Josh Wolfe & Brett McGurk – Venture, Geopolitics, and the Next Frontier (EP.476)

25:22If you're making a hard demand, say, with Iran on whatever it might be, whether it's on the nuclear file or the hostage negotiation or warning them against ongoing attacks against us or else type demand. They point to our constant turnover in Washington. The anecdote I heard across the table was, if I'm taking a bus from Tehran to Isfahan, and it might be a three-hour bus ride, and someone gets on the bus that is obnoxious, making a mess of things, I'm going to stay on the bus, I'll get to my destination. And he said, United States, you're that obnoxious guy on the bus. You're going to have a new president in four years.

26:00You have a new administration, totally new policies. Whether I like him or don't like him, we're to stay in the course. That's how Xi Jinping thinks. That's how Putin thinks. That's how a lot of leaders around the world with longevity think. That's why it's important for us as a country to get back to some sense of things stop at the water's edge when we're out there in the world. would try to speak generally with the same voice. That's an ideal, but we've kind of lost some of that. That is how a lot of leaders around the world see us. It makes getting things done difficult, particularly when you have these wild swings between one administration to the next.

26:37I've spanned those administrations. Some of it, I say to foreign counterparts, is exaggerated. Our interests still are, generally speaking, within a zone of agreement between both parties of whatever comes into office. That is not just a cultural difference, but a geostrategic difference that we have to grapple with as the United States. As an American official, as an American business person sitting across the table from somebody in one of these countries. I'm curious of the transition from public sector to private sector. In this period of time, you've been at Lux. What are some of the contrasts that you felt most acute?

27:15Looking back in the last 10 months, I've done a few things. First priority has been my family. I have a second grader, a little girl. I missed much of her last four years. I've sought to reconnect and it's been great. Family is beginning, middle, and the end, as President Biden used to say to us all the time, very much true. I'm writing a book about the last 20 years, might come out next year. I'm a CNN analyst. I keep current with things. I joined Lux and to become a full integrated member of that team. I'm learning so much. What's so fascinating about it is the thought process. I've developed a disciplined way of approaching problems.

27:54Strategy, Ted, is the most overused word in Washington. What is strategy? The essence of it is aligning your objectives with your resource and how you're going to do it. And if that's not aligned, you're going to find yourself in trouble. I've found if you're in the Oval Office or in the situation room, asking those hard questions. What exactly are we trying to achieve? How are we going to do it? What's our resource base? What are unintended consequences, risks, assumptions, I mean, all that, especially in military type decision-making, war and peace decision-making. The way Josh described venture, it might look like you're painting on such a broad canvas.

28:29You can make a lot of bets, some of which don't pan out. And yet the same thought process comes to bear, I have found. We've had examples of a founder coming in with crazy ideas that at first brush might sound nuts. Then you get to know the founder, you get to know their background. Counter to what I just said about that more disciplined strategic thought process, take JFK's moon speech in 1962. People say, look, a leader's got to point to where we want to go. He studied JFK's moonshot speech. Before he made that speech, he had on his desk a commissioned study completed, overseen by LBJ as vice president, with the best rocket scientists in the world, Wernher von Braun and these Nazi scientists who came over to work with us.

29:15That's another lesson of history. And they said, we can get to the moon. Here's what it takes. It's going to take bigger engines. It's going to take big resources. Not with our risk, but you can do it. When JFK set that objective, you had an ends ways and meet, a way to tell that story. That's very important in making investment decisions. It's got to be grounded in the founder of the team and where it's going to go. Reflect Orbital is a company that is launching satellites to redirect the sun to the dark side of the earth for all sorts of applications from solar to military applications. They're launching their first satellites here and second quarter of the coming year.

29:56That's just one example. This is a great example to what Brett said. Half or more of the team, when they heard the pitch, they're like, that's ridiculous. We want to be at the cusp between what seems impossible. And then you hear the pitch from the founder. Oh, no, no, this is inevitable. That gap from impossible to inevitable is powerful. Believing in this founder who came from SpaceX, got these extraordinary people who were some of the best and brightest material scientists, satellite builders, rocketeers, et cetera, it all made logical sense. And it was all obeying the laws of physics. The way to think about this simply is you have those Mylar birthday balloons, those mirror-like, super thin, but super reflective.

30:39If you could have that unfurl into a giant mirror on a big boom that sprawls out and redirect the sun to flash and have a satellite hovering, shining a flashlight in the middle of darkness, anywhere you want in the world. Tasking a satellite, sunlight as a service. Solar cells could be performative 100 % of the time instead of 50 % of the time. Yes, you can shine it on something trivial like a concert in the middle of the desert at night or search and rescue for military operations when you don't want the cloak of darkness. If it works, it could be huge. There's a lot of countries where Brett and others have relations that might really care about this.

31:19we are a priori thinking about how could we light up, no pun intended here, their business by opening doors that they might themselves never have access to. Brett, I'd love to ask under this thought process of the duration of venture investing long-term and the macro perspective you have, what do you think are the biggest geopolitical risks that you're now thinking about as part of Lux over a long period of time, not just the next year or two? Counterterrorism keeps me up at night. I go back to that foundational experience. I'm on one track and the world's on one track. And then we got hit out of left field.

31:58There's still groups out there that want to do that again. We got to stay on top of that in a smart, sustainable way. There's increasing nuclear risk. We're entering a world of all the guardrails of arms control are eroding. If you project ahead 10, 15 years, particularly as the Chinese built up their arsenal. And of course, AI and where that's heading and military applications of AI is something that jumped to the top of our national security agenda when I was in the White House in the last four years, particularly starting in 2022. And that intersected a lot with the Middle East because of the advancements and the ambitions of some of these Middle East partners.

32:34The risks out there, particularly the risks I've been associated with, are quite acute. However, the opportunities are massive. We've been talking a lot about the Middle East and we at the G20 back in 2023 rolled out this initiative called the IMEC, India, Middle East, through Jordan, Israel, into Europe, overseen by the dynamic American entrepreneurial ecosystem. It is a ship to rail, transport, it's tech, it's trading, ideas, and technology. That is very much the wave of the future, trading relations between, say, UAE and India, how they've spiked since they closed a pre-trade agreement a few years ago.

33:12Once we emerge from this horrific Gaza crisis, that agenda is very much the direction of travel. I hear it on almost every capital. That's where they want to head. People want to build stuff that's going to change the world in the heart of our national security space. There is a recognition that we have to have our best minds in tech, our best companies, our best startups, working seamlessly with our national security sector and the advancement and protection of our country, working with allies, partners, strengthening them as a comparative advantage against China and others' adversaries. The other vectors that are interesting, space as a vector is going to be hotly geopolitically contested.

33:51There's great opportunities for collaboration, also great near-peer competition. The ISS was this technological and scientific diplomacy that has now fractured. China and Russia are out. China has gone to the dark side of the moon. Russia is going to the moon with China. U.S. is going back, and we've got our internal competition between SpaceX and Blue Origin and others. Space is going to be a contested domain in a very big way. Huge opportunity, but huge risks. There's going to be sabotage, counter-sabotage, detection, deflection. It literally is going to be Star Wars, and there's going to be a lot of contested battle space there.

34:24Josh, I'd love to turn to the other side as Brett's talking about progress and trends. The fun way to ask you about what you're seeing is your line of directional arrows of progress. Everyone talks about AI. You just threw out early on this apex of absurdity. Please go on with that. Wind me up. We've known each other for a very long time. I think it was back in 2015 or 2016. We were in a company called Zoox, which was doing self-driving cars. It was inside of the doors of Zoox where we recognized that they were training these cars outside, ingesting all the sensor data, and they were processing it at the rate of one second per second, what we call reality.

35:02I got upset because I thought that they were messing around and playing Grand Theft Auto. It looked like they were literally playing video games. And they said, no, no, no, relax. What we're doing is training the vehicles in silico on computers using physics engines. Are those supercomputers? I asked. They said, no, they're NVIDIA chips that aren't even on the market yet. Now, this was at a time when NVIDIA was a$15 billion market cap company. NVIDIA was ascendant and it made a ton of sense to see with convolutional neural nets and the rise of the transformer and ultimately what we know as large language models, transformers being the T of GPT.

35:31All of that made sense. Fast forward to today, the amount of money that is being spent and encouraged by Jensen and NVIDIA, that you need hundreds of thousands of H100 chips and these clusters of data centers, which in turn fuel a demand for next generation energy of people funding modular reactors, speculative companies that I think are borderline fraudulent. All of that as premise that you need all of this. I use five different large language models at any given point in time. I open up five different tabs in the browser, or if I'm on my phone, I'm using five different apps. I've got Grok, Gemini, Perplexity, ChatGPT, Claude.

36:02They're all competing for 100 % of my mindshare. I'm allocating them 20%, but they're spending 500 % on CapEx. When Anthropic trained Claude, the first model was$100 million to train, they got$100 million of revenue. Amazing. Then they spent$1 billion to train the next model. Got$1 billion of revenue. Still amazing. Then they spend$10 billion. Now they have$7 billion of revenue. The asymptote of revenue growth starts to hit this plateau. It will continue to grow, but the pace and the rate of it slows down while CapEx is still going super linear. You have a scenario where what one company does individually and rationally collectively becomes irrational.

36:35What are the things that will break it? There are at least two major things. One of them in preview is memory and one is Google. On the memory side, I feel the same way today where Jensen wants you to believe that you need all these chips for training and inference. You do need them for training, but for inference, meaning the prompts that you do. I'm of the belief that 50 % of your inference and your queries will be on device. They will not be on device going to the cloud. They will be on device, locally hosted on models that are either etched into the silicon or using flash memory. There was a paper a year ago that gave me that insight.

37:06The paper said we could do large language models on small devices using flash memory. I immediately went and looked and said, okay, you've got SK Hynix, Samsung on the Korean side. SK Hynix has been delivering through the roof, backlog for the next two years. Samsung will probably come in and take some of that. Micron, US, I was worried that they were going to face some export restrictions from Trump. So far, doing well and thriving. Two main markets for all these memory guys. High bandwidth memory that's attached to the GPUs and is needed. And then this next cycle of on-device. Small chips, flash memory, being able to run locally.

37:37That's the first thing that cracks the consensus narrative of endless demand for chips and data centers and power. The second is Google. Google can Microsoft these large language model players, and the presumptive revenue growth at OpenAI and Anthropic, et cetera, would get hit. Today, roughly$13 billion of revenue at OpenAI, 70 % of that revenue is from$20 a month paying subscribers like me. Google, 20 years ago, looked at Microsoft's main cash cow of Office, PowerPoint, Excel, Word, and Outlook, and they gave it all away for free because they could, because they have this advertising model in search.

38:14Google can give away Gemini, which was the laughing stock when it was called BARD about two years ago. Today, Gemini is the most performative model, largest context window. Young companies run on G Suite. I allow it and trust it to search my Google Drive and my files and my Gmail and so forth. I don't trust OpenAI to do that. In the same way, I didn't trust Meta and many of their products. I did trust Apple. I did trust Google. I did trust Amazon. Trust is a feature of these companies. Google can give it away for free and completely undercut. And people will say, why am I spending 20 bucks a month for this when I'm getting it for free?

38:47They've got Nano Banana for images. They've got Veo 3.1 for video. They've got Notebook LM to be able to put in hundreds of PDFs and audio files and be able to generate stuff. It truly is getting better and better and more cohesive from what seemed a bit chaotic two years ago. Those are my two bets that both public market and private market, you will see edge inference chips and memory chips becoming more and more important. And then the second is Google is a sleeper here. Apple is a little bit of a sleeper. I was sharing this with an LP, this thesis around Google. And they're like, Josh, I got to tell you, I just got a Bloomberg alert that said that Apple had just chosen Google to be the main AI as a replacement for Siri.

39:26That's a big deal. Google was paying Apple$10,$20 billion a year for the search inside of Safari, that ecosystem is a lot more vulnerable than consensus linear predictions will make. The other adjacency that I talked about is something that we call at Lux Life Cording. There will be lots of little devices. They will start with third-party pendants and rings and necklaces, hardware microphones, recording passively 24-7 every day. The young people will be comfortable with this as digital natives. You can start to see the breadcrumbs. Some people, when they're on Zoom, they have fireflies or granola or otter that are recording things.

40:03You can then summarize the transcript. You can unearth insights. There's going to be more demand for the flash memory in addition because there's going to be so many devices, whether they're integrated into watches, jewelry, wearables, 24-7 recording. Same thing with glasses. Glasses will be able to detect if a scene is changing, you got to record every three seconds, little snapshots, little videos. If it's static and not changing that much, you only have to record every three minutes or 30 minutes. You will have entire daily archives of your daily experience ingested, processed by AI, unearthing insights that you didn't realize.

40:39And there's going to be a whole new demand for the components that go into all that stuff. Those are some of the contrarian takes on AI. If we're at the apex, where does it go? On the capital market side, you see hundreds of billions of dollars destroyed of people that play the same game that you did back in 2000 with fiber optic cables. The counter to that, if you listen to some of my friends, would say, well, back then you had dark fiber. Today, there's no dark GPUs. I still beg to differ. What the individual company does rationally, collectively, is irrational. There's going to be a glut and there's going to be a collapse.

41:11Debt has not really entered the system until now. Debt is the thing that causes bubbles to collapse. You've got Facebook issuing$25 billion of debt. You have this crazy structure with Grok and XAI from Elon. You've got CoreWeave. These things are messy. Dave Einhorn had a good recent snippet in one of his recent quarterly letters that was talking about how a dollar of OpenAI revenue results in$8 of seeming revenue. It's a dollar to OpenAI. Then they pay Microsoft$2, so they're losing money, 50 % negative margins. Microsoft then pays CoreWeave. CoreWeave then pays NVIDIA. that$1 that I'm paying translates into$8 of revenue and$100 of equity market.

41:54That's a house of cards that's going to collapse. The next wave that I'm bullish about is going from two-dimensional AI in this current wave, which is pretty much saturated. That is everything from voice, video, images, text, code, anything that's popping off a screen to thinking about the next wave of three-dimensional AI. What that means is going into biology and robotics. Biology and robotics are interesting because there is scarce data sets. If I wanted to raise money and train the next large language model, I could take the repository of the internet, everything on Reddit, everything on X, and be able to train the model.

42:29If I wanted to train a new robot, there is no repository that I can go to. I have to create it. What is scarce becomes valuable. There's a company called Physical Intelligence known as Pi that is the leader doing that. We had a company called Evolutionary Scale. It's a quick, amazing story. I backed this guy 10 years ago in a biotech company. His name is Alex Reeves, called Caliove, focused on the gut-brain axis. Alex then goes, gets a PhD in computer science at NYU. Why? Because he believed that the future of AI was going to be computer science and AI meeting bio. Amazing, prescient. He goes to Meta, builds a group, has$50 million a year of budget, builds a 12-person team, is competing with AlphaFold Deep Mind, publishes a paper on protein structure prediction.

43:10Amazing. You should spin this out of the company. He says, Josh, love you. I'm getting a lot of budget here at Meta. I'm good. Six weeks later, he comes and visits me with Eric Lander of the Broad Institute. He says, Josh, they're shutting down my group in a few weeks. I offer money on the spot. He does the right responsible thing. Says, thank you very much. Goes to market, gets me to pay a higher price and own less. We bring in Nat Friedman, Daniel Gross, Amazon, NVIDIA, and a bunch of other investors and CZI just bought it. We funded them. And then the nonprofit arm ends up acquiring them. Why?

43:39Because Mark is so competitively motivated to beat Demis, Hasibis, and DeepMind and wants to win a Nobel Prize. And he thinks that this is the group to do it. Crazy story. Bio and AI is going to continue to be a big deal. What are some of the other directional arrows of progress that are leading your mind share? One of the big ones is maintenance. Why am I excited about maintenance? The past 10 years, driven by the cost of capital, was an abundance of growth capex. Everybody was funding new buildings, new HVAC systems, rail, train, planes, automobiles. They were putting data centers, military installations, satellites, anything that you can imagine, people were deploying.

44:17Now, if you are a good CFO or a good capital allocation committee or a good activist investor or good governance, you are like, enough with the spending. Why don't we take the assets that we've put in the ground for the past 10 years and maintain them? If you are a student of accounting or financial statements, there's two parts of CapEx, growth and maintenance. The pendulum is going to shift. Maintenance is going to be in high demand. That means that there's an opportunity for new technologies to help maintain existing assets. Software for asset maintenance, low-hanging fruit. Software plus sensors for acoustic detection, aberrations of machines that might be breaking down.

44:50You start to hear a rattle, a fan belt, something's off a little bit from its normal pattern. Robots for remote repair. We're not going to send astronauts off. Remember, there's only eight people in space to maintain these satellites. We're going to send robots that can go and wayfind, detect them, manipulate them, and in some cases, prevent or conduct sabotage. That's another directional arrow of progress for maintenance. On the AI and bio side, there's one that I feel very strong conviction on. And the thing about these directional arrows of progress, you don't know who the entrepreneur is, you don't know who the company is, but you have high conviction it's inevitable.

45:23That thinking led us from seeing an energy. We went from carbohydrates to hydrocarbons to uranium. Directional arrow progress, more and more energy density per unit of raw material. I got excited about nuclear. What's the big unsolved problem with nuclear? Nuclear waste. We started a company called Curion. Big success for us in Japan. Green machine interfaces. Idea here, we went from standing up next to a giant ENIAC computer 50 years ago, then 25 years ago, having desktops and towers. 12 years ago, having laptops where you give rise away from the mouse to a trackpad. You're still tickling the keys, but now it's touching your lap and your thighs.

46:00Then phone held in your hand 18 hours a day. First thing you touch in the morning, last thing you touch at night. Then watch and AirPods. The next thing was gestures. I'm sitting here gesticulating wildly while I'm talking. The ability to go like this and turn a computer on or off of change a song by swiping in free space without a camera. That became the basis for our investment in Control Labs. Coincidentally, Meta bought that for a little under a billion dollars, started by this guy Reardon. He joined along with Brett as a venture partner about to start his next company, another directional hour of progress.

46:28My newest directional hour of progress, the way that science itself is done. If you are a musician, you take your instrument and you go to a studio and you play guitar. If you want to write a song, you don't need to get the instrument. You You can go on to GarageBand and Pro Tools and Logic. You can put a virtualized instrument into the machine in a temporal sequence and play the song, collaborate with others, digitize the entire experience. Turned it from a physical CapEx decision to basically paying a cloud fee to access these instruments. Biology and science are going to experience the same thing as happened in computing.

47:06Compute, you used to have to go have a data center. Then you went to co-located. Then you went to cloud, going from a CapEx decision to an OpEx decision, treating it like utility that you can get off of Amazon or Azure, Microsoft or Google Cloud. Today, if you're a scientist, you have to go to a scientific lab, a wet bench, a white lab coat. You're working with pipettes and beakers, centrifuges, shaking tables. All of that is going to be done by robots. Completely automated labs operated in the cloud where a scientist can pick up the iPad on the beach in the Bahamas, dial up the experiment. it runs.

47:38The results come out. They are immediately published in a paper. The AI scientist then contacts the scientist who came up with the experiment and says, hey, a paper from two years ago, a paper from 20 years ago, these might be spurious correlations. Would you like to rerun the experiment tweaking these parameters? And you just, having been reverse prompted, click yes. And then you go back to sleep, wake up in the morning, new scientific paper. The companies and the countries that get that right are going to produce massive amounts of knowledge in materials, in drugs. They're going to produce massive amounts of wealth, and they're going to have competitive advantage.

48:11So there is going to be a race, but that is an absolute directional arrow of progress of decoupling the scientist from the wet bench and putting them in front of the computer. Brett, as you've gotten exposed to Josh's way of thinking of directional arrows of progress over the last 10 months, I'm curious, as you put that lens onto the perspective that you bring, what have you seen in your world as some directional hours of progress? A recent experience, we had the largest aerial attacks, Iran launching against Israel. 200 ballistic missiles in the air at the same time. That was a worst case scenario to get that many missiles in the air at the same time.

48:48They launched them in waves of like 30 to 40. You're in the situation room with the president on screen as General Eric Krill, our CENTCOM commander, our secretary of defense, all the way down to commands in the theater. We had built a coalition of destroyers and aircraft to coordinate the defense. It's a 13-minute flight from Iran to Israel. For the first eight or nine minutes, you can see the streaks coming across the screen, like the movie War Games back in the 80s. There's nothing you can do. You're captive to what's about to unfold. And as the defenses engage, the screen lights up white. You're just hoping that the technology that's been developed over decades works in that fog of war.

49:36Our first report that came into the Situation Room was that we had a direct hit on a dormitory at Tel Aviv University. That turned out not to be true, but that was what we were facing. When the dust settled, The defenses we had set up and established and things that are known and not known were extremely effective. I was thinking in that moment, all the work that went into anti-missile systems, radars, technology, being able to coordinate across multiple militaries in real time, all the practice that that took, not knowing it would ever be used, it worked. The missiles will get better. The drones will get better.

50:17They're going to get smaller, faster. more lethal. If we don't keep up with this threat, the world is going to be a more dangerous place. That's why I feel so strongly about our defense tech sector and investments in that sector, making sure we're doing everything we can. One thing that always attracted me to Lux was their early investment in Andrel and our great neoprimes and everything that is happening in that space. But that directional area of progress, adversarial systems are going to get better and better and better and harder to defeat. We need to keep up with that and live that in real time, that sinking feeling.

50:53I've been in a situation a lot, over 20 years. That's the one time I bowed my head and said a little prayer because you just didn't know what was about to happen. It was complete silence. That ended up being a fairly successful defense. The Israelis responded by taking out all of Iran's air defenses last October, which set the stage to what happened this past June. Progress is not the word, directional arrows of threats and having to keep up with them. The very attack that Brett was involved in helping to defend massive coordination amongst multiple countries, multiple technological layers, drones, cruise missiles, ballistic missiles, the speeds at which they move, the heights and the altitudes.

51:34You think about Israel's different systems from Iron Dome, Iron Beam, David's arrows, sling. This is a long arrow, pun intended. People used to punch each other, and then they threw rocks at each other, and then they threw arrows at each other, and then they shot bullets, and then rockets, and then missiles, and then hypersonic missiles. Mankind, unfortunately, has been throwing projectiles at each other since the dawn of time. What's different now is economic and technological asymmetry. Economically, to fire the drones, which take roughly, Brett can correct me here, but nine hours maybe to get from Tehran to Israel, whereas the cruise missiles are maybe 90 minutes and the ballistics are like nine, 10, 12, 15 minutes.

52:15Cruise missiles, 90 minutes, the drones, a few hours, the ballistic missiles, yeah, about 12, 13 minutes. The cost is not just the time asymmetry of having to react quick to something that's 15 minutes away. It's crazy. That's why they have shelters and they can rush to the shelters and sound an alarm, assuming that you can do early detection. Then to be able to counter that, to scramble jets with missiles or to fire missiles that can counter decades of technological development. The cost of those things, if you're talking about one to$5 million a missile to shoot down something that's$500 ,000 is a big problem.

52:45Josh, if we bring this back to when you're discussing bringing in venture partners that have a different perspective from anyone else on the table at Lux, I'd love to get your perspective on how your eyes have been opened over the last 10 months by Brett coming onto the team. First and foremost, Brett is a diplomat. He just has a really nuanced appreciation for things that aren't black and white. Some of that is over a timeframe, thinking about something in a 10-year period versus a one-year period. Some of it is a deep understanding of history, of the personalities, of what's motivating them.

53:21I have a nuanced appreciation in the same way years ago when we did all kinds of family therapy that made me just a better person and interacting with partners at work. Having an appreciation that you can't just be zero sum on these things. You can't be black and white, that there's a lot of nuance, that these things are really complex. Enemies yesterday can be allies tomorrow. There are coalitions and alliances that form understanding interests and means and ends and capabilities. Who has leverage and where and why? Having an appreciation for the distinctions of the assets and the liabilities that many of these countries have.

53:57All these groups have leverage in different ways, and they all have different interests to have geopolitical power. Some of that is going to be through economics. Some of that is going to be through information and media influence. Some of it is going to be through religion. Some of it is going to be through technology and trade. So having an appreciation for how complex the world is and not taking what I would consider very naive black and white views is really important. And then seeing the primacy and value of relationships. We had venture partners early on that were senior people when we were younger and they had Gravitas.

54:25We wore suits instead of t-shirts because we were trying to show we're adults. Brett is a real doer. I get off the phone with a shake in one of these regions. Within four hours, Brett has a document ready to seize the moment in the window. Time is not our friend. We don't want to let this slip to entropy. Let's get it done. Let's move. He's biased towards action in ways that create win-win-win for everybody. He's helped people in ways that the public will never know, that I will never know. has created a lot of goodwill and trust. That is a really valuable currency. I admire that a lot about Brett.

54:59Josh, I'd be remiss as we get closer to wrapping up, since we only get to do this publicly every so often. What else is on your mind that we might not have talked about so far? On an economic vector, we've got extraordinary guy in Scott Besson in treasury. I have no idea what's going to happen with the Fed. The cost of capital is going to rise in venture. When you have so much money, 60-something percent of venture dollars went into AI, you have to start looking at where's their scarcity of money. Because the number one determinant of future returns is never the hockey stick curve that the banking sell side or the consultants put out.

55:35It's how much capital is going into a sector. The more capital that goes in, the lower the future expected returns. Biotech is interesting. 91 ,000 people laid off in Boston alone in August from pharma and biotech companies, higher cost of capital, stock prices down, working on 10 different programs, they got to shut down to one. Even if you just take 1 % of that 91 ,000 people, you've got 910 people who are absolutely brilliant and amazing that you can put into companies and start new things and spin out, bullish on whatever others are bearish on. Personally worried about private credit, it gets overdone.

56:07Why is due in the beginning? The fool does in the end. Secondaries are going to do really well. It's part of our strategy to do these special situations, investing in a late stage business at an early stage price where somebody else took the risk but didn't get paid for it. There are some secondary funds that are going to do really well because LPs are overexposed, don't have DPI, don't have returns of cash coming in from their managers, are going to divest from some of the ones that are wrongly positioned. Worried about the horseshoe theory of the far left and the far right domestically. Why can't we have moderate candidates?

56:38They don't win elections. We need strong leaders that are proud Americans, but are not in the extremes. United we stand, divided we fall. We got to find the things that bring us together, whether that's innovation, economic growth, common enemies. I am worried about the information vector. I'm worried about anybody that is able to weaponize the very things that we create. People did it with our commercial airliners 24 years ago. They're doing it with our information access today. The thing that is optimistic for me is I have no idea in two years where we're going to be investing. I know we will be at the cutting edge.

57:14A lot of it is finding people that have amazing vision and signals. Some of them are inspired by science fiction. Some of them have a chip on their shoulder. You look at Palmer Luckey. This is a guy who founded Andruil. He was 20, 21 years old, made nearly a billion dollars with the sale of Oculus to Meta, got fired because he was an early Trump supporter back in 2016. I decided I'm going to start the real-life Stark Industries out of Ironman. There is no secret lab that the U.S. government has, but DARPA or the Pentagon that is developing the James Bond Q stuff, it is going to rely on the next Howard Hughes type person, the next person that is going to build the general dynamics and general atomics.

57:50And that's what Andrew and I hope many others will do. Palmer six and a half, seven years ago was saying that the next domain that he wanted to pursue was subterranean warfare. I laughed. I'm like, that's ridiculous. Underground, boring. And you look at Gaza envelope in Israel, and this is real. Whether they're inspired by sci-fi or have this prescient speculation, that's the business that we're in. Find the low probability, high magnitude impacts. Either identify them as threats and do things to thwart them or get behind them as investors and get along for the ride. Our closing questions are brought to you by Oldwell Labs or OWL.

58:27OWL is the very best software I've seen for allocators to find and track managers. And I've seen a lot of them. If you O-L-D-W-E-L-L-L-L-B-O-B-E.com slash TED. And trust me, it's worth the look. Brett, I've subjected Joss to this in the past, but a couple questions that we'll close with that people get to know you a little bit better. What's your first paid job? What'd you learn from it? Well, I mowed lawns. That was the first dollars in my pocket. I learned to mow lawns in a square so that all the grass is in the middle and you can rake it up easily. I scooped ice cream, Baskin-Robbins. I was a lifeguard.

59:08The value and the feeling of doing something well, I still remember that smell of grass mowing some of those first lawns. Brett may not want to go here, but I'll just say we have similar family situations, relationships with dads. He, to me, is also the epitome of somebody that had a chip on his shoulder, put chips in his pockets in different kind of way. Wrong side of the track, origin, went to school and important schools with people that grew up very differently. I think he's probably rejected by some of those peers. I think the vast majority of them envy him, look up to him today. I was in Jacksonville, Florida, program for the World Affairs Council.

59:45As part of the program, they run with their local high schools and all these top kids are interested in history and foreign affairs. They're high school kids, many of whom are first-generation families, about 80 kids in there. I just spoke to them about this very point. One of the kids asked, how do I not doubt myself? Such a revealing question, because I wouldn't even have thought to ask that question in high school. The attributes of discipline I got from football coaches or hockey coaches, you didn't realize you're developing your mind for the future. Don't doubt yourself. Prepare. You got to work hard.

1:00:23Nothing beats hard work. I was a law clerk after Columbia in the Second Circuit for Dennis Jacobs. I remember him telling me, the secret to this job is ass power. You got to sit and work. He was right. Ass power comes in handy. You got to know what you're talking about. When I'm first in the Oval Office briefing President Bush a lot, I was talking to Josh Bolton, who was the chief of staff in 2006 timeframe. He had such a good rapport with the president, good news, bad news. I remember asking Josh, like, hey, what's your advice? He just said, hey, Brett, it's simple. Tell the president the truth in five minutes every morning.

1:01:00To do that takes a lot of work. You got to know what you're talking about. That was good advice. Now, as a father, I have a young daughter. Thinking about how to instill that in her is something I think about all the time. Brett, what's one thing most people don't know about you that you find interesting? I picked up guitar in COVID. That has become my little hobby. Drives my wife a little crazy because I like to play electric guitar. I got one more for both of you. What's a mystery that you wonder about? Josh, why don't you start? The default obvious one, how do you get something from nothing is confounding.

1:01:39I still find human nature endlessly perplexing. Markets change and business change and technology change, but human nature is a constant. What makes us tick? Why we do the things we do? There are compelling arguments, evolutionary psychology, status, tribalism. Danny Kahneman was a friend, and he would say that the reason you believe the things you believe is not because you reason through them and empirically calculate in your head, but because the three or four people in your life that you care about or respect or whose respect or caring you want, you believe what they believe. That's probably the most important thing in our business, and it's something that I don't think I'm ever going to feel like I solved, is just understanding people, what drives them, what motivates them, why they do what they do, how to get people to change their minds.

1:02:23Really hard thing to do. When you're in a marriage, when you have kids, you learn these things all the time. You're negotiating with your kids and then with your spouse. The origin of the universe, how do you get something from nothing without a theistic explanation? And just understanding people, it's still a confounding mystery. Right. Lots of mysteries in the immediate. The great thing about parenthood, you see the world through these very fresh little eyes. My daughter had a phase of planets and learned the planets, and we'd go out and look at stars, asking the basic question, where does space end?

1:02:55Can't explain. It's kind of infinite. How do you explain that? Not a physicist, not a scientist, but those basic human questions that people have always been asking are still unanswerable, and that to me is fascinating. I'll give you two more quick ones. Do aliens exist? My answer is yes, but not extraterrestrial aliens. If you ever watch any of these Nova shows, you see these underwater creatures. There are aliens. They're here. They didn't come from outer space, but my gosh, truly Darwinian endless forms most beautiful. It's insane. And the last mystery, which I think I've solved, and I'm going to upset a lot of people.

1:03:34I am convinced that Tiramisu, which dates back to the 1960s, here's what I think happened. They ran out of chocolate powder. They were trying to make a cake and the guy was like, oh shit, let's just use coffee. They won't know the difference. We'll give it a fancy name, charge them more money. We'll convince them that it's really a delicacy. I think that tiramisu is one of the great hoaxes pulled on us. And I'd like to know the mystery of where it actually came from because it's disgusting. All right. With that, Josh, Brett, thanks so much. Always a pleasure. Great to be with you, Ted. Thanks, guys.

1:04:05Thanks, Ted. Thanks for listening to the show. If you like what you heard, hop on our website at capitalallocators.com, where you can access past shows, join our mailing list, and sign up for premium content. Have a good one, and see you next time.

1:04:39this podcast.