit Israel. They're, have new missiles long enough to hit Europe. And thinking of them as an AI country is quite frightening, actually. And they might be. They're very smart people there. And I think what is naive about what Alex said is that you can keep it closed. I just think it's tough. You know, the US put a ban on NVIDIA chips going to China. I, I, I don't sense that it slowed China down in a meaningful way. I think there's enough kinda semi-degraded chips that you can, that you can kinda string together, or open-source designs. You can kinda figure this out. I think the world is far more open than we give it credit for today. And so, it's really, really, really tough to shut this down. I think we just have to be ever more competitive technologically to stay ahead of the bad guys.

Harry Stebbings 26:42 I can't remember who it was, but someone said recently that China is two years behind the US on AI. Would you agree with that?

Michael Eisenberg 26:48 I think they were 10 years ahead of the UA, uh, th- uh, the, the US on AI and PSYOPs using TikTok. I think they were 10 years ahead of the A- uh, US in sophistication of how to manipulate the minds of young people using AI algorithms in TikTok. I would assume, I generally assume of my competitor, of my enemy, that they're much smarter than I am. That's the only way to stay ahead. So, even if, by some bizarre notion, you think that they're two years behind, you should think

Michael Eisenberg 27:19 they're two years ahead.

Harry Stebbings 27:20 I slightly worry, honestly. We mentioned a little bit about kind of, you know, geopolitics and global conflicts. In the investing world, I feel a little bit uncomfortable when I have these, like, 30 to 35-year-old MBAs, nothing against them, but who are, like, getting super jacked up about the margins of defense companies and how defense is, like, the best place to be right now. And I'm like, "You are aware of the weight of your words." Like, I'm, I, I, uh, yeah, I know that we have enemies and, you know, eh, people have different rights or whatever, whatever. But, like, this is still about closing kill chains. And you're, like, talking in your MBA language about margin optimization. It just feels a bit off. How do you feel about, like, this new wave of excitement towards defense and also being aware of the weight of what that actually means to fund with billions of dollars more efficient killing?

Michael Eisenberg 28:12 As a father of four children who have served in Gaza and on the northern border, which Hezbollah has been burning nonstop, uh, in Israel, war is a real thing. It's a very serious thing. It's serious because people get killed. It's serious because it shows what you're fighting for, which is freedom and the values of the West. And there are enemies out there. I've, I've written publicly that I think this generation of Israeli kids is the defining generation. They stood up and fought for what matters. I think there aren't enough people willing to stand up and fight. Now, these kids, because they've been in battle, have a level of maturity that your average MBA student does not have, and they understand what it means to hold a weapon. Holding a weapon is a responsibility. You know, Israel has a tiny homicide rate, despite the fact that most of the population is armed and many of them walk around with automatic machine guns on their back. Like, you can go to a wedding, I was just at one, there are eight girls in wedding dresses of sorts, and they have machine guns slung on their back. But you learn to respect it. You learn to understand that this is a serious matter, and that if you're not careful and you're not responsible, then people get killed. And at the same time, I think that there was always a, an alliance between Silicon Valley and the defense industry, and patriotism to use... You know, I'm from United States originally, is, is a real thing, and it should be in tech. You should be patriotic. I think patriotism is a great thing, a wonderful thing. People should be proud of their people and proud of their country and power- proud of their cultures. That's not the same thing as, uh, my margi- y- y- you know, your margin is my opportunity. Um, I think there will be a lot of money made in defense going forward. I think people way underestimate what it tells, what it takes to sell defense. You know, the government buyers, the regulations, how hard it is to sell. If you've never sold defense technology and you just come out of an MBA, good luck. You know, Palantir took 20 years to crack this code. Andrew cracked it much faster because of Palantir. Other people have done it. This is, this is a profession. It's like, it's not a couple of guys, you know, in, in... Palmer Luckey, people kind of, because he's dressed in a Hawaiian shirt, don't take him seriously enough. Trace Stevens is there, who was at Palantir. Trace Stevens of Founders Fund. That guy...... is incredible. And he knows more about selling defense technology tha- than everyone's forgotten. And that's just different. And so, people will lose a lot of money chasing the new, new thing, which today happens to be defense in addition to AI.

Harry Stebbings 30:42 But we're seeing this, like, move to hard tech. It's, like, people seem to agree with you actually. SaaS is, you know, per- over-proliferated, too competitive, whatever we wanna say on that. But actually, we need to move to hard tech. We need to move to, uh, whether that's climate or... but much more physical, challenging, I think challenging, uh, style of investing. How do you see that, Michael? 'Cause I think a load of software investors are gonna lose their shirts investing in a completely different type of company.

Michael Eisenberg 31:09 Listen, I'm unencumbered by any knowledge of technology or- or- or software. And I... Nor management experience, by the way. I've never managed anything and I can't write a line of code. The last line of code I wrote was in seventh grade. This is the venture business. You're gonna lose a lot of money. You're gonna lose a lot of companies. The question is actually, why does a founder want to take your money if he comes from something else other than software? And I think that'll be interesting and very Darwinian. Some founders, or the great founders, are likely not taking SaaS investors' money.

Harry Stebbings 31:37 Hm.

Michael Eisenberg 31:38 Because the SaaS playbook, as it's become called, doesn't apply to any of these harder technologies. And by the way, you know, we- we've kind of dichotomized this. Either you do SaaS or you do this, like, ridiculously hard technology like Anduril. There- there's also in the middle, whether it's, you know, semiconductors or these kind of new emerging fields in, you know, synthetic biology and synthetic chemistry. I've coined this term. You know, when- when I grew up in this business, uh, fabless semiconductors became a thing in the late, in the late '90s. I was a early guy around a company called Galileo, by Avigdor, uh, Wilens, which Mellanox came out of, and a bunch of others. Fabless started to be a thing in the '90s. I think we've now, uh, gotten to labless. So, uh, you can do synthetic biology, you don't have to build a lab. You can outsource. Part of what happened... You know, I mentioned before, that in early 2000s, late '90s, we built out this giant fiber optic infrastructure. ZIRP built out a huge amount of biological infrastructure. And so now, you can just rent labs for cheap. And so, I think we're moving to a labless model where you do kind of AI, you know, in your office, and then send out things for testing to labs, uh, and maybe even some manufacturing to capacity that's been built over the last decade.

Harry Stebbings 32:49 But transparently, Michael, I just don't feel intellectually equipped to underwrite a AI synthetic biology company.

Michael Eisenberg 32:56 Yeah.

Harry Stebbings 32:56 I'm- I'm humble enough to be like, "I just, I don't understand it."

Michael Eisenberg 33:00 That's- that's my advantage. I don't understand anything about software either. So, uh, but I-

Harry Stebbings 33:04 But you do, but you do. Like, you're- you're no, you're- you're- you're-

Michael Eisenberg 33:06 No, I've gone back to learning chemistry now.

Harry Stebbings 33:07 You're very humble, but you do, you understand it very well.

Michael Eisenberg 33:10 No, I've gone back to learning chemistry and studying chemistry now. Uh, but I didn't do very well in high school chemistry. I- I had, I had a teacher who I think thought, and I... he was right. I- I- I wasn't very good at chemistry.

Harry Stebbings 33:21 (laughs)

Michael Eisenberg 33:21 Um, but I've used actually ChatGPT now to- to teach me chemistry, and I've been doing kind of high school chemistry again, uh, through this thing. He's the- the founder and... What I did was, I- I brought this guy back from America. Uh, he's an Israeli guy with a PhD from the Weizmann Institute, who had built a synthetic biology lab at Harvard Medical School, and then went to Amazon to their Moonshot factory. And I convinced him to come back to Israel. I put him in the office next door to me for a year. And so I learned by osmosis the very little amount of extra chemistry I now know. And, uh, we built this company together. And I think that's how you build companies. I don't really know anything. I- you know, I'm invested in- in Empathy.com. You had Ron Gore on your show, I think?

Harry Stebbings 33:56 Yeah. No, I haven't had Ron on yet, but-

Michael Eisenberg 33:56 You should have him on your show. You know, I didn't-

Harry Stebbings 33:58 I actually just had Danny Rimer on from, uh, Index, and he was laying about Ron the other day.

Michael Eisenberg 34:02 Yeah. You know, we're so fortunate to have Danny, um, lead The Last Round, and it's a great band. We got, we got Joel Cutler from Catalyst and Danny, and myself and Ron, and it's- it's just a lot of fun. I don't know anything about the life insurance market, or, you know, how next of kin handle- handle grieving. It's okay. You know, hopefully it keeps me humble enough to say, "I'm really sorry, but I can't help you terribly much, but here- here's who might be able to."

Harry Stebbings 34:25 You- you said before, when we were chatting before the recording, that I could just exploit you for your wisdom and your... more experience. Let's put it that way. I think you said age.

Michael Eisenberg 34:34 My age.

Harry Stebbings 34:34 Uh, yeah. But I- I... increased wisdom is what I'll go for.

Michael Eisenberg 34:37 You know, on my podcast with Beazer, I said, "We should talk about ageism," because she should probably get rid of

Michael Eisenberg 34:41 me from the fund. I'm too old to be doing this business.

Harry Stebbings 34:43 (laughs) Um, listen, I wanna, I wanna start on, like, my big concern right now. And I sound like I'm the old one, but I'm wondering, where ha- where is liquidity gonna come from? Like, I'm looking at the different avenues, which is IPO, it's M&A, and then it's like PE. And I just want to hear your wisdom and experience on this, because I'm concerned. M&A seems totally shut. Lina Khan seems like, you know, the ultimate, uh, challenge-

Michael Eisenberg 35:07 Socialist.

Harry Stebbings 35:08 ... yeah, to- to M&A. How do you see M&A, Markus? Is it completely shut? Am I right to be worried?

Michael Eisenberg 35:13 Lina Khan, I would say, openly, is I think a threat to American capitalism. Whether Biden or Trump gets elected, she needs to go. Um, because we need to be able to have M&A for the big companies. And as best I can tell, she hasn't redefined what antitrust is. And so, I think she's taken a lot of liberties that are, that are just bad for capitalism globally. Where's liquidity coming from? So, I have bad news for you. Most people don't get liquid. Actually, by the way, not even most people, like 98% of people don't get liquid. This is a very difficult environment, number one. But number two, this actually is a hard business. I know there's been like an explosion of venture capitalists. It- it's a hard business.

Harry Stebbings 35:55 Le- let's just go back a step. You- you said that most people don't get liquid. One way to get liquid ahead of time is secondaries, or to sell p- parts of positions.

Michael Eisenberg 36:04 If there are buyers.

Harry Stebbings 36:05 If there are buyers. We-

Michael Eisenberg 36:06 You know, one of the things I've discu- so- so let's just back up. A tiny number of companies get sold or go public. I don't buy that the IPO window is closed. The IPO window is wide open. The question is, what's the price you're willing to take? And so, you can go pu- ... Reddit went public. It's done very well in the public markets. It develo- delivered a stellar first quarter after going public. It's doing very well. It's like a $7 or $8 billion company right now. You don't need to have a half a billion or a billion dollars-... of revenue to go public. It, we took Lemonade public with 60 million of revenue. You should go public early an- and the markets are open. Take the price the public markets will give you and build your company in the public markets. You will be better off for it. I am a huge believer in that.

Harry Stebbings 36:46 Would you really say that, uh, to your companies today, who are at 200 million now and they're going, "Ah, you know what? Actually, public markets will probably give us, uh, 2x premium in 18 months' time?"

Michael Eisenberg 36:56 Uh, well, I don't know if they'll give you 2x premium in 18 months' time, but go. And by the way, if, if the bankers are playing this game, "We want you to have half a billion dollars," et cetera, find a second-tier banker. They're just as good. You don't need to raise $500 million and go public. Go get, take your company public, let's say, at 100 million of revenue. Take your company public at 600 million, raise $80 million, build an efficient business, and go. Just go build it there. I, it's like, this is...

Harry Stebbings 37:21 I think people forget that your Shopify went public at 700 million.

Michael Eisenberg 37:25 Yeah.

Harry Stebbings 37:25 Like, some of the best businesses in the world.

Michael Eisenberg 37:27 No, Amazon, the best business in the world, went public... and look, by... Everyone's forgotten that Facebook went down 50% after their IPO. It's one of the biggest companies in the world today. Now called Meta.

Harry Stebbings 37:37 (laughs)

Michael Eisenberg 37:37 Maybe that helped. The... It's okay. And everyone's, like, expecting, like, PE to turn up and buy these companies. I got bad news. Number one, they're gonna buy a tiny percentage of companies. Number two, interest rates are way up, they can't borrow anymore.

Harry Stebbings 37:49 (laughs)

Michael Eisenberg 37:49 Number three, they're incredibly price-sensitive. And so most companies waiting for PE to come save them, I hear th- there's, like, this whole meme right now that PE is gonna be the buyers of the future.

Harry Stebbings 37:58 Well, li- listen, we had Mark Suster on the show, and he said, "Listen, PE will be the primary buyers for the next generation of software companies." I, I just want to touch on a couple of different components that you mentioned, that you said about the cost of capital. Uh, how does that impact how they think about buying? For the people that don't know.

Michael Eisenberg 38:13 It drives prices down. Increased cost of capital drives multiples down, just what it does. When interest rates go up, multiples go down. It's the laws of physics.

Harry Stebbings 38:20 Mm-hmm.

Michael Eisenberg 38:21 And so they'll pay a lower multiple, because their cost of capital is higher. And, you know, people are just not gonna get liquid the way they think they are. And the top part of the cap table... This happened, by the way, in, oh, 2000, '01, '02, '03, '04. The top part of the cap table got its liq pref back. People have forgotten that liq prefs matter. And they got the liq pref back, and in some cases, there used to be an 8% coupon on these things. Many cases now, there isn't. They took an 8% coupon, and then what was left went to the rest of the cap table.

Harry Stebbings 38:48 Uh, you can tell them we're gonna see many more Pluralsights. Pluralsight obviously happened last week, which Vista wrote down from three and a half billion to zero, because, you know, the cost-

Michael Eisenberg 38:55 I'm not concerned, I'm certain. It's happening, we just don't know. By the way, in new venture portfolios, there's a lot of this also. And I, I also wonder a lot about how many SaaS companies get undermined by AI right now. And there's a lot of SaaS portfolios sitting at hefty prices in people's portfolios, hefty values in people's portfolios that aren't worth that anymore.

Harry Stebbings 39:12 You're certain. So that means that there's a ton of companies that PE have bought where the cost of that debt is now way too high, and they are in trouble. Does that mean we have a generation of, as I think Jason Lampkin puts it, Mulligan PE funds? And do they just get let off the hook for it?

Michael Eisenberg 39:26 I have no idea whether they get let off the hook for it. I don't know if it's a ton of companies, but it's not one.

Harry Stebbings 39:32 Okay.

Michael Eisenberg 39:32 It's not just Pluralsight. There's a lot more like that out there. There's a lot more like that out there. And, but also think about the following. If you are a PE fund that just took a significant write-down on one or two big software purchases, are you gonna have an appetite to buy something? I don't think so. There's a small number of really great businesses that are built in venture. It's, like, 10 a year. Um, and some other ones get smaller liquidity events, which is great. We should have that. Um, but if you can get paid 10 to 20 times on your medium companies, or even five to seven times on your medium companies, it's not a bad outcome. You know, we talk... I don't want to mention the company name. We talked before, on another show, we just took 100% of our equity off the table in a company. I think we did 11x or so. We have a different view of the company than the management and some of the other investors. Dude, that's fine. It's totally fine. And we're very happy, and we're very happy... We'll be super happy if the founder makes it into 30x and we were wrong. We'll be so happy for him, and so happy for the other investors. But liquidity is a thing. And by the way, the last thing I'll just say about the secondary markets. We've discovered, you know, you get these emails like... You probably get them also. Two a day that says, "I got a buyer in the market for so-and-so company." So, over the last three or four weeks, we've decided to do a little experiment. We reach out to them now. We used to ignore them. But now it's interesting, because, like, you discover there are actual buyers there. There aren't. Okay? There are not actual buyers there. It's, "Hey... Okay, we can hold this price if you get us due diligence material on the company." These are just people looking for data. And so, uh, I think most of those buyers aren't

Michael Eisenberg 41:03 out there either.

Harry Stebbings 41:03 Very rarely do you hear the advice to sell all of your position. The kind of canonical advice is, "Well, I'll sell 33% at a time in increments." Um, selling all of a position is, respectfully, a strange take. Like, I would never sell all of a position, Michael. I would sell 80%, maybe, even when I didn't believe, but I would still want to ride the upside on 20% in case I was wrong. How did you think about that 100%?

Michael Eisenberg 41:29 You asked me before we got on the show what is the biggest mistake I made during ZIRP.

Harry Stebbings 41:33 Right.

Michael Eisenberg 41:33 One of them was not selling 100% of one or two positions-

Harry Stebbings 41:36 Mm-hmm.

Michael Eisenberg 41:37 ... uh, when the prices were, were high. Uh, I, I'm actually in the opposite camp generally, which is ride your winners. That's my... the general camp I'm in. I think the question you need to ask yourself if you're a venture capitalist and you sit on the board is, "Do I have, uh, an information advantage or a unique insight that somebody else does not have?" And if I think I have that, I should act on it. And if I should act on it, I should act on it with conviction. It's like when I go into companies, I don't want a half position. If I have conviction, I'm gonna take a whole position. This business is out- an outlier business. It's just the nature of it. So take the whole position, and if you think you've hit a good return, take the money off the table. I don't think you average into positions in the venture capital business. You do right in the public market, so you kind of average into a position and then average out of a position.

Harry Stebbings 42:25 Mm-hmm.

Michael Eisenberg 42:26 I, I don't think this is like this. This is a binary industry. Very, very binary, uh, investments. And I'm, I'm in that binary camp.

Harry Stebbings 42:33 Do you do pro rata?

Michael Eisenberg 42:35 Rarely.

Harry Stebbings 42:36 Why?

Michael Eisenberg 42:37 I kinda drank the Benchmark Kool-Aid. We don't get the same kind of credit that Benchmark gets, that because Aleph invested, so someone's gonna invest, so we kinda just have to do one more round. But we try to kinda peel back on it. We think we, we have to buy our ownership in the first round. Uh, our unique insight is in seed and A. Buy it. Write the check, buy it. We wrote this giant check to a company called Dream, uh, Security, which is, uh, AI infrastructure for protecting critical assets in count- countries. Right? It was the biggest check we wrote.

Harry Stebbings 43:10 (exhales deeply)

Michael Eisenberg 43:10 Crazy competitive deal. Incredible team.

Harry Stebbings 43:14 When you say the biggest check you wrote, it's like proportion of fund, like 10%?

Michael Eisenberg 43:17 No. It was... Initial check, 6.5, 7% of the fund.

Harry Stebbings 43:22 Okay.

Michael Eisenberg 43:23 They're not gonna need to raise any more money, it looks like, after the A. And so, they can, but they probably don't need to. If I didn't get my ownership in the A by writing this big check, I, I, I wouldn't have a position. And I, I still feel like I'm sub low. I'd love to have more.

Harry Stebbings 43:37 I think one of the biggest misnomers is, "Hey, we'll place a load of bets, and then we'll be able to concentrate capital into our winners." Whenever I hear this, I'm like, "Are you high?"

Michael Eisenberg 43:46 (laughs)

Harry Stebbings 43:46 Like, wha- when does that ever... My best performing companies, generally speaking, like 99% of the time, have and all the best all over them. I'm not like, "Oh, sure, I'd love to go treble pro rata."

Michael Eisenberg 43:56 I was talking to a venture capitalist, Jesse Beyroudi at IA Ventures.

Harry Stebbings 43:59 I like him. He's .

Michael Eisenberg 44:00 I like him too. He made a very smart observation. H- he said that, "We do pro rata into our, uh, middling companies." Super interesting. He said, "The good ones just kinda get, you know, they become runaways and you kinda can't get more capital in. Bad ones, you wanna put money into." But sometimes one out of some number of these kinda middling companies where most people don't see what's going on there at the beginning, but I do. You give them a little extra capital or you kind of double down once or twice, and you end up with a winner out of this kinda middle part of the portfolio that's a giant winner. I thought that was super insightful.

Harry Stebbings 44:31 I look at my, my Fund One, which is a tiny, you know, $8 million fund. Actually, our best value drivers today were all in the meh.

Michael Eisenberg 44:40 Yeah, it happens a lot.

Harry Stebbings 44:41 And the worst performing companies were all in the rocket ship category. (laughs) They're the ones that went to zero quickest.

Michael Eisenberg 44:47 Those are the ones you should've sold 100% of as they were on their way up.

Harry Stebbings 44:50 I should've done.

Michael Eisenberg 44:51 (laughs)

Harry Stebbings 44:51 I know, I know. Absolutely, I should've done. But actually, I think there's a big difference between consumer and enterprise there. Like consumer, sell, because like, I mean, what I found was like, it's much more transient of whether it works or not. Enterprise, you've got a lot more predictability.

Michael Eisenberg 45:05 Consumer is ephem- ephem- ephemeral.

Harry Stebbings 45:07 Yeah.

Michael Eisenberg 45:08 And, um, sometimes works, sometimes doesn't, sometimes goes out of style, sometimes goes in. So I think of consumer as fast fashion-

Harry Stebbings 45:14 (laughs)

Michael Eisenberg 45:14 ... uh, in many cases, unless you build a moat. And I think there are rare consumer companies who, who build a moat. You know, I've brought up the AI conversation, somebody from Target Consumers. It's really hard to tell the difference visibly between, um, OpenAI and Anthropic, and any one of a number of models. So, you kinda ask yourself, "What's the moat?" So I'll tell you, the moat's the data. And over time, some people have, some people w- will note that it's significantly better. It's like Elon Musk is talking about XAI is gonna be the place of truth. I hope he's right. I'm not sure how I'll know. That's kind of a question. And so, how do you build a moat on those businesses? I, it's not clear to me right now. Uh, Google built a moat 'cause they were by far the fastest, and then they had this kinda reinforcing ad business, uh, that added onto it. But who gets to build a moat in these businesses is not obvious to me.

Harry Stebbings 46:03 We spoke about the three different exit sources being M&A, IPO, PE. When you're making investments, do you think about kind of value creation and pathway to exit?

Michael Eisenberg 46:13 Nope.

Harry Stebbings 46:14 Nope, nothing at all? So we don't outcome scenario plan of, "Okay, if HoneyBook does X, Y, and Z, then we have this outcome." We don't think that?

Michael Eisenberg 46:22 Nope. For a variety of reasons. Number one, because companies pivot. Number two, because you convince yourself of all sorts of stupid things by, by kinda dreaming about the exit. And number three, it never works out like that, so why bother?

Harry Stebbings 46:35 Do you think we're gonna see a generation of funds die, Michael?

Michael Eisenberg 46:38 Of course. Bruce Dunlevie from Benchmark used to say that venture capital has the highest b- barrier to exit of any business out there. No one wants to leave.

Harry Stebbings 46:43 (laughs)

Michael Eisenberg 46:43 You know, the management fees kinda keep coming. I, I think that's true. So it takes longer than people think, but yes, of course.

Harry Stebbings 46:48 Um, which segment is that?

Michael Eisenberg 46:49 But you know, a whole good generation of people grow up during and you know, now reality is biting. And so, yeah, of course.

Harry Stebbings 46:56 Uh, but who is that-

Michael Eisenberg 46:57 That's not a bad thing, by the way.

Harry Stebbings 46:58 (laughs)

Michael Eisenberg 46:58 It's like startups going out of business. You recycle the talent into something better.

Harry Stebbings 47:01 Who is that segment though? Is that kind of old, ailing funds that haven't innovated in the Valley style? Or is it the microfunds where it was like, you know, uh, respectfully, p- operators who had $5 million funds on the side and were doing it part-time 'cause it was kinda cool?

Michael Eisenberg 47:17 I think the first style of funds that goes out of businesses is funds where there's three, four, five partners, and only one of them is good. And then that person gets tired of holding everybody on their shoulders, and they leave and set their own thing up, and everyone knows. So I think that's one. Um, there's just some people who won't deliver capital back to LPs, and they'll be gone. And some will be big and some will be small, and it's fine.

Harry Stebbings 47:36 I've been shocked by how much focus LPs still place on TVPI. Like, you know, I speak to a lot of LPs, and I thought given the last few years, they would reduce the importance of TVPI and just focus on TV. No, they're still like, "Oh, wow."

Michael Eisenberg 47:51 Well-

Harry Stebbings 47:51 I'm like, "Really? You still give a shit about that?"

Michael Eisenberg 47:53 Well, no, they should actually. And, and this, you know, I had this tweet in back and forth with Ibrahim Ajami of, uh, Mubadala. You know, the, the DPI is what matters. And that's true. If you didn't take money off the table during, you know, the era, there's something wrong. Let's just be clear about that. But TVPI actually matters because, uh, I had this conversation with one of our LPs at our, at our annual meeting, which is, if the thing is compounding at a higher rate and keeps going, it, that's not a bad thing. And I can kind of think through, if I kinda, uh, wait another two or three years, that this will be worth a significantly larger amount of money. The question is one level beneath TVPI, which is, how real is it and how sustainable is it? People in the venture business, and this goes back to the previous competitation- conversation about competitive, uh, markets versus noncompetitive markets, underestimate...... um, moats on businesses, the value of moats and competitive advantage on business. There is insufficient time spent on, "How do I deepen competitive advantage in this business?" To take a financial rendering of this, multiple is my expectation for future cash flows. So, the question of future cash flows is not just how much cash it is per year, but how many years this goes on for. How many years this goes on for is a function of competitive advantage. So, you need to think through how deep my competitive advantage is in a given category. You know, there's a lot of talk about Google right now. Does AI undermine their ad model? Yes. But that competitive advantage is actually really deep, I think, for, for, for the time being. How long that lasts, though, is really the question. And so, um, and that depends on consumer behavior, which is fickle, right? How much of it moves to a different modality in, in search? And so, the question that I think needs to be asked in TVPI land is, these underlying portfolio companies, how deep is their competitive advantage? How sustainable is the growth? And that's key. And I think when you kind of dig that level deeper, you'll find that a small number of companies matter to TVPI, and, and that matters.

Harry Stebbings 49:57 You said how sustainable is the growth there. We're seeing this kind of transition from a lot of companies, from hyperscalers or hypergrowth, to actually just quite slow growers. And maybe sustainable growth, but just low growth companies that need to just grind it out. What happens in that world?

Michael Eisenberg 50:15 I think, you know, we talked about like, 100% growth like it was easy for a decade, so, but it's not. 30-40% growth is great business.

Harry Stebbings 50:21 Yeah.

Michael Eisenberg 50:22 Um, so let's just say that, first of all. Second of all-

Harry Stebbings 50:24 But very few have it. (laughs)

Michael Eisenberg 50:25 But very few have it. 10 to 15 to 20% growth is tricky. That's a tricky middle place if you're a venture capitalist, 'cause it's not clear how you get out of it. These would have been great businesses if they didn't take venture capital. If they took venture capital and, you know, stacked a large preference on the, on the cap table, it's challenging. What happens to them? Some of them will grind it out for a very long time until some sort of exit market opens up. As long as they're in control of their own destiny and they're profitable, that's not a terrible way to think about this. Others will be traded in large secondary trades and por- por- portfolios. Um, and some of them, if they're good enough, will buy back stock. I just had a company, actually, that bought back stock from early investors. I, I didn't sell 'cause I thought the price was too low. But I have a company that generates more than enough cash, we bought back stock. So, I reduced dilution in this company, which already wasn't much, uh, 'cause it was a profitable business for a while. And we'll see what happens to the asset over time. Uh, you know, w- we will make, by the way, a significant multiple on our capital whenever it comes out. It's not growing... It's probably, the CAGR's probably 30-35%. And it's had a slower growth year, but generates a lot of cash. We'll see what happens. By the way, one of my bigger exits in, in my career, I, we distributed hundreds of millions of dollars in dividends. It's a strange... It doesn't happen, but... It's never happened to me again, but...

Harry Stebbings 51:39 Final one. If you were to analyze yourself, say, as an investor, where do you most need to improve?

Michael Eisenberg 51:44 That's a great question. Um, two areas. It has been said about me multiple times that I am, uh, w- very, very direct, and therefore an acquired taste.

Harry Stebbings 51:56 (laughs) That's, that should be the name of the podcast, An Acquired Taste.

Michael Eisenberg 52:01 (laughs) And it takes time for entrepreneurs to understand that I'm just, I'm asking the hard questions and being very direct because this is important and urgent and they need to get going about this.

Harry Stebbings 52:10 Mm-hmm.

Michael Eisenberg 52:10 And there's no sense in beating around the bush. And so, probably, I should find a better way to deliver messages, um, that is not as...

Harry Stebbings 52:18 Do you care? And what I mean by that is-

Michael Eisenberg 52:20 No, I don't.

Harry Stebbings 52:20 ... a lot ... Uh, because truth be told, what I find is the best founders appreciate the directness. Uh, take it or not, they appreciate the directness and move on.

Michael Eisenberg 52:28 You know, one of the things I found about myself as I'm getting older, I'm getting cranky about board meetings. There's a lot of pontificating at board meetings.

Harry Stebbings 52:35 (sighs)

Michael Eisenberg 52:35 And I just think most board meetings... I've come to the conclusion that with good preparation, most board meetings can be finished in 45 minutes. Um, there's real governance that needs to go on there. There's core strategic issues that you need to kind of pound on. Um, but most of them can be finished in 45 minutes. Some need two hours. None need three to four hours. And there's just a lot of kind of endless sharing of opinions.

Harry Stebbings 53:03 For boards that are done in 45 minutes, what is the prep that is done? Is it a Google Doc with three cool questions that people think about ahead of time?

Michael Eisenberg 53:10 It's generally a letter and not a presentation.

Harry Stebbings 53:12 Mm-hmm.

Michael Eisenberg 53:12 It's rich in data. But it upfront identifies, these are my core one, two, or three strategic issues that I got to get board input on. And, uh, you prep for it, you do calls beforehand, and then the conversation is, uh, super concentrated. And because you're attacking the question, I find that people di-... kind of don't expand. I, all these boards, they show endless amounts of numbers, and people, you know, want to dig into the margins. And then you show a product plan, and everyone wants to be the product manager on the board. It's like, just no. I, I'm getting cranky about this, and I probably should moderate my behavior, so there's another area for, for, for improvement.

Harry Stebbings 53:48 Do you find Zoom board meetings are as effective as in-person board meetings?

Michael Eisenberg 53:51 Not at all. I don't, I don't think, I don't like Zoom at all. I hate Zoom. I do... This is better?

Harry Stebbings 53:56 Way better.

Michael Eisenberg 53:57 You know, I didn't invest over Zoom dur- during the pandemic. I just, I met people. Out of a farm, I met people. And I do, I do everything in person that I can. It's harder now because you have board members who've invested from all over the world, and it's just hard to get everyone in one place. But I think in-person board meetings matter a ton. And by the way, I'll spend two hours in person, uh, because I like people. I actually love the people, but... And I think it's harder for people to pontificate in person.

Harry Stebbings 54:26 Yeah.

Michael Eisenberg 54:26 And so I, I'm... Uh, by the way, the other area I got to get better at by, by, by a long shot, you can get really absorbed into these companies.

Harry Stebbings 54:35 Sure.

Michael Eisenberg 54:36 And sometimes I need to kind of pull myself out at a high level, 'cause I think, you know, you could just be too intense around these companies. And my job is actually to sit at 10,000 feet.

Harry Stebbings 54:47 Why is that bad?

Michael Eisenberg 54:50 'Cause I think it can rattle entrepreneurs.

Harry Stebbings 54:52 Mm.

Michael Eisenberg 54:52 Particularly first-timers. Second-timers don't get rattled. You know, Tom Barr didn't get rattled by anything.

Harry Stebbings 54:56 Mm.

Michael Eisenberg 54:57 Ron Gordon didn't get ma- rattled by anything, and so-

Harry Stebbings 54:59 Do you prefer second-time founders?

Michael Eisenberg 55:00 No. No, I love the whole first-time thing. We ha- I have this company that's kind of building the modernization layer for LLMs now in a very unique way using blockchain. And, um, this is a company that started during the pandemic. We were about four kilometers from each other, and when it was shut down, I had a pass to get around Israel because I chaired the largest volunteer organization in Israel. And I would pick him up and bring him to my house, and we'd sit face-to-face, and slowly we built this company. Hasn't been launched yet, but they have couple of hundred customers already. He's an incredible entrepreneur. Uh, he was at Google beforehand. And, uh, you know, but he was a first-timer, and I think it took us time to find out, find our rhythm. And we would spend a couple of hours just pounding on this idea month after month after month during the pandemic, and, uh, I think he probably would've gotten started earlier if I hadn't been so aggressive.

Harry Stebbings 55:46 Final one before our quick fire. Do you prefer founders who've got experience in a market or who are net new naive?

Michael Eisenberg 55:51 Oh, net new naive.

Harry Stebbings 55:52 Yeah?

Michael Eisenberg 55:52 Net new naive. Uh, the fact that Shai and Daniel of Lemonade didn't know anything about insurance like helped a ton. When we first invested in Lemonade, I called a bunch of FinTech investors and said, "Who's the insurance guy?" They said, "Nobody. That's the beauty."

Harry Stebbings 56:04 (laughs)

Michael Eisenberg 56:04 So, none of them would invest, until Sequoia turned up. And, uh, these guys have broken every rule, every rule of how you build an insurance company, for the better, because they knew nothing about it. By the way, everyone told us, "Don't go to New York first. You'll never get a license there. That's the hardest DFS." And we said, "Go to New York." Or actually, I didn't say it. Daniel said, "Let's go to New York." It was against every insurance expert's advice. I generally have a healthy skepticism for expertise.

Harry Stebbings 56:29 (laughs)

Harry Stebbings 56:32 Listen, I want to move into our quick fire.

Michael Eisenberg 56:33 Yeah.

Harry Stebbings 56:33 So, I say a short statement. What have you changed your mind on most in the last 12 months, Michael?

Michael Eisenberg 56:38 I was concerned about younger people being too absorbed in TikTok and Instagram. And watching what happened in Israel over the last 12 months and how the younger people went out to fight for what mattered and fight for their brothers and sisters and fight for women who were raped, I think younger people, uh, particularly in Israel but maybe elsewhere, are gonna surprise us. And this is gonna be an incredible generation who's gonna be forced to stand up and fight because it... You know, Israel's one place, but there's Ukraine, and there's more of this going on around the world right now. And we saw what happened in Germany last week. Um, and so, I think younger people are gonna surprise us.

Harry Stebbings 57:08 What's the biggest misconception of the Israeli startup ecosystem?

Michael Eisenberg 57:12 I'll tell you this, by the way. Uh, we've raised an incredible amount of foreign money into our company since October 7th, an insane amount of money. It's wide open for business right now, and investors are on the ground every day. That's right now. But the... I got, you know, the, the biggest misconception is, uh, people think Israelis know how to scale things. And for the most part, when you grow up in a small country, it's harder to think of scale. And so, when you're a foreign investor, you need to help Israelis think in big scale.

Harry Stebbings 57:39 A weird and hard one. Which competitor do you most respect, and why them?

Michael Eisenberg 57:45 I said before, I respect all competitors, and I think they're 2X better than I am. I have to work harder than anyone. If I... I said this to Beezer. If I'm not the hardest working venture capitalist out there, shame on me, and I'm gonna get killed. These guys are all 2X better than I am. I was on email with Mamoon Hamid from, uh, Klainer this morning. That guy is so good.

Harry Stebbings 58:03 Mm.

Michael Eisenberg 58:04 I think, "How do I become Mamoon?" Uh, I, I just think they're all better than I am, and m- and, you know, more naturally suited for this business. So, I got to work harder.

Harry Stebbings 58:14 What's the biggest piece of BS you hear most often?

Michael Eisenberg 58:16 Board members giving product advice.

Harry Stebbings 58:19 (laughs) I, I completely agree with you. What's the biggest sin of the zero interest rate environment?

Michael Eisenberg 58:25 Probably not having sold enough of my portfolio. (laughs)

Harry Stebbings 58:28 (laughs)

Michael Eisenberg 58:28 But, but in general, you know, when things get too easy... If something is not hard, it doesn't matter. If something is not hard, it doesn't matter. And it... This is supposed to be hard, and life is supposed to be challenging. And when we succeed in challenging things, whether it's relationships or hardships or defending freedom or building a startup company that's hard to build, like Anderil or you know, Or which is the synthetic chemistry company, we feel satisfied afterwards. This is meant to be challenging. Life is not supposed to be necessarily easy, and you got to keep working at it, and keep working at it. The Protestant work ethic is a real deal. It matters, and you got to keep this... Hard work matters. And so, you know, in ZIRP... Uh, uh, the other thing, I think, by the way. Look, I never invested a nickel in China. I think s- I think China was a ZIRP phenomenon also, China investing.

Harry Stebbings 59:22 Why?

Michael Eisenberg 59:22 Yeah, I think people look for yield in other places and ignored the risk, what I call stroke of pen risk. I'm not the only guy who calls it that. Where, you know, the government will just take away your assets. And so... But 'cause people were chasing yield, they just kind of, you know, sent it over.

Harry Stebbings 59:37 (laughs) That's like a mic drop at the end. Uh, tell me, my friend, where do you want to be in 10 years? Where is Aleph in 10 years time? You, you now do the content platform as well.

Michael Eisenberg 59:47 Yeah.

Harry Stebbings 59:48 And then your partner does the data platform.

Michael Eisenberg 59:50 Yeah.

Harry Stebbings 59:50 What do you want for Aleph in 10 years?

Michael Eisenberg 59:52 Well, I don't know. Um, I hope they'll keep me around. Like, I'm old for this business at 53. And, uh, you know, I'm, I, I don't-

Harry Stebbings 1:00:00 Can-

Michael Eisenberg 1:00:00 I don't ignore the data which says as you get older in this business, you don't do as well. And so, I hope they'll keep me around.

Harry Stebbings 1:00:04 Okay, you've got sourcing, selecting, servicing.

Michael Eisenberg 1:00:06 Yeah.

Harry Stebbings 1:00:07 There's three components of venture, I should say. Um, where do you think you're best and worst? So, Keith Rabois on the show said he's worst at sourcing, and as time, relevance in that sourcing is his biggest challenge. How do you think about sourcing?

Michael Eisenberg 1:00:19 I've stopped sourcing. I, I, I get all my deals right now through referrals of 30 years of being in this business. And so, I actually don't do proactive sourcing anymore. I don't know if I was good at it or bad at it, but I don't do it anymore. Uh, I need to do two or three deals a year. There's enough that come my way that I think are super interesting that I don't do that anymore. Uh, I don't think I provide enough kind of management help to companies because I don't know how to manage my way out of a paper bag. Um, I think what I, uh, do better than average is networking. I, I have large global network, and I think that is valuable. I think if you're... If you have good chutzpah and are not bashful and you know enough people, you can reach anyone in the world. And I think that's valuable. I get, I had to get the CEO of a Fortune 10 company in the last 48 hours. It took two hops.

Harry Stebbings 1:01:07 Uh, final one. What question do you not get asked that you think you should get asked more?

Michael Eisenberg 1:01:12 I get asked the h- the first-level question of, "How do you balance your time?"

Harry Stebbings 1:01:16 Mm.

Michael Eisenberg 1:01:16 And I tell... I think there's no such thing as work-life balance. You've heard me say that 100 times. Um-

Harry Stebbings 1:01:20 I get asked that one too. (laughs)

Michael Eisenberg 1:01:21 The second-level question is, "How do other people think about the way you balance your time?" That's the better question to ask. And I think if I'm self-reflective, there are many people who I love and respect who probably think I don't give them enough time. Um, and that I don't have enough time for the things I should be having time for. Uh, whether it's my day job as a venture capitalist, community work, family, et cetera. And I think I'll... I think, as they say in, in Hebrew after 120, which is when you go to the pearly gates, I'll probably have to give judgment for not having given people who wanted that time the time.

Harry Stebbings 1:02:04 Michael, I, I always love our chats. I'm so glad we could this in person.

Michael Eisenberg 1:02:07 Yeah, thank you.

Harry Stebbings 1:02:08 So, thank you so much for joining me.

Michael Eisenberg 1:02:10 Harry, it's so good to see you. Keep doing what you're doing. You're providing just an, just an incredible service. And thank you for taking me in as an LP in your fund.