Pod Insights Summation with Auren Hoffman Beniamin Mincu CEO, MultiversX - building the trust layer for AI About this episode Beniamin Mincu is the CEO and co-founder of MultiversX, a next-generation blockchain protocol used by governments, enterprises, and developers. He previously co-founded Metachain Capital, which was an early investor in projects like Polkadot and Binance. He’s also the host of the Full Shard Podcast and one of the most followed voices in Web3. In this episode of World of DaaS, Beniamin and Auren discuss: Why AI needs a blockchain trust layerAI sovereign chains and agent accountabilityThe geopo...
Transcript Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.
0:02 Auren Hoffman? Hello fellow data nerds. Welcome to World of DAS. I'm your host, Orin Hoffman, CEO of Incubate and GP of Flex Capital. Discover more episodes, get weekly data as a service news, original content, articles on data and more at worldofdas.com. That's world of das, D-A-A-S, dot com. Hello fellow data nerds. My guest today is Benjamin Mincu. Benjamin Mincu is the CEO of Multiverse X, a next generation blockchain protocol. Before that, he founded Metachain Capital, which was an early investor in projects like Polkadot and Binance. He's also the host of Full Shard Podcast, which is really one of the more followed voices of Web 3. Benjamin,
0:46 Auren Hoffman? welcome to World of DAS.
0:49 Speaker 01 Orin, it's really great to be here. Pleasure.
0:52 Auren Hoffman? Great to have you. Why does AI need a trust layer? Why are people talking about blockchain and AI together?
0:59 Speaker 01 The most important point here is AI is not stopping. It's accelerating and has increasing capabilities to the point where it will very soon feel like a genius and then almost gradually like an oracle that trends to a god. Of course, you could argue how long that will take and so on. But if you grant that AI will have this superhuman capabilities and that it will be able to do good and bad, the most important point is how do you ground that such that you know it's safe, especially when you imagine you have an AI assistant, it has access to your entire life, everything you do, everything you say and so on. Question
1:44 Speaker 01 is, how can you sleep well at night, knowing that there's someone that's maybe brighter, smarter and so on. And I think this is how you ground it, a trust layer that sets some boundaries, some guardrails for what the AI can do and cannot do.
2:01 Auren Hoffman? Why is it overlap with a blockchain?
2:03 Speaker 01 I've framed this in two ways. I thought in the future, you're going to only have two types of technologies. Creation machines, which are different versions of AI, where you can just take an AI and it will gradually do expanding capabilities to do almost everything we can do and then go beyond that. And then trust machines or truth machines, which is a set of properties that a computation layer has, which incidentally is the blockchain, which is censorship resistance, verifiable, transparent, and then can scale with the scale of AI.
2:43 Auren Hoffman? Digital signing type of thing or digital fingerprint or something?
2:48 Speaker 01 Some type of digital fingerprinting is just a very simple form of it. Imagine that every interaction with AI, every response, request and all of that can be verified, cannot be counterfeited. So neither the company nor the AI can then change the response to make it seem like they've answered something else and so on. If something bad happens, suppose that asking for medical advice and then you listen to that and then someone dies. Who's at fault in that type of situation? The point is just having a kind of accountability layer, something that establishes the ground rules, establishes who's at fault if something bad happens
3:30 Speaker 01 and then who's responsible for maintaining that type of transparency. That's what the blockchain can do in a way that has not been done up until this point.
3:39 Auren Hoffman? We're moving to a world very quickly where a very large portion of the content on the internet is AI-generated. Maybe already we're at over 50% today and soon we'll be at 80, 90% of the content. How does that trust problem evolve as more and more of the content is not generated by a person?
4:01 Speaker 01 There's this dead internet theory that, interestingly enough, Sam Altman just discussed about recently, which just states that more and more of what you perceive and see there is actually generated by bots, some type of AI, more sophisticated AIs, that make it seem like this is humans speaking to you, but it's bots. How do you ensure that there's someone there, that you're not talking to a bot, that you understand the identity, and then again can trace what the implications of what's happening?
4:37 Auren Hoffman? Sometimes I prefer to talk to a bot, but I might want them to let me know that I'm talking to a bot.
4:42 Speaker 01 I think this is precisely the point. Knowing with whom you discuss is the very fine point that then makes you calibrate some of the things.
4:52 Auren Hoffman? Yeah, I was on a call the other day where I called customer service and basically said, hi, by the way, I'm a bot, just so you know, because it wasn't obvious. They let me know right at front and it was great. It was an amazing experience and it was great. I got to talk to someone like immediately and they were able to deal with my problem faster than potentially a human would be.
5:14 Speaker 01 Indeed, the identity problem has been a very core problem to the internet. When something is fully digital, the question is how do you validate who's there?
5:25 Auren Hoffman? The first internet cartoon was No One Knows You're a Dog on the Internet.
5:29 Speaker 01 Hundred percent, and those meme have a great grounding in the reality because people tend to be something else. And then how you solve this, there's a problem that's called the SIBIL attack problem, where someone can direct thousands of bots toward the website, toward the registration form, toward something, and then overwhelm that and you don't even know. As we scale to this sophisticated agents, the question is how do we verify who's human, who's an agent, and then who's a malicious bot? Because if you do not have an identity, that's a way of creating this separation between
6:10 Speaker 01 people or agents which have validated and identified as X, and then these malicious actors that do not want to verify or identify because it's much more advantageous for them to have the pseudonymity there.
6:26 Auren Hoffman? I started my career in like the ad tech world and in the ad tech world, there's a lot of bots. The frustrating thing was so many legitimate companies in the ecosystem were benefiting from the bots. They had a disincentive to root out the bots because they were making more money because the bots were there. There's so many bots that are in social media. In some cases, in social media, there's more bots commenting than real people. And often the social media companies benefit from these bots. How do you deal with those types of cases?
7:00 Speaker 01 You're very spot on because when you said ad, the first thing I thought was misaligned incentives, perverse incentives. And I think the next thing that you said is the social networks, sometimes there may be 80% with bots that looks so plausibly real that you never think, hey, this might be actually a bot. And this is why, in some sense, you want to ground this solution independent of them. Because if they make sufficient profits, they will not want to solve this problem. It's contrary to what their profits would be if they eliminate all the bots.
7:39 Speaker 01 Having an independent, trusted, transparent, very provably verifiable layer, like the blockchain, is just a simple way of actually solving this problem and then enabling anyone to just verify this. Because there you cannot modify the traces, you cannot delete the tracks. That gives it a different kind of trust perspective.
8:04 Auren Hoffman? What is this idea of like an AI sovereign chain? How does that work?
8:08 Speaker 01 The idea is very simple in some sense. If you think about AI and specifically agents. Agents are this actors that can operate and do actions for some type of trusted party. The question is, is the internet built for agents? If you think about the internet, we had three phases: read, write, and own. And I think we can certainly see the agents going from page to page and so on. But we've not seen them have a great type of substrate where the actions can be tracked and verified deterministically. And then they can have objectives that then can be verified and so on.
8:51 Speaker 01 As you may know, AI as it is right now is probabilistic. It's non deterministic, so it may or may not tell you the truth. It may hallucinate from time to time. So you need to ground all the AI interactions in a substrate that is deterministic. Where you say, A cannot be both A and non A, and we're not sure. It can be either A or B. And then we're sure that it's A or B, and I can track, monitor, and so on. And the point with this is more to think, suppose that I have 100,000 agents, 1 million or 1 billion agents. How do I facilitate one, this type of transparency and interaction with them?
9:34 Speaker 01 That's very common sense with identity and everything that comes through it. But then even more important, how do I facilitate a coordination between multiple such agents in a way that can be deterministic? And I can say in a chain of 10 agent interaction, I can say, no, this guy has actually stolen the money. Audit, in a way. Yes, audit is one level. But then the second level is, you can think about it as single player interactions with agents are between you and them, and you audit and verify them. Multiplayer interaction, they need to coordinate also between each other. They need to understand even more between each other because if they don't operate with a deterministic function,
10:18 Speaker 01 then you cannot actually instruct 100 agents to do something and discuss between them and so on.
10:26 Auren Hoffman? So many people today are talking about MCP, the Model Context Protocol. Can you help us understand what that is?
10:33 Speaker 01 Hundred percent. Maybe a simple way of looking at that is the USB for AI interaction? And what does the USB actually mean? It's just this standard that translates what the AI can do and understands between the AI and any other applications. It's almost like in any other socket, in any other device, if you have a USB, you plug it in and then it connects to the computer. In the same way, the MCP is plugged into what applications you have, and then they can interact in this predictable way with the agents, AI, and so on.
11:13 Auren Hoffman? If these agents start using blockchain for everything, is that going to start slowing everything down? Or is it going to be like HTTPS where we thought it would slow everything down and then it got fast enough where it just became the default for everything?
11:26 Speaker 01 The point here is you can think about it as in HTTPS, which is more defensive. And then offensive, it's the TCP IP Communication Protocol between agents. Both of them, in order to do them at scale, you need a blockchain that can process hundreds of thousands of transactions per second, that can scale. And in some sense, everything we've built is built on this technology that we call sharding, that is built for internet scale. If you have no adoption, then of course you don't need scalability. If you have adoption on this scale, at that point, it's crucial to have the type of architecture that enables you to process hundreds of thousands of transactions per second and eventually
12:10 Speaker 01 millions of transactions per second in this type of interactions.
12:15 Auren Hoffman? World of DAS isn't just a podcast. World of DAS membership is a private invite-only community for founders and executives of important data and AI businesses. It's a place for off-the-record, high-trust conversations focused on the biggest questions data leaders are facing. The community is about peer learning, curated intros, and access to private events. Everyone in the group is building or leading something meaningful in data. If that sounds like you, apply to join at worldofdas.com. That's world of D A A S dot com.
12:57 Auren Hoffman? We're moving to world two where every person, every application, every company is having dozens, potentially hundreds, maybe even thousands of agents doing things on their behalf. And those agents are starting to interact with other agents. We have a company that is helping doctors call the insurance companies to verify if somebody is insured for a specific thing. And at first, 100% of the people that they were talking to were humans at the insurance company. But now already 10% of the time, the AI is talking to another AI. They're still talking in English, but you can imagine at some point they'll talk in a more efficient language, no reason just to talk in English to one another.
13:42 Auren Hoffman? How does that evolve over time?
13:45 Speaker 01 I wanted to say before, when you asked me all these other questions, that I think for simple users, this may sound very technical, abstract, and it's almost sci fi. But now that you come and say, hey, these are companies that are doing all this today, not in the future, and discussing between each other, my AI is discussing with your AI on the phone to do something, that tells you what crazy world we live in. So I do agree with you that this will not stop and this will accelerate. But I think a guarding against AI switching to a much more efficient language that we don't understand anymore, that becomes a black box for us, that would be an important checkpoint.
14:31 Speaker 01 Because if it becomes super effective, I've heard this metaphor where if you take the speed at which AI can process things and compare it to humans, humans will be like trees, almost in slow motion, doing something that seems so ridiculous that you're wondering, why are we even having this conversation? A question that we should pose is, how can we stay grounded and ahead, do not be left behind? And then how we keep the AIs grounded? And that's the entire point with the accountability layer, that some things should be grounded such that the AI cannot do whatever it wants to the point where it becomes malicious.
15:12 Auren Hoffman? As this transition plays out, who are the winners and losers?
15:17 Speaker 01 You can see it in a few ways. Simple users or builders, I would just encourage to play with this technology because it's extraordinarily useful and valuable today. Whatever may happen in the future, today they can unlock value out of it in whatever they're doing, whether it's passion, hobbies and so on, or professional stuff. Second, companies even more. There's an imperative in looking at this because if you're not using it, you risk leaving a lot of opportunity on the table and then being left behind. And I do think that there's also a kind of risk and challenge here
16:00 Speaker 01 that if these three, four biggest AI companies in the world race and do not have any grounding, any supervision, they may become dictators. So the question will be then, how do we ensure that it's a kind of benevolent dictator? And you don't end up with China owning this technology and the entire world going to communism and so on.
16:26 Auren Hoffman? OpenAI has recently said that they're going to lose about 20 to 30 billion per year. That's going to be their deficit every year. Back in the day, not that long ago, it was like a big deal to lose 20 to 30 million a year as a startup. And then it became like a huge thing, maybe for Uber to lose $300 million a year. And then maybe a billion dollars a year. So when we're talking about 20 to 30 billion dollars every year, that's more than market cap of most companies. And even the biggest tech companies are in a hard time competing with that amount of capital. How does that play out?
16:59 Speaker 01 I think the most important question is, what would warrant? What would justify this type of insane investment of capital? Such that you bet everything you have and then some, and you take more capital and you arm yourself. Because you almost perceive it, if this is the final war that you have to fight, then of course, every investment is justified. In practice, this is what these companies are doing. They're recognizing that there is so vast a reward and potential outcome that it's all or nothing in some sense.
17:36 Auren Hoffman? It's potentially also existential for the other tech companies that are out there.
17:41 Speaker 01 Hundred percent, and they recognize it. This is the only plausible reason why someone would tell you that they'll lose 50 billion a year for the next few years, because the outcome that justifies this is, hey, we're going to unlock a kind of oracle or genie or demigod unlike anything that we've had up until this point.
18:03 Auren Hoffman? To have this audit or digital signature, fingerprints, or blockchain interacting with AI, there's this bottoms up approach, there could be like companies adopt certain things, and maybe because they get customers or certain standards like they do with the USB and stuff. But at some point, there's a regulatory state that gets involved, there's governments that get involved. How do you think those interactions start playing out?
18:27 Speaker 01 I believe that this is beyond a certain point, just a geopolitical national security issue for every country. When something is so important that it's comparable to an atomic bomb in terms of potential, both beneficial in some sense and negative, then me developing an atomic bomb would threaten any other country that doesn't have an atomic bomb. I believe that bringing it down to grounding, every country will look at how they defend against this type of AI, and they themselves have a weapon, because if you're left without one, then you basically do not exist, you're not even at the table having a conversation.
19:11 Speaker 01 So I think there's, of course, the user perspective, how do you defend, how do you reason navigate that? There's the business perspective, but the biggest perspective is the national security one.
19:23 Auren Hoffman? Germany doesn't have an atom bomb, Japan doesn't have an atom bomb, these are very large countries, Italy, etc. Somehow they're under someone's umbrella, or someone's tutelage, or some sort of thing for them to exist today.
19:37 Speaker 01 There's a kind of game theory that plays out here where you say in the atomic bomb age, it was mutually assured destruction. You needed several big powers to arm themselves and then to say, hey, I'm allied with this guy. I'm with the US, I'm with Russia, I'm with China. And those being, let's say, three big parties, whoever met with the other one, they know that there's going to be some repercussions. Here, the thing is, in some sense, playing in a similar way. So definitely, US and China are in an arms race, publicly declared where everything goes, and they're trying to both attack and deceive themselves.
20:19 Speaker 01 You see new models coming up from China deceiving people that they've been made with less money, without chips from the US. The more practical point here is, can you wield this technology today as a positive, almost weapon? Can you use it in whatever you're doing as a user or as a business? And then what's the point at which you need to be able to defend yourself? Defend yourself as a business, defend yourself as a user, or defend your country as a kind of sovereign state? Because it's not only benevolent actors that get this technology, there's also terrorists,
21:00 Speaker 01 there's also thieves, there's also very malicious people that will try to harm you with 10x precision, less, 10x less cost. That's the huge positive potential, but then also huge negative potential that we should defend against.
21:17 Auren Hoffman? In the nuclear age, the governments were the ones who built it, created it. The core expertise, if you think of the US government, you have the Manhattan Project, you have national labs. The government is the one with all of the expertise. In AI, it's really the opposite. It's developed outside of government, sometimes even opposition, sometimes to government. Governments really is a late adopter, essentially, when it comes to AI, when it comes to blockchain. How do they get involved in the right way to actually harness it? They could mess it up, they could impair it. There's a lot of different types of things that could be happening there.
21:54 Speaker 01 There's almost two types of games. There's this kind of clear arms race that is happening right now. Where if you're one of the top countries in the world and you're not competing, you're already losing and will be losing more and more every day. Recognizing that and taking a stance around that. When it comes to the biggest two power players here are US and China. One of them wants freedom, open markets and so on. The other one wants communism and it's a much more challenging scenario to just imagine what they want. The point is beyond being one of the top three. What options do you have to play?
22:36 Speaker 01 What are your advantages for playing this game? And what are your costs or penalties for not playing this game? And I think the advantages and disadvantages are very much tied to economic opportunity. So if you're not using AI, of course, everything takes more. Whatever you're doing in the knowledge space takes 10x more. Whatever you're building in terms of coding takes 10x more. Whatever you're doing in terms of automation that now gradually eats almost everything. It's just a very clear, immediate opportunity cost. But there's also some intangibles that are not clear to people, which is the biggest outcome that these companies are targeting, for instance, is
23:20 Speaker 01 a kind of new understanding of the universe. Imagine that you unlock a new paradigm in physics, a new fundamental physics model. What would that enable? Would that enable faster than life travel? Would that enable some type of teleportation? Would the normal standard rules not apply anymore? Chemics, medicine, all that stuff that's much more fundamental than how I earn money today? That's the big outcome that these people are targeting where they're essentially saying it's not only a kind of war like scenario where this is a weapon, but this is also the most important breakthrough or discovery
24:01 Speaker 01 that if we solve would grant us a new industrial revolution that we could apply across all the fields that we have. Putting it short, there's the very competitive offensive frame where there is an arms race and you need to be there because otherwise you don't matter and you'll not have a say and you'll not be able to wield this technology. And then there's a much more creative, value-driven frame where however you look at things, you would never in your right mind actually wish to go back in history before the Industrial Revolution.
24:41 Speaker 01 However nostalgic it may sound and interesting it may sound, actually having medicine is good. We know that is good by now. Having the basics, food and water and so on, having rights and liberties, that's sort of the type of thing that we're fighting for and then we're expanding as countries, companies and users or citizens.
25:03 Auren Hoffman? The blockchain revolution was highly, in some ways, decentralized all over the world. You saw really interesting things and some of the most interesting developments on blockchain were just in like random corners of the world that are out there. AI seems to be a bit different. It's not just in the US, it's in the Bay Area. It's not just in the Bay Area, it's in San Francisco. It's not just in San Francisco, it's in like two or three neighborhoods of San Francisco. It's so, so centralized in the very, very, very small corner. Why is that happening? Is it just the capital is so different than blockchain?
25:40 Speaker 01 There's a outcome or reward-driven arms race that's being understood by very few people at the scale that it really is. And there's very few companies and individuals that have the resources, capacity, talent to actually seize that opportunity and make it a reality. Because if you look at it, to be very honest, there have been many AI summers and winters throughout the last 50 years. But then when OpenAI actually came with GPT-3, then GPT-3.5 and so on, that was a breakthrough that
26:21 Speaker 01 Took even the most important scientists by surprise that first they were denying it. They were actually minimizing it and saying, hey, yeah, this is not so important, blah, blah, blah. We've been doing this for years. There's an element of insane talent density, capital requirements, because even if you have a few billions of dollars, you cannot compete in this anymore. It's not only a human capital, the most brilliant people in the world working on this, but it's the type of capital required to be deployed constantly to stay ahead, ground that. And then you need an entire economy with legal and market structure to translate that into something that works.
27:06 Speaker 01 You almost have a kind of missing chicken and egg problem in any other part of the world where if the US would not exist, many other technologies would never reach the market because there's not enough scale to actually ground them and sell them properly and make money out of them.
27:25 Auren Hoffman? You're obviously a very ambitious person who cares about AI, but you started Multiverse X in Romania, which is not just far from US, also even further from these neighbors in San Francisco. What are the advantages and disadvantages of building a little bit more on the periphery?
27:39 Speaker 01 You can start anywhere in the world. That's the beauty of it. If you are and have an insight, you don't need to be in Silicon Valley to start doing some stuff. The more serious you are, the more ambitious you are, the larger the scale of what you're trying to build, the harder it will be anywhere else. And this is why I am in the US right now and everything we're doing is positioned for the US. We've sort of reached in Europe, a kind of upper invisible ceiling toward what that ecosystem can offer. And I do believe that beyond a certain threshold, suppose you're multi-billion dollar already and so on. It's almost like the other parts of the world cannot satisfy the needs of growth that you want.
28:24 Speaker 01 And this is why you're doing it in the US.
28:27 Auren Hoffman? Let's wind the clock back to two thousand and seven, which isn't that long ago. Two thousand and seven, the GDP of the Europe zone is higher than the US. Europe's as fast, if not faster, growing than the US. Now, we're in two thousand and twenty five, eighteen years later, and there's just like a massive delta between the US and the Eurozone. What happened? If you ask my prediction back in the US, I would have said, well, Europe's already bigger than the US and Europe's maybe disjointed. I wouldn't have thought in two thousand and twenty five it would have been so much smaller.
29:03 Speaker 01 This is a very important and hard question that we should pose every once in a while. It's very hard to argue why that happened as they did. The grounded reasons are at the root of this is, I feel in some sense, the European zone has lost the hunger. In some sense, you don't feel that type of ambitious, crazy drive where people believe in something, go until the end of the world to make that happen. You don't feel there's a national ambition in most of the countries. You don't feel there's a national project and pride to make something happen. That then cascades into all these ways downstream.
29:48 Speaker 01 The alternative to that, if you look at Europe when it was in its heyday, hundreds of years ago, the most brilliant people in the world were all in Europe. The most brilliant people in physics, math, sciences, and so on. In fact, the US, in order to do the Manhattan Project.
30:05 Auren Hoffman? They're all Europeans.
30:07 Speaker 01 Hundred percent. They were living there and then they were migrated to the US. But all I'm saying is, you need a type of insight and intentionality that's extraordinary to actually understand that those are the most brilliant people in the world, to fish for each of them, to get them a package to come here, and then to direct them to an ambitious project like the Manhattan Project, which is national security. There's several steps and it's almost underrated how much effort goes before you get to execute the Manhattan Project, to have that type. Of resources and I feel this is what's missing in Europe. People, even when they're ambitious or successful.
30:49 Speaker 01 It's almost like there's a bit of complacency or maybe hate toward that. Where people do not understand that you do not get to have the benefits you have without someone fighting for them. Without someone working their ass off at some point, without the revolutions happening at some point. And I feel that's for every generation, it's a puzzle. How do you convey that to your children? Let's say you have a good life now and you've worked your entire life. How do you convey that struggle and message and challenge to your children? That's difficult.
31:24 Auren Hoffman? It's interesting. I would say there are more ambitious French people in the tiny city of San Francisco than there are in the large country of Fritz. Somehow it's this magnet that the super smart Eco-Polytechnique graduate just shows up in San Francisco. And then somehow ends up staying. I don't know why this happens.
31:48 Speaker 01 Paul Graham had this brilliant essay where he framed it like this. Every city has a whisper. And if you listen carefully, in every city, you may hear a whisper stronger or lighter. And then there are some cities in the world, very few, where you hear the whisper wherever you may be in the world. And that you're ambitious. You'll have very few cities in the world. And I think there's this example where the whisper of L A is be famous. Wherever you are, you come in the U S specifically for that city. Because that's the place to be. If you're in New York, New York has the whisper make money, money above everything.
32:32 Auren Hoffman? Make money through moving money.
32:34 Speaker 01 Exactly. And then San Francisco and the Bay Area and so on has this whisper, build a startup. That's the way to have an impact on the world. What's playing here is not only that the whispers locally are barely heard, but you have a whisper that's conveyed very strongly everywhere in the world. And then people, as they wake up, they just want to be with other people that are ambitious, that are pursuing these goals and that are aligned. And I feel that type of alignment will find you wherever you are.
33:09 Auren Hoffman? It's hard to get that critical mass. There's a lot of super smart tech people in Tallinn, but basically 100% of them are Estonian. There's a lot of super smart people in Stockholm, but basically 100% of them are Swedish. There's way more super smart people, maybe than almost any other place in Tel Aviv, but they're all Israeli. If you're a Spaniard, you usually don't move to Stockholm. You don't move to Tel Aviv. You don't move to Tallinn, but you can still move to San Francisco. There's something about it where the US is still this weird place where you could just show up. And I don't know why, but these other places are just not hospitable to bring in random folks.
33:46 Speaker 01 I think it's not necessarily about hospitality because in some sense, all other places you can get easier in. But it's not enough. In some sense, what you're saying is exactly this. The whisper of that place is so strong and comparatively stronger than anything else that it still is the top place to be if you're in that category. You're this young person, you wake up and then you say, I want to do AI. There's nothing like the Bay Area and so on for AI. And I think that's the thing that the other countries cannot compete with. And the thing that every other country ships in that region, if the people are brilliant enough.
34:32 Auren Hoffman? When I meet a super smart, ambitious 25 year old who's interested in tech.
34:38 Auren Hoffman? They are either staying where they are, which might be Tallinn, Stockholm, Tel Aviv, wherever, London, Hyderabad, wherever. They're either staying where they are to be close to family for whatever, that's fine, or they're moving to San Francisco. There's no other place they're moving to. They're not moving to Berlin. They're not moving to Lisbon. They're not moving to Austin or Miami. There's no other scenario that I have seen. I don't know if you disagree.
35:03 Speaker 01 No, no, I very much agree. And this is just a compounded premium that you instantly unlock there. It's almost like in your place, what you're building is 1X. In this other place called SF and so on, it's 100X from the get go.
35:19 Auren Hoffman? Maybe even 1 billion X.
35:21 Speaker 01 It can become 1 million X, but your 100 to 1,000 X valued differently just because you're there. And so people understand that.
35:30 Auren Hoffman? That's crazy. What do you think venture capital firms are getting wrong about blockchain?
35:36 Speaker 01 They got some stuff wrong in the past, more than right now. Right now, there's a kind of silent revolution where the entire US economy is doing something that was considered, let's say, illegal up until this point. Because the silent bottleneck of this entire space was what the US called Operation Chalk Point. Imagine that a state, a government actually creates a program that it calls Operation Chalk Point. That was the previous administration in the US. I figure the most important advantage today is, on the other hand, the exact opposite, that you have a regulatory structure, that you have a market structure, and the US is
36:18 Speaker 01 all in on making the US the capital for crypto and AI. That's the thing that has now paved the way for Circle to go public and have a 200X revenue to IPO multiple, and then that translating a bit into all these other stablecoin companies that are now the most profitable in the world per employee, printing money better than Tesla, SpaceX. As long as you have this great prospect. Of making money, VCs are in. And that's now the case, I think, more and more. We just need the macro green light to kick in. And then with that macro green light, this is definitely not a investment device.
37:02 Speaker 01 We could have one of the golden ages of the public markets, blockchain, AI and so on, in a way that's even significantly better and faster than everything we've seen up until this point.
37:16 Auren Hoffman? I'm super into blockchain. I've been into it for 12 plus years now, but I don't use it for anything. I own some coins, but I don't use it for anything, literally nothing today in my life. When am I going to start using a tech forward person who can code and do other types of things? When will actually help me beyond just something in my portfolio?
37:39 Speaker 01 I do agree with that and I appreciate the sentiment. I think two use cases will very likely prompt you to use it even without knowing in some sense. And then to just seeing the benefits very, very concretely. One is stable coins being embedded in any type of payment that you do that settles instant anywhere in the world.
38:00 Auren Hoffman? Because it's so cheap to settle, it will become more of a way of settling transactions.
38:07 Speaker 01 Yes, money is just a packet of data that move at the speed of light. Why hasn't it?
38:13 Auren Hoffman? Why do I have to do ACH?
38:15 Speaker 01 Yeah, it was simply legal stuff. We did not have the foresight to see how we can integrate this and so on. So once you have this stable coins, then that would be a very simple use case where you don't have to think about it.
38:30 Auren Hoffman? Stable coins are already getting pretty big, but maybe I have at some point in my life. I don't know if I've ever paid for anything in stable coins or anyone ever paid me or any of our companies. Every once in a while, we might take it from somebody or something. It's not a common thing today. Is it going to move quickly, you think? Or do you think it's going to just be a very slow moving thing?
38:50 Speaker 01 The point is, we need to stress this. This was semi-illegal.
38:55 Auren Hoffman? Good point. We're really just a few months into this is being legal in the US.
38:59 Speaker 01 And then, once it's legal, you have all the biggest companies, Stripe, Robinhood, and so on, all of them racing to implement this.
39:09 Auren Hoffman? Even like NASDAQ's now all in on it. So it's a good point.
39:13 Speaker 01 Yeah, the six to 12 months immediately, you'll see it integrated almost invisibly in many things, and they just work faster. But I also think that this conversation around agents will rely so much more on blockchains, again, in a way where you don't have to think about it. You'll not know that it's blockchains, but you can verify everything that the agent has done. It comes back to you. You granted money to do some stuff. And all of that will almost be done on a substrate that is blockchain focused.
39:46 Auren Hoffman? Ben, you made two personal questions before we leave. First of all, what is a conspiracy theory that you believe?
39:51 Speaker 01 I believe the idea that you can actually conspire to change the world, and you can actually do it. That's so outrageous, that in fact, conspiracies, I believe, started from this idea that a few people had this secret through which they moved some mountains. And I think that's true, and it can be done not only by the sharpest in Silicon Valley, but even bright young people anywhere in the world.
40:21 Auren Hoffman? It's so interesting that when people say, I don't believe in conspiracy theories, it's almost saying, I don't believe things can change. It's almost like a pessimistic view of change in the world.
40:32 Speaker 01 I do think people conspired, some in a good way, some in a bad way.
40:37 Auren Hoffman? In the crypto world, you see conspiracies all the time. There's always pump and ducks. To me, there's classic conspiracies going on in the crypto world, more than almost anywhere. It's like these private Telegram channels coordinating things. It's tough.
40:51 Speaker 01 Yes, but one thing that looks like a conspiracy, but it's actually real, is the operation Chop Point.
40:57 Auren Hoffman? I was crazy. It's still hard for me to even believe that that happened.
41:01 Speaker 01 And so unconstitutional and against everything that the US stands for. This is not China.
41:08 Auren Hoffman? How do you think that even happened? I don't think the President of the United States at the time even knew what was going on. I don't think he directed it to happen. Like, he probably couldn't care either way. I don't know that anyone like high up was coordinating this. What's your conspiracy about how it happened?
41:24 Speaker 01 I think there's a very simple explanation for this, and that's financial incentive.
41:28 Auren Hoffman? JP Morgan, those types of places, they have a lot of incentive to stop this.
41:33 Speaker 01 The entire space has an incentive to keep it as it is and not muddy everything.
41:38 Auren Hoffman? Yeah, that's true. They had enough power to push some things around, scare some people, et cetera.
41:45 Speaker 01 Yeah, and the way, just to be very honest, to what the previous president knew or did not know, I think he did not know many things.
41:53 Auren Hoffman? I agree. He wasn't in control, so that was for sure.
41:56 Speaker 01 Not even in control of himself anymore. You would look at him and say, is this the strongest country in the world? Is this the person that governs the entire known world? What's up with this?
42:07 Auren Hoffman? It makes me mad that that happened. To me, that is just doing something that's clearly not in the best interest of the United States. Very good for United States adversaries, and it doesn't actually make sense. Probably so many things happen that way because of fiefdoms and other stuff, but it does make me very upset that it did happen.
42:26 Speaker 01 It almost seems too simple right now when I say that the strongest president in the world with the administration and so on has course corrected not to something neutral, but something where the US wants to lead. And I think that requires both big respect, big brains and so on, but also balls. Very rare that you see politicians actually taking this risk.
42:51 Auren Hoffman? The current president did change his mind on crypto. So that's also something you have to respect. He reassessed it and changed his mind.
42:59 Speaker 01 Hundred percent. So I give a lot of credit to the US for where it is and where it continues to push itself to be, even after mistakes, because every country makes mistakes and so on. But the idea of having the presence of spirit to recognize an opportunity, to take risks, and then define regulation and market structure and all of that because you see how it will contribute to your people, to your institutions and to businesses all around your country. That's a spirit that every country should try to emulate a bit more of.
43:34 Auren Hoffman? Our last question we ask all of our guests. What conventional wisdom or advice do you think is generally bad advice?
43:40 Speaker 01 I do not believe in this idea of the work-life balance. If you're an ambitious person, if you're not obsessed with something, you're not telling me anything.
43:50 Auren Hoffman? Isn't that the conventional advice? That you don't have work-life balance if you're ambitious? That is the conventional advice.
43:56 Speaker 01 I think most of the people still have not recognized what spectrum that can go to. And I feel Musk, in some sense, again, has demonstrated that you can sit and be the wealthiest man in the world and then go in your factory and slip on the floor. And until that problem is not solved, you'll not get there. That's one version, just that the young people today maybe do not appreciate it today, just because you have a good life, especially you do a famous university, Ivy League.
44:28 Auren Hoffman? Even Elon Musk, he's usually, you gotta be all in on something. He's got 10 jobs and then he also spends many hours a week becoming a great video game player in the world. And then he spends time thinking about government and other stuff and tweeting. In some ways, you'd be like, you shouldn't be tweeting. You shouldn't have other jobs. You shouldn't be playing video games. You should just be like working. He's not working on any one company more than 20 hours a week, on average. Maybe it is 100 hours one week and zero the next or something.
44:58 Speaker 01 I do agree that it's very unequal in that respect. I think it's much more on the extent of the energy outcome and dosage of what he puts in that then makes this again nearly impossible things seem possible, like even Neuralink. Very few people maybe realize how fundamental that may be for not only science, engineering and so on, but health, well-being and solving some terminal illnesses.
45:28 Auren Hoffman? To me, the Musk thing shows that you don't really need to spend that much time in your company. If you're CEO, weirdly, on any given company, he's not spending that much time. But you do need these weird bursts of concentrated time. We have to do like a hundred hours in one week, sleeping on the floor. And then you have to pick your spots and then you have to super micromanage in this concentrated time. And then outside you're like, things are just going. And then you jump in again and you have to like super micromanage in a weird way. He's the best at that thing. And most people, it's kind of like steady. Okay, I'm gonna do sixty hours a week, every week, very, very, very steady. I'm gonna have the same meetings every Monday, the same meetings every Tuesday.
46:11 Auren Hoffman? I'm gonna go fly to China, look at the supply chain on Wednesday, whatever it is. It's a different way of thinking about running a company.
46:19 Speaker 01 I do think that Musk points to almost a different model of living life, doing business, doing productive stuff, because without him, you almost have things figured out. There's conventional, then this is what it dictates. You definitely don't do several companies, you'll blow up. It's painful to do one company. Why you do three companies or four companies, take that type of risk? But I feel at minimum, he was the type of guy that made me think, hey, conspiracies are not true as people think they're true. And then they are true in a very contrarian way than you think. Namely, what people want to conspire to do doesn't work out most of the times,
47:03 Speaker 01 even if they're ill-intended and so on. And then you can conspire with two, three people. People to change something extraordinarily that seems impossible like reusable rockets. And the entire field contest you and says you're ludicrous and it will never happen. And then 20 years down the road, 20 years, not two years, not five years. Pain, if you survive, if you're still alive, if you can still burn that pain and look so stupid so many years. I think you might have a chance to prove the world that he was wrong.
47:38 Auren Hoffman? Well, thank you, Benjamin Mincu, for joining us on World of DAS. I follow you at Benjamin Mincu on X. I definitely encourage our listeners to engage with you there. This has been super interesting, super enlightening and a ton of fun.
47:53 Speaker 01 Thank you very much. Great pleasure of being here.
47:58 Auren Hoffman? Thanks for listening. If you haven't already done so, please subscribe to World of DAS wherever you consume your podcasts, YouTube, Apple, Spotify and more. And please help us get discovered by leaving a review and check out worldofdas.com. That's worldofdas.com. And of course, connect with me on Twitter, at Auren, that's A U R E N. Would absolutely love to hear from you.
Previous The LM Brief: How to Actually Look for a Job (Without Wasting Everyone’s Time) Next The LM Brief: How Leaders Can Close the AI Intelligence Gap Page updated November 1, 2026 · transcript and summary are machine-generated