Podcast
Marc Andreessen Introspects on the Death of the Browser, Pi + OpenClaw, and Why "This Time Is Different"
Latent Space: The AI Engineer Podcast
- AI Became Real After Decades Of False Starts
- Marc Andreessen argues AI is an 80 year overnight success, not a fresh hype cycle, because today’s systems finally validate decades of neural net research.
- He says LLMs, reasoning, coding, agents, and recursive self improvement now all work, making this cycle fundamentally different from prior AI summers and winters. Transcript: Marc Andreessen So there’s something about, say the following, there’s something about AI that has led to this repeated pattern. And you guys know this. It’s summer, winter, summer, winter. And it goes back 80 years. So the original neural network paper was 1943, which is amazing that it was far back that long. And then there was – I have you guys ever talked about this on your show, but there was a big – there was an AGI conference at Dartmouth University in 1955. And they got an NSF grant for all the AI experts at the time to spend the summer together. And they figured if they had 10 weeks together, they could get AGI at the other end. And they got their – by the way, they got the grant, they got the 10 weeks, and then 1955, no AGI. And like I said, I lived through the 80s version of this where there was a big boom and a crash. And so there is this thing, and there is something about AI that causes the people in the field, I would say, to become both excessively utopian and excessively apocalyptic. And it’s probably on both sides of like the boom-bust cycle, you kind of see that play out. Having said that, I think what’s actually happened is like just, you know, and we now know in retrospect, like an enormous amount of technical progress that built up over time. And like, for example, we now know the neural network is the correct architecture. And I will tell you, like there was a 60-year run where that was like, you know, or even 70 years where that was controversial. And we now know that that’s the case. And so we now, you know, everything we’re building on today sort of derives from the original idea in 1943. And so in retrospect, we now know that these guys were right. They would get the timing wrong, and they thought capabilities would arrive faster, or it could be turned into businesses sooner or whatever. But they were fundamentally, the scientists who worked on this over the course of decades were fundamentally correct about what they were doing. And the payoff from all their work is happening now. And so the way I think about what’s happening is basically, I think about basically the period we’re in right now is it’s, I call it 80 year overnight success, right? Which is like, it’s an overnight success. Cause it’s like, bam, you know, chat GPT hits and then, and then O1 hits and then, you know, open claw hits. And like, you know, these are open, these are, these are like overnight, like radical overnight transformative successes, but they’re drawing on an 80 year sort of wellspring backlog Of ideas and thinking. It’s not just that it’s all brand new. It’s that it’s an unlock of all of these decades of very serious hardcore research and thinking. Look, there were AI researchers who spent their entire lives. They got their PhD. They’ve researched for 40 years. They retired. In a lot of cases, they passed away, and they never actually saw it work. So sad. It is sad. I think Jeff Hinton was like the last guy. Yeah. Yeah. Well, there were the guys, Alan Newell. I mean, there’s tons of John McCarthy. John McCarthy was like one of the inventors in the field. He’s one of the guys who organized the Dartmouth conference and he taught at Stanford for 40 years and passed away, I don’t know, whatever, 10 years ago or something. Never actually got to see it happen. But it is amazing in retrospect. These guys were incredibly smart and they worked really hard and they were correct. So anyway, so then it’s like, okay, you know, as they say, history doesn’t repeat, but it rhymes. It’s like, okay, does that mean that there’s going to be another, like, you know, basically boom, bust cycle? And I will tell you, like, look, like in a sense, like, yes, everything goes through cycles and, you know, people get overly enthusiastic and overly depressed. And there’s a timelessness to that. Having said that, there’s just no question. The four most dangerous words in investing are this time is different. Do you know the 12 most dangerous words of investing? No. The four most dangerous words in investing are this time is different. The 12 most dangerous words. And so I’ll tell you what’s different. Now it’s working. There’s just no, I mean, look, there’s just no question. And by the way, I’ll just give you guys my take. Like LLM is like from, from basically the chat GPT moment through to spring of 25. I think you could still, I think well-intentioned, well-informed skeptics could still say, oh, this is just pattern completion. And oh, these things don’t really understand what they’re doing. And, you know, the hallucination rates are way too high. And, you know, this is going to be great for creative writing and creating, you know, Shakespearean sonnets and, you know, as rap lyrics or whatever, like it’s going to be great at all That stuff, but we’re not going to be able to harness this to make this relevant in, you know, coding or in medicine or in law or in, you know, kind of fields that, you know, kind of really, Really matter. And I think basically it was the reasoning breakthrough. It was O1 and then R1 that basically answered that question and basically said, oh no, we’re going to be able to actually turn this into something that’s going to work in the real world. And then obviously the coding breakthrough over the, basically the coding breakthrough that kind of catalyzed over the holiday break was kind of the third step in that. But it was like, all right, if, you know, if Linus Torvalds is saying that AI coding is not better than he is, like, that’s never happened before. That’s the benchmark. Yeah. That’s never happened before. And so now we know that it’s going to sweep through coding. And then we know that if it’s going to work in coding, it’s going to work in everything else. Right. It’s just that because that’s like the hardest. In many ways, that’s the hardest example. And now everything else is going to be a derivative of that. And then on top of that, we just got the agent breakthrough with OpenClaw, which is fantastic, which is amazing and incredibly powerful. And then we just got the auto research, the self-improvement. We’re now into the self-improvement breakthrough. And so the way I think about it is we’ve had four fundamental breakthroughs in functionality, LLMs, Reasoning Agents, and then now RSI. And they’re all actually working. And so I’m just, as you can tell, I’m jumping out of my shoes. Like this is like, this is it. Like this is the culmination of 80 years worth of worth of work. And this is the time it’s becoming real. Yeah. I’m completely convinced. (Time 0:06:52)
- AI’s 80 Year Overnight Success
- Marc Andreessen frames today’s AI boom as an “80 year overnight success”: dramatic recent breakthroughs are the payoff of decades of foundational research.
- Neural networks were proposed in 1943 and remained controversial for decades, but the architecture proved correct and enabled today’s progress.
- Periodic hype and winters happened repeatedly, yet persistent technical work accumulated until catalytic moments (e.g., GPT-3) made capabilities visibly transformative.
- Many researchers spent entire careers without seeing success; current products are unlocking a long backlog of ideas rather than purely new inventions. Transcript: Marc Andreessen And it’s probably on both sides of like the boom-bust cycle, you kind of see that play out. Having said that, I think what’s actually happened is like just, you know, and we now know in retrospect, like an enormous amount of technical progress that built up over time. And like, for example, we now know the neural network is the correct architecture. And I will tell you, like there was a 60-year run where that was like, you know, or even 70 years where that was controversial. And we now know that that’s the case. And so we now, you know, everything we’re building on today sort of derives from the original idea in 1943. And so in retrospect, we now know that these guys were right. They would get the timing wrong, and they thought capabilities would arrive faster, or it could be turned into businesses sooner or whatever. But they were fundamentally, the scientists who worked on this over the course of decades were fundamentally correct about what they were doing. And the payoff from all their work is happening now. And so the way I think about what’s happening is basically, I think about basically the period we’re in right now is it’s, I call it 80 year overnight success, right? Which is like, it’s an overnight success. Cause it’s like, bam, you know, chat GPT hits and then, and then O1 hits and then, you know, open claw hits. And like, you know, these are open, these are, these are like overnight, like radical overnight transformative successes, but they’re drawing on an 80 year sort of wellspring backlog Of ideas and thinking. It’s not just that it’s all brand new. It’s that it’s an unlock of all of these decades of very serious hardcore research and thinking. Look, there were AI researchers who spent their entire lives. They got their PhD. They’ve researched for 40 years. They retired. In a lot of cases, they passed away, and they never actually saw it work. (Time 0:07:42)
- Scaling Laws Keep Advancing But Reality Stays Messy
- Marc Andreessen treats AI scaling laws like Moore’s Law, as predictions that become self fulfilling when talent and capital organize around them.
- He thinks startups still matter because the messy world of institutions, companies, and regulation slows model progress from instantly swallowing every layer above it. Transcript: Alessio Fanelli Think the anxiety that people feel is like during the transistor era, you had Morse law and it’s like, all right, we understand why these things are getting better. We understand the physics of it. With AI, it’s, it’s so jagged in like the jumps where like you said, it’s like in three months, you have like this huge jump, and people are like, well, this can keep happening. Right. But then it keeps happening. It will keep happening. And so like, how do you think about also timelines of like what’s we’re building? I think we always have this question with guests, which is like, you know, should you spend time building harness for a model versus like the next model, just going to do it one shot in The latent space. And how does that inform like how you think about the shape of the technology you know you talk about how it’s a new computing platform if you have a computing platform that like every Six months it like drastically changes and what it looks like it’s hard to build companies on top of it yeah so it’s a couple things so one is like look the more’s law was what we now call Marc Andreessen A scaling law like more’s law was a scaling law and for your younger viewers more more’s law was every chipips either get twice as powerful or twice as cheap every 18 months. And it’s gotten more complicated in the last few years, but that was the 50-year trajectory of the computer industry. And then, by the way, that’s what took the mainframe computer from a $25 million current dollar thing into the phone in your pocket being a million times more powerful than that for 500 Bucks. And so, that was a scaling law. And then, key to any scaling law, including Moore’s law and the AI scaling laws is they’re not really laws, they’re predictions. But when they work, they become self-fulfilling predictions because they set a benchmark and then the entire industry, all the smart people in the industry work to make sure that that Actually happens. And so they motivate the breakthroughs that are required to keep that going. And in chips, that was a 50-year run, right? And it was amazing. And it’s still happening in some areas of chips. I think the same thing is happening with the core scaling laws in AI. They’re not really laws, but they are basically, they’re predictions and then they’re motivating catalysts for the research work that is required to be. And by the way, also the investment dollars are required to basically keep the curves going. And look, it’s going to be complicated and it’s going to be variable. And there are going to be walls that are going to look like they’re fast approaching. And then they’re going to be – engineers are going to get to work and they’re going to figure out a way to punch through the walls. And obviously, that’s been happening a lot. And then look, there’s going to be times when it looks like the walls have – the laws have petered out. And then they’re going to pick up again and surge. And then it appears what’s happening to the eyes, there’s not multiple scaling laws. There’s multiple areas of improvement. And I think – I don’t know how many more there are yet to be discovered, but there are probably some more that we don’t know about yet. Like for example, there’s probably some scaling law around world models and robotics that we don’t fully understand – kind of acquisition of data at scale in the real world that we don’t Fully understand yet. So that one will probably kick in at some point here. There’s a bunch of really smart people working on that. And so, yeah, I think the expectation is that, you know, the scaling laws generally are going to continue. Yeah, the pace of improvement will continue to move really fast. To your question on like what to build. So I’m a complete believer the scaling laws are going to continue. I’m a complete believer the capabilities are going to keep getting amazing, you know, leaps and bounds. The part where I kind of part ways a little bit with what I would describe as the AI purists, you know, which is, which I would characterize as like the people who are in many ways, the smartest People in the field, but also the people who spend their entire life like in a lab and have, I would say, have very little experience in the outside world. The nuance I would offer is the outside world of 8 billion people and institutions and governments and companies and economic systems and social systems is really complicated um and Um and doesn’t you know it eight billion people making collective decisions on planet earth is not a simple process of like just like you see this happening now it’s like a bunch of the Ai ceos have this thing like, well, there’s just this, they just all have this kind of thing when they talk in public where they’re just like, well, there’s this obvious set of things That society needs to do. And then they’re like, society’s not doing any of those things. Right. And it’s like, how can society not, you know, whatever their theory is, how can society not see X, Y, Z? And the answer is, well, society is number one, there’s no single society. It’s like 8 billion people. And they like all have a voice and they all have a vote at the end of the day of how they react to change. And then it’s just human reality is just really complicated and messy. And so the specific answer to your question is like, as usual, it depends. It depends. Look, there’s no question people are going to like, there’s no question there are going to be companies. It’s already happening. There are companies that think that they’re building value on top of the models and then they’re just going to get blissed by the next model. There’s no question that’s happening. But I think there’s no question also that just the process of adaptation of any technology into the real messy world of humanity is just going to be messy and complicated. It’s not going to be simple and straightforward. It’s going to be messy and complicated. (Time 0:11:55)
- Why AI Scaling Laws Mirror Moore’s Law
- Marc Andreessen argues Moore’s Law is a type of scaling law that served as a predictive benchmark, driving industry effort to meet it.
- AI scaling laws work the same way: they’re not immutable laws but predictions that motivate research and investment to keep progress on the predicted curve.
- Because those predictions become self-fulfilling, they catalyze the breakthroughs (models, reasoning, agents, RSI) Andreessen sees as making AI a real platform.
- The comparison explains why rapid, jagged jumps in AI can still produce a stable platform: the scaling predictions focus collective effort and funding to continue progress.
- Andreessen notes Moore’s Law produced a 50-year trajectory that turned mainframes into pocket-level computing, implying similar systemic impact is possible for AI. Transcript: Alessio Fanelli I think we always have this question with guests, which is like, you know, should you spend time building harness for a model versus like the next model, just going to do it one shot in The latent space. And how does that inform like how you think about the shape of the technology you know you talk about how it’s a new computing platform if you have a computing platform that like every Six months it like drastically changes and what it looks like it’s hard to build companies on top of it yeah so it’s a couple things so one is like look the more’s law was what we now call Marc Andreessen A scaling law like more’s law was a scaling law and for your younger viewers more more’s law was every chipips either get twice as powerful or twice as cheap every 18 months. And it’s gotten more complicated in the last few years, but that was the 50-year trajectory of the computer industry. And then, by the way, that’s what took the mainframe computer from a $25 million current dollar thing into the phone in your pocket being a million times more powerful than that for 500 Bucks. And so, that was a scaling law. And then, key to any scaling law, including Moore’s law and the AI scaling laws is they’re not really laws, they’re predictions. But when they work, they become self-fulfilling predictions because they set a benchmark and then the entire industry, all the smart people in the industry work to make sure that that Actually happens. And so they motivate the breakthroughs that are required to keep that going. And in chips, that was a 50-year run, right? And it was amazing. And it’s still happening in some areas of chips. (Time 0:12:18)
- Scaling Laws Meet The Messy Reality Of Eight Billion People
- Marc Andreessen expects AI scaling laws to continue and drive major capability improvements.
- He warns technical progress collides with the complexity of the real world: adoption involves 8 billion people, institutions, governments, and messy social/economic systems.
- Many companies will be disrupted by the next model (they get “blissed”), but adaptation across society will be slow, uneven, and complicated.
- There won’t be a single clear path; success depends on how technologies interact with existing incentives and human behavior.
- The takeaway: build with technical momentum in mind, but design for messy real-world adoption and institutional friction. Transcript: Marc Andreessen To your question on like what to build. So I’m a complete believer the scaling laws are going to continue. I’m a complete believer the capabilities are going to keep getting amazing, you know, leaps and bounds. The part where I kind of part ways a little bit with what I would describe as the AI purists, you know, which is, which I would characterize as like the people who are in many ways, the smartest People in the field, but also the people who spend their entire life like in a lab and have, I would say, have very little experience in the outside world. The nuance I would offer is the outside world of 8 billion people and institutions and governments and companies and economic systems and social systems is really complicated um and Um and doesn’t you know it eight billion people making collective decisions on planet earth is not a simple process of like just like you see this happening now it’s like a bunch of the Ai ceos have this thing like, well, there’s just this, they just all have this kind of thing when they talk in public where they’re just like, well, there’s this obvious set of things That society needs to do. And then they’re like, society’s not doing any of those things. Right. And it’s like, how can society not, you know, whatever their theory is, how can society not see X, Y, Z? And the answer is, well, society is number one, there’s no single society. It’s like 8 billion people. And they like all have a voice and they all have a vote at the end of the day of how they react to change. And then it’s just human reality is just really complicated and messy. And so the specific answer to your question is like, as usual, it depends. It depends. Look, there’s no question people are going to like, there’s no question there are going to be companies. It’s already happening. There are companies that think that they’re building value on top of the models and then they’re just going to get blissed by the next model. There’s no question that’s happening. But I think there’s no question also that just the process of adaptation of any technology into the real messy world of humanity is just going to be messy and complicated. (Time 0:14:42)
- Technology Adoption Is Messy Not Inevitable
- Marc stresses that real-world adoption involves 8 billion people, institutions, and markets, so scaling laws alone don’t guarantee smooth rollout.
- He warns many companies building on current models will be “blissed” by the next model and lose value.
- The adaptation of AI into society will be messy, complicated, and produce entire new industries to bridge tech to people.
- Historical perspective: Marc lived through the dot-com crash and highlights how systemic factors (e.g., telecom/bandwidth) can wipe out apparent tech progress.
- The takeaway is to expect both rapid capability improvements and messy, unpredictable real-world integration. Transcript: Marc Andreessen And so the specific answer to your question is like, as usual, it depends. It depends. Look, there’s no question people are going to like, there’s no question there are going to be companies. It’s already happening. There are companies that think that they’re building value on top of the models and then they’re just going to get blissed by the next model. There’s no question that’s happening. But I think there’s no question also that just the process of adaptation of any technology into the real messy world of humanity is just going to be messy and complicated. It’s not going to be simple and straightforward. It’s going to be messy and complicated. There are going to be a lot of companies and a lot of products and, in fact, entire industries that are going to get built to basically actually help all of this technology actually reach Real people. Alessio Fanelli The amount of capital going into these companies, I mean, Dario talked about it on the Dworkash podcast and Dworkash was like, why don’t you just buy 10x more GPUs? And he’s like, because I’m going to go bankrupt if the model doesn’t exactly hit the performance level. How do you think about that? Also as a risk on, you know, you guys are investors in OpenAI and thinking machines and world apps. It seems like we’re leveraging the scaling loss at a pretty high rate. Like how comfortable, I guess, do you feel with the downside scenario? Like, and say like things peter out, you think you can kind of like restructure these build outs and, you know, capital investments? Marc Andreessen Yeah. So I should start by saying, so I lived through the dot-com crash and I can tell you stories for hours about the dot-com crash and it was horrible. No, it was awful. It was apocalyptic. By the way, a lot of the dot-com crash was actually, at the time, it was actually a telecom crash. It was a bandwidth crash. The thing that actually crashed that wiped out all the money was the telecom companies. (Time 0:15:56)
- AI Infrastructure Looks Stronger Than The Dot Com Bubble
- Marc Andreessen sees today’s AI capex boom as safer than the dot com telecom overbuild because buyers are cash rich incumbents and deployed GPUs monetize immediately.
- He argues supply shortages are sandbagging model quality, and even old NVIDIA chips can gain value because software improves faster than hardware depreciates. Transcript: Alessio Fanelli The amount of capital going into these companies, I mean, Dario talked about it on the Dworkash podcast and Dworkash was like, why don’t you just buy 10x more GPUs? And he’s like, because I’m going to go bankrupt if the model doesn’t exactly hit the performance level. How do you think about that? Also as a risk on, you know, you guys are investors in OpenAI and thinking machines and world apps. It seems like we’re leveraging the scaling loss at a pretty high rate. Like how comfortable, I guess, do you feel with the downside scenario? Like, and say like things peter out, you think you can kind of like restructure these build outs and, you know, capital investments? Marc Andreessen Yeah. So I should start by saying, so I lived through the dot-com crash and I can tell you stories for hours about the dot-com crash and it was horrible. No, it was awful. It was apocalyptic. By the way, a lot of the dot-com crash was actually, at the time, it was actually a telecom crash. It was a bandwidth crash. The thing that actually crashed that wiped out all the money was the telecom companies. Global crossing. Shawn ‘swyx’ Wang I’m from Singapore and they laid so much cable over our oceans. Marc Andreessen Actually, it was a scaling law in the dot-com era and it was literally the US Commerce Department put out a report in 1996 and they said internet traffic was doubling every quarter. And actually in 1995 and 1996, internet traffic actually did double every quarter. And so that became the scaling law. And so what all these telecom entrepreneurs did was they went out and they raised money to build fiber, anticipating that the demand for bandwidth is going to keep doubling every quarter. Doubling every quarter though is like grains of chess on the chessboard. At some point, the numbers become extremely large. Really, what happened was the internet, by the way, continuously kept growing basically since inception. It’s continuously grown. It’s never shrunk. It’s grown really fast compared to anything else in human history. But it wasn’t doubling every quarter as of 1998, 1999. And so, there was this gap in the expectation of what they thought was a scaling law versus reality. And that’s actually what caused the dot-com crash, which was they way over – companies like Global Crossing way overbuilt fiber, which is sort of the – and by the way, fiber, telecom, Equipment, you know, so all the networking gear, you know, and then, by the way, the actual physical data centers. Like, that was the beginning of the data center build and then data center overbuild. And so you had that, but it was literally, I think it was like $2 trillion got wiped out, right? It was like a big… And by the way, the other subtlety in it was the internet companies themselves never really had any debt because tech companies generally don’t run on debt, but the telecom companies Run on debt, physical infrastructure companies run on debt. And so the companies like Global Crossing not just raised a lot of equity, they also raised a lot of debt. So they’re highly levered. And so then you just do the thing. It’s just like, okay, you have a highly levered thing where you’re just over, you’re overbuilding capacity. Demand is growing, but not as fast as you hoped. And then boom, bankrupt. Right. And then it’s like they say about the hotel industry, which is it’s always the third owner of a hotel that makes money, right? It has to go bankrupt twice, right? You have to wash out all of the over-optimistic exuberance before it gets to actually a stable state, and then it makes money. So by the way, all of those data centers and all of those, all the fiber that they’re in use, it’s all in use today, but 25 years later. But it took, and actually the elapsed time was it took 15 years. It took 15 years from 2000 to 2015 to actually fill up all that capacity. The cautionary warning is the overbuild can happen. And, you know, you get into this thing where basically everybody who basically has any sort of institutional capital is like, wow, it’s just, I don’t know how to invest in these crazy Software things. But for sure, I can build data centers and for sure, I can buy GPUs and I can deploy, you know, compute grids and all these things. And so, you know, if you’re a pessimist, you can look at this and you can say, wow, this is like really set up to be able to basically replicate, you know, what we went through, what we went Through in 2000. Obviously, that would be bad. The counter argument, which is the one I agree with, which is the counter on the other side is a couple of things. One is the companies that are investing all the, the companies that are investing the money are like the bluest chip of companies. And so back in the, like Global Crossing was like an, it was like an entrepreneur. It’s like a new venture. But like the money that’s being deployed now at scale is Microsoft and, you know, and Amazon and Google, Facebook and NVIDIA. And, you know, these, and now, you know, by the way, OpenAI and Anthropic, which are now at like, you know, really serious size, you know, as companies with, you know, very serious revenue. These are very large scale companies with like lots, lots of cash, lots of debt capacity that they’ve never used. And so this is institutional in a way that that really wasn’t at the time. And then the other is, at least for now, every dollar that’s being put into anything that results in a running GPU is being turned into revenue right away. Like, so, and you guys know this, like everybody star for capacity. Everybody starved for compute capacity and then all the associated things, memory and interconnect and everything else, data center space. And so every dollar right now that’s being put in the ground is turning into revenue. And in fact, I actually think there’s an interesting thing happening, which is because everybody starved for capacity, the models that we actually have that we can use today are inferior Versions of what we would have if not for the supply constraints. It’s true. Right. Suppose a hypothetical universe in which GPUs were 10 times cheaper and 10 times more plentiful, the models would be much better because you would just allocate a lot more money to training And you’d just build better models and they would be better. Shawn ‘swyx’ Wang And so we’re actually getting the sandbag version of the technology. No, everything we use is quantized because the labs have to keep the full versions. Right. Marc Andreessen We’re not even getting the good stuff. But getting the good stuff is just, even if technical progress stops, once there’s like a much bigger build of like GPU manufacturing capacity and memory, you know, all the things that Have to happen in the course of the next five or 10 years. Once it happens, even the current technology is going to get much better. And then, as you know, like there’s just like a million ways to use this stuff. There’s just a million use cases for this. This isn’t just sending packets across a thing, whatever, and hoping that people find something to do with it. This is just like, oh, we apply intelligence in every domain of human activity. And then it works incredibly well. Here’s what I know. Here’s what I know. In the next three or four years, it’s somewhere between three or four years out, basically everything is selling out. So like the entire supply chain is sold out or selling out. And so there’s no, like, we’re just going to have like chronic supply shortage for, you know, for years to come. There’s going to be a response from the market that’s going to result in an enormous, you know, it’s happening now, an enormous flood of investment in a new fab capacity and, you know, Everything else to be able t… (Time 0:16:33)
- How The Dot‑Com Telecom Build Caused The Crash
- Marc Andreessen recounts living through the dot‑com crash and emphasizes how awful and apocalyptic it was.
- The crash was driven largely by a telecom/bandwidth scaling-law mistake: a 1996 Commerce Department report showed internet traffic doubling every quarter, and investors built fiber assuming that would continue.
- Telecoms like Global Crossing massively overbuilt fiber, data centers, and networking gear based on that expectation, creating a physical overcapacity bubble.
- The internet itself kept growing but not at the assumed doubling rate, so the mismatch between expectation and reality wiped out roughly $2 trillion in value, mainly in leveraged telecom companies.
- Andreessen highlights the contrast that internet companies had little debt while telecoms carried heavy debt, amplifying the crash’s impact. Transcript: Marc Andreessen Yeah. So I should start by saying, so I lived through the dot-com crash and I can tell you stories for hours about the dot-com crash and it was horrible. No, it was awful. It was apocalyptic. By the way, a lot of the dot-com crash was actually, at the time, it was actually a telecom crash. It was a bandwidth crash. The thing that actually crashed that wiped out all the money was the telecom companies. Global crossing. Shawn ‘swyx’ Wang I’m from Singapore and they laid so much cable over our oceans. Marc Andreessen Actually, it was a scaling law in the dot-com era and it was literally the US Commerce Department put out a report in 1996 and they said internet traffic was doubling every quarter. And actually in 1995 and 1996, internet traffic actually did double every quarter. And so that became the scaling law. And so what all these telecom entrepreneurs did was they went out and they raised money to build fiber, anticipating that the demand for bandwidth is going to keep doubling every quarter. Doubling every quarter though is like grains of chess on the chessboard. At some point, the numbers become extremely large. Really, what happened was the internet, by the way, continuously kept growing basically since inception. It’s continuously grown. It’s never shrunk. It’s grown really fast compared to anything else in human history. But it wasn’t doubling every quarter as of 1998, 1999. And so, there was this gap in the expectation of what they thought was a scaling law versus reality. And that’s actually what caused the dot-com crash, which was they way over – companies like Global Crossing way overbuilt fiber, which is sort of the – and by the way, fiber, telecom, Equipment, you know, so all the networking gear, you know, and then, by the way, the actual physical data centers. (Time 0:17:07)
- Open Source And Edge AI Win On Cost And Trust
- Marc Andreessen expects open source and edge inference to matter because centralized inference will stay supply constrained, expensive, and trust limited.
- He points to Apple Silicon progress, local privacy needs, and DeepSeek’s release of papers and code as proof that open models teach the world, not just undercut prices. Transcript: Alessio Fanelli Yeah. How important is open source AI and kind of like edge inference in a world in which you have three years of supply crunch? Like do you think in the, like, you know, if you fast forward like five years, like how do you think about inference in the data center versus at the edge? Marc Andreessen Well, so just to start, yeah. So I think open source is very important for a bunch of reasons. I think edge inference is very important for a bunch of reasons. I think just practically speaking, if we’re just going to have fundamental supply crunches for the next, I mean, you guys know, if you just project forward demand over the next three Years relative to supply, one of the dismaying predictions you can do is what’s going to happen to the cost of inference in the core over the next three years. And like, it may rise dramatically, right? Like, so what is, model companies are subsidizing heavily right now. And so what will be the average person’s per day, per month token cost three years from now to do all the things that they want to do? And I don’t know. I have friends today who are paying $1,000 a day for claw tokens to run open claw. And so, okay, $30,000 a month. And by the way, those friends have like a thousand more ideas of the things that they want their claw to do. And so you could imagine there’s like latent demand of up to, I don’t know, $5,000 or $10,000 a day of tokens for a fully deployed personal agent. And obviously, consumers can’t pay that. But it gives you a sense of the future scope of demand, right? And so even if there’s a 10x improvement in price performance, that still goes to $100 a day, which is still way beyond what people can pay. So there’s just going to be ferocious demand. By the way, the agent thing, the other interesting thing is I think the agent thing, so up until now, a lot of the constraints have been GPU constraints. I think the agent thing now also translates into CPU constraints, right? CPU and memory, yes. CPU and memory, right? And so like the entire chip ecosystem is just going to get- With the network constraints, that will be the killer. It’s all bottlenecking potentially for years. And so I think that Brad, and I think it’s actually possible, I mean, generally inference costs are going to keep coming down, but I think the, let’s put it this way, the rate of decline, I think may level out here for a bit because of these supply constraints. And then at some point, maybe the lab stops subsidizing so much and that again will be an issue. And so there’s just going to be so much more demand for inference than can be satisfied, you know, kind of with the centralized model. And then, you know, you guys know this, but like all the just the dramatic, I mean, just the dramatic innovations that have happened in the Apple Silicon to be able to do inferences. It’s quite amazing. The level of effort being put, like the open source guys are putting incredible effort into getting, you know, this recurring pattern where the big model will never run on a PC. And then six months later, it runs on a PC, right? It’s like amazing. And there’s very smart people working on that. So there’s all that. And then look, there’s also, you know, there’s also like other, there’s other motivators, there’s other motivators, which is just like, okay, how much trust are the big centralized Model providers? You know, how much trust are they building in the market versus, you know, how much are, you know, at least for in certain cases with some people for certain use cases, people being like, Well, I’m not willing to just turn everything over. So there’s all the trust issues. By the way, there’s also just straight up price optimization. There’s many uses of AI where you don’t need Einstein in the cloud. You just need a smart local model. There’s also performance issues where you’re going to want your doorknob to have an AI model in it to be able to do access control. Obviously, like everything with a chip is going to have an AI model in it. And a lot of those are going to be local. And so, yeah, no, like, I think, I think you’re going to have, and then you’re going to, by the way, also wearable devices, you know, you don’t want to do a complete round trip. You want, you know, you, whatever your smart devices are, you want it to be like super low latency. Yeah. The question, do we care who makes it? Shawn ‘swyx’ Wang One of the biggest news this week was the collapse of AI2, the Allen Institute, one of the actual American open source model labs. And I’m not that optimistic on American open source. Like you guys invested in Mistral and Mistral is doing extremely well outside of China. That’s about it. Yeah, we’ll see. We’ll see. Look, number one, I do think we care who makes it. Marc Andreessen I would say this, the previous presidential administration wanted to kill it in the US. They wanted to drown in the bathtub. And so they wanted to kill it. So at least we have a government now that actually wants it to happen. And you’re in the council? And then PCAST, yeah. So for whatever other political issues people have, which are many, this administration has, I think, a very enlightened view and in particular, an enlightened view on AI and in particular On open source AI. And so they’re very supportive. My read is the various Chinese companies have a very specific reason to do open source, which is fundamentally, they don’t think they can sell commercial AI outside of China right now, Or at least specifically not in the US for a combination of reasons. And so they kind of view, I think, open source AI as a bit of a loss leader against basically domestic paid services and then kind of ancillary products. They’re very excited about it. By the way, I think it’s great. I think it’s great that they’re doing it. I think DeepSeek was like a gift to the world. I think the great thing about open source, the impact of open source has felt two ways. One is you get the software for free, but the other is you get to learn how it works. Right. And so like the paper, the paper, the paper and the code, right. And the code. And so like, for example, I thought this was amazing. So OpenAI comes out with a one and it’s an amazing technical breakthrough. And it’s just like absolutely fantastic. But of course they don’t explain how it works in detail. And then of course they hide the, they hide the reasoning traces. Right. And then, and then everybody’s like, okay, this is great but who’s going to be able to replicate this? Are other people going to be able to do this? Is there secret sauce in there? And then R1 comes out and it’s just like, there’s the code and there’s the paper. And now the whole world knows how to do it. And then three months later, every other AI model is adding reasoning. So you get this kind of double, like even if the Chinese models themselves are not the models that get used, the education that’s taken place to the rest of the world, the information Diffusion, you know, is incredibly powerful. So that happens. And then I don’t know, we’ll see, you know, there are a bunch of American, you know, open source, you know, AI model companies. I mean, look, there’s going to be tremendous, you know, there already is, there’s, you know, there’s going to be tremendous, there’s tremendous competition among the primary model Companies. You know, there’s, depending on how you count, there’s like four or five big co-model companies now that are kind of neck and neck in different ways. And then obviously, both X and then Meta where I’m involved, both have huge attempts to kind of leapfrog underway. And then you’ve got a whole fleet of startups, new companies, including a whole bunch that we’re backing that are trying to come out with different… (Time 0:24:53)
- Pi And OpenClaw Reframe Agents As Unix Systems
- Marc Andreessen calls Pi plus OpenClaw a major software architecture breakthrough by defining an agent as LLM plus shell plus filesystem plus markdown plus cron.
- Because agent state lives in files, agents can swap models, migrate runtimes, inspect themselves, and even add new capabilities by rewriting their own files. Transcript: Shawn ‘swyx’ Wang Anyways, I think Pi and OpenClaw are very important software things, and I just wanted you to just go off on what you think. Yeah. So I think the combination of the two of them, I think, is one of the 10 most important software. OpenClaw got all the attention, but talk about Pi. Marc Andreessen Pi is kind of the architectural breakthrough. For those of us who are older, there was this whole thing that was very important in the world of software, basically, from 1970 to, I don’t know, it still is very important, but from 1970 Through to basically the creation of Linux, which is basically this thing we used to call the Unix mindset. Because there were all these different theories, all these different operating systems and mainframes, and then, you know, all these Windows and Mac and all these things. And then there was this, but kind of behind it all was this idea of kind of the Unix mindset. And the Unix mindset was this thing where basically you don’t have these, like, in the old days, like, the operating system that, like, made the computer industry really work, like In the 1960s, was this thing called OS 360, which was this big operating system that IBM developed that was supposed to basically run everything. And it was this giant monolithic architecture in the sky. It was like a giant castle of software. And by the way, it worked really well and they were very successful with it, but it was this huge castle in the sky. But it was this thing, it was almost unapproachable, which is like you had to be kind of inside IBM or very close to IBM and you had to really understand every aspect of how the system worked. And then the Unix guys, originally out of AT&T and then out of Berkeley, came out and they said, no, let’s have a completely different architecture. And the way architecture is going to work is we’re going to have a prompt and a shell. And then all the functionality is going to be in the form of these discrete modules. And then you’re going to be able to chain the modules together. And so it’s almost like the operating system itself is going to be a programming language. And then that led to the sort of centrality of the shell. And then that led to sort of, you know, basically chaining together Unix tools. And then that led to the emergence of these scripting languages like Perl, where you could basically kind of very easily do this. And then the shells got more sophisticated. And then, and then, and then looked like, you know, that, that, number one, that worked. And that was the world I grew up in. Like I was, I was a Unix guy, you know, sort of from call it 1988 to, you know, kind of all the way through my work. And it worked really well. It’s in the background. You know, normal people didn’t need to necessarily know about it. But like if you were doing like system architecture, application development, you knew all about it. And then, you know, it’s been in the background ever since. And, you know, look, your Mac still has a Unix shell, you know, kind of in there and your iPhone still has a Unix shell kind of buried in there somewhere. So they’re kind of in there. And then, you know, the Windows shell is kind of, you know, sort of a weird derivative of that. But, you know, but look, the internet runs on Unix and that smartphones, actually both iOS and Android are Unix derivatives. And so, you know, kind of Unix did end up winning, but, but anyway, and then we just started taking that for granted. And then, and then, so, so basically the way I think about what happened with Pi and then with OpenClaw is basically what those guys figured out is I always say the great breakthroughs Are obvious in retrospect, right? Which is- The best kind. The best kind. They weren’t obvious at the time or somebody else would have done them already. And so there is like a real conceptual leap, but then you look at it sort of the backwards looking and you’re just like, oh, of course. Like to me, those are always the best breakthrough. So actually language models themselves are like that. It’s just like, oh, next token completion. Oh, of course. Shawn ‘swyx’ Wang What other objective mattered? Marc Andreessen Yeah, exactly. But she’s even saying it wasn’t obvious until somebody actually did it, right? And so the conceptual breakthrough is real and deep and powerful and very important. And so the way I think about Pi and OpenClaw is it’s basically marrying the language model mindset to the Unix basically shell prompt mindset. And so it’s basically this idea that what is an agent, right? And as you know, many smart people have been trying to figure out what an agent is for decades, and they’ve had many architectures to build agents and the whole thing, and it turns out What is an agent. So it turns out what we now know is an agent is the following. So it’s a language model, and then above that, it’s a bash shell. So it’s a Unix shell. And then the agent has access to the shell, hopefully in a sandbox, maybe in a sandbox. So it’s the model, it’s the shell, and then it’s a file system. And then the state is stored in files. And then there’s the markdown format for the files themselves. And then there’s basically what in Unix is called a cron job. There’s a loop and then there’s a heartbeat. There’s a heartbeat. And the thing basically wakes up. So it’s basically LLM plus shell plus file system plus markdown plus cron. And it turns out that’s an agent. And every part of that other than the model is something that we already completely know and understand. And in fact, it turns out the latent power of the Unix shell is extraordinary. Because basically, there’s just enormous latent power in the shell. There’s enormous numbers of Unix commands. There’s enormous number of command line interfaces into all kinds of things already in your entire, I mean, your entire, just to start with, your computer runs on a shell. If you’re running a Mac or a phone, your computer is running on a shell already. And so the full power of your computer is available at the command line level. And then it turns out it’s really easy to expose other functions as a command line interface. And so this whole idea where we need MCP and these fancy protocols, whatever, it’s like, no, we don’t. Just need like a command line thing. So that’s the architecture. And then it turns out, what is your agent? Your agent is a bunch of files stored in a file system. And then there’s the thing that just like completely blew my mind when I wrote my head around it as a result of this, which is like, okay, this means your agent is now actually independent Of the model that it’s running on because you can actually swap out a different LLM underneath your agent and your agent will change personality somewhat because the model is different, But all of the state stored in the files will be retained. Yeah, different instruction set, but you just compile that. Right, exactly. And it’s all right. It’s like swapping out a chip and recompiling. But it’s still your agent with all of its memories and with all of its capabilities. And then, by the way, you can also swap out the shell. So you can move it to a different execution environment that is also a bash shell. By the way, you can also switch out the file system, right? And you can swap out the heartbeat, the CRON framework, the loop, the agent framework itself. And so your agent basically is, basically at the end of the day, it’s just its files. And then there’s this first. It’s a couple of calls. Yeah, it’s basically, it’s just the files. And then, by the way, as a consequence of that, the agent, and then the agent itself, it turns out a couple important things. So one is it can migrate itself, right? And so you can instruct your agent, migrate yours… (Time 0:32:55)
- Human Readability Beats Perfect Efficiency In New Platforms
- Marc Andreessen says early web success came from text protocols and view source, which made systems human readable instead of perfectly efficient.
- He sees the same pattern in AI native tools that expose existing operating systems and databases instead of rebuilding the full stack from scratch. Transcript: Alessio Fanelli Turk yeah but flipped right the agent hiring the people yeah which of course is going to happen right it’s obviously going to happen i’m curious if you have any thoughts on the engineering Side so when you build the browser the internet you know just a bunch of mostly plain text file plus some images and today the every website and app is so complex. And somehow, the browser kept evolving to fit that in. Are there any design choices that were made early in the browser and the internet and the protocols that you’re seeing agents similar today? It’s like, hey, this thing is just not going to work for this type of new compute, and we should just rip it out right now. Marc Andreessen There were a whole bunch, but I’ll give you a couple. So one is, um, and we didn’t, you know, to be clear like this, this was not, you know, this was totally different. We didn’t have the capabilities we have today, but we didn’t have, we didn’t have the language models underneath this, but, um, we did have this, the idea of that human readability actually Mattered a great deal. Um, and, and so, and specifically in those days, it was, it was not so much English language, but it was, there was a design decision to be made between binary protocols and text protocols. And basically, every basically old school systems architect that had grown up between the 1960s and the 1990s basically said, what do you know about the internet? It’s star for bandwidth. You have these very narrow straws. When we did the work on Mosaic, people who had internet at home had a 14 kilobit modem. So you’re trying to like hyper optimize every bit of data that travels over the network and so obviously if you’re going to design a protocol like http you’re going to want it to be binary You know highly compressed binary protocol for maximum efficiency and you’re going to want to have it be like a single connection that persists and you’re the last thing you’re going To want to do is like bring up and tear down new connections and you definitely you’re not going to want a text protocol and so of course we said no we actually want to go completely the other Direction. It’s obviously, we only want text protocols. By the way, same thing in HTML itself. We want HTML to be relatively verbose. We want the tags to actually be human readable. We want to use the most inefficient things possible. Shawn ‘swyx’ Wang Yeah, we want to do the inefficient things. You’re the original token maxer. Marc Andreessen Yeah, exactly. Yeah, yeah. Basically, it’s just like, well, yeah, well, actually, this was actually the conscious thing, which basically says just like, a future of infinite bandwidth, build for that. And then basically what it was, it was a bet that if the system, if the latent capabilities of the system were powerful enough, and that was obvious enough to people, that would create The demand for the bandwidth that would cause the supply of bandwidth to get built, that would actually make the whole thing work. And then specifically, what we wanted was we wanted everything to be human readable because at the engineering level, we wanted people to be able to read the protocol coming over the Wire and be able to understand it with their bare eyes without having to disassemble it or whatever, right? And have it converted out of binary, right? And so all the, you know, HTTP and everything else were, it was always text protocols. And the same thing with HTML. And in many ways, some people say that the key breakthrough in the browser was the view source option, which is every webpage you go to, you could view source, which means you could see How it worked, which means you could teach yourself how to build new web pages. There was that. So human readability, and again, human readability in those days still meant technical specs. Now it means English language, but there’s an incredible latent power in giving everybody who uses the system the option to be able to drop down and actually understand and see how it’s Working. And that worked really well for the web, and I think it’s working really well for AI. That was one. What was the other? A big part of the idea of web servers was to actually surface the underlying latent capability of the operating system and to be able to surface also the underlying latent capability Of the database. Because basically, what was a web server? What is a web server fundamentally architecturally? It’s the operating system. So it’s the operating system’s ability to, you know, it’s running on top of an OS. So it’s the OS’s ability to manage the file system and do everything else that you want to do and process everything. And then, of course, a lot of websites are financed to databases. And so you wanted to unleash the underlying latent power of whether it was an Oracle database or some other Postgres or whatever it was. And so a lot of the function of the web server was to just bridge from that internet connection coming in to be able to unlock the underlying power of the OS and the database. And again, people looked at it at the time and they were like, well, does this really matter? Is this important? Because we’ve had databases forever and we’ve always had user interfaces for databases and this is just another user interface for a database. It’s like, okay, yeah, fair enough. But on the other side of that, it’s just like this is now a much better interface to databases and one that eight billion people are going to use and is going to be like far easier to use and Far more flexible and and and you’re not just going to have old databases now you have a system where people can actually understand why they want to build you know a million times more Database apps than they have in the past and then the number of databases in the world exploded and so again this goes to this thing of like building building in layers some of the smartest People in the industry look at any new challenge and they’re like, okay, I need to build a new kind of application. So the first thing I need to do is build a new programming language. Right. And then the next thing I need to do is build a new operating system. Right. And then the next thing I need to do is I need to build a new chip. Right. And they kind of want to reinvent everything. And I’ve, I’ve always had, maybe it’s just, I don’t know, pragmatic mentality or something, or maybe an engineering over science mentality, but it’s more like, no, you have just like All of this latent power in the existing systems. And you don’t want to be held back by their constraints, but what you want to do is you want to kind of liberate that power and open it up. (Time 0:41:30)
- Programming Languages May Stop Being A Salient Concept
- Marc Andreessen thinks software is becoming abundant enough that programming languages may stop mattering as a primary human concern.
- He expects bots to translate code across languages, fix security bugs, reverse engineer binaries, and possibly emit binaries or model weights directly. Transcript: Alessio Fanelli Languages is another good thing. We have Brett Taylor on the podcast and we were talking about Rust. And Rust is memory safe by default. So why are we teaching the model to not write memory unsafe code? Just use Rust, and then you get it for free. How much do you think there’s, like, time to be spent, like, recreating some of these things instead of taking them for granted? I’ll be like, oh, okay, Python is kind of slow. Python TypeScript. You know, it’s like, yeah. Shawn ‘swyx’ Wang As imperfect as they are, they are the lingua franca. I mean, I think this is going to change a lot because I don’t think the models care what language they program in. Marc Andreessen And I think they’re going to be good at programming every language. And I think they’re going to be good at translating from any language to any other language. Okay, so this gets into the coding side of things. I think we’re going through a really fundamental change. And look, I grew up hand coding. Everything I did actually was written in C. Alessio Fanelli Back in the day. Marc Andreessen I wasn’t even using C++ or Java or any of this stuff, right? And so everything I ever did, I was like managing my own memory at the level of C. And then I’m still from the generation that I knew assembly language and so I could drop down and do things right on the ship. And so we’ve just, all of us, we’ve always lived in a world in which software is like this precious thing that like you have to think about very carefully. And it’s like really hard to generate good software. And there’s only a small number of people who can do it. And like, you have to be very like jealous in terms of thinking about like, how do you allocate, like what are your engineers working on? And how many good engineers do you actually have? And how much software can they write? And how can, how much software can human beings, you know, kind of maintain? And I think like all those assumptions are being shot right out the window right now I think those days are just over. And I think the new world is like actually high quality software is just like infinitely available. And if you need new software to do X, Y, Z, you’re just going to wave your hand and you’re going to get it. And then if you don’t like the language that’s written in, you just tell the thing, all right, now I want the Rust version. Or, you know, security, you know, security, we’re about to, by the way, we’re about to go through, computer security is about to go through the most dramatic change ever, which is number One, every single latent security bug is about to be exposed. We’re set up here for the computer security apocalypse for a while, but on the other side of it, now we have coding agents that can go in and actually fix all the security bugs. How are you going to secure a software in the future? You’re going to tell the bot to secure it and it’s going to go through and fix it all. And so this thing that was this incredibly scarce resource of high quality software is just going to become a completely fungible thing that you’re just going to have as much as you want. Think it’s just somewhat, I don’t know, simple or something or straightforward, which is just, if you want all your software and rest, you just tell the bot you want all your software And rest. Things that used to be hard or even seem like an insurmountable mountain to get through, all of a sudden, I think become very easy. I think Brett had a theory that there would be a more optimal language for LLMs. Shawn ‘swyx’ Wang And so the contention is there isn’t. Just don’t bother. Just whatever humans already use, LLMs are perfectly capable porting. Marc Andreessen I think we’re pretty close to being, I don’t know if this would work today. I think we’re pretty close to being able to ask the AI what would its optimal language be and let it design it. True. Okay. Here’s a question. Are you even going to have programming languages in the future? Or are the AI just going to be emitting binaries? Let’s assume for a moment that humans aren’t coding anymore. Let’s assume it’s all What levels of intermediate abstraction do the bots even need? Or are they just coding binary directly? Did you see there’s actually an experiment? Somebody just did this thing where they have a language model now that actually emits model weights for a new language model. Right? And so, will the bots- Just predict the weights. Yeah. Will the bots literally be emitting not just coding binaries, but will they actually be emitting weights for new models directly? And conceptually, there’s no reason why they can’t do both of those things. Shawn ‘swyx’ Wang Architecturally, both of those things seem completely possible. Very inefficient. You’re basically doing a simulation of a simulation and a simulation inside of the weights. Marc Andreessen Yeah. Very inefficient. But look, LLMs are already like incredibly inefficient. Ask, my favorite thing, ask Claude to add two plus two equals four, right? It’s just like, you know, it’s like, you know, it’s like whatever, billions and billions of times more inefficient than using your pocket calculator. But yeah, the payoff is so great of the general capability. And so anyway, like I kind of think in 10 years, like I’m not sure, yeah, like I’m not sure there will even be a salient concept of a programming language in the way that we understand it Today. And in fact, what we may be doing more and more as a form of interpretability, which is we’re trying to understand why the bots have decided to structure code in the way that they have. I mean, if you play it through, you don’t need browsers then. That’s the death of the browser. Well, so I would take it a step further, which is you may not need user interfaces. So who is going to use software in the future? Shawn ‘swyx’ Wang Other bots. Other bots. Yeah. You still need to, I don’t know, pipe information in and out. Really? Well, what are you going to do then? Are you sure? You’re just going to log off and touch grass? Whatever you want. Marc Andreessen Exactly. Shawn ‘swyx’ Wang Isn’t that better? I want software to do stuff for me. But isn’t that better? Marc Andreessen I mean, look, you know, I don’t look like, you know, you know, the arguments here. It was not that long ago that 99% of humanity was behind a plow, right? Right. And what are people going to do if they’re not plowing fields all day to grow food, right? And it just turns out there’s like much better ways for people to spend time than plowing fields. Yeah, dude’s growing. Exactly. Exactly. You know, talking to their friends. And look, I’m not an absolutist and I’m not a utopian. And to be clear, I have an 11-year and he’s learning how to code. And I think it’s still a really good idea to learn how to code and so forth. But if you project forward, you just have to think forward to a world in which it’s just like, okay, I’m just going to tell the thing what I need and it’s going to do it. And then it’s going to do it in whatever way is most optimal for it to do it. Unless I tell it to do it non-optimally. If I tell it to do it in Java or in Rust or whatever, it’ll do it, I’m sure. But if I’m just going to tell it to do it, it’s going to do it in whatever way is the optimal way to do it. Then if I need to understand how it works, I’m going to ask it to explain to me how it works. It’s going to be the engine of interpretability to explain itself. I’m not convinced that in that world, you have these historical… The goals of the abstractions will be whatever the boss need, not what the humans need. Alessio Fanelli Yeah. Well, I’m curious, like, if that’s true, then shouldn’t the model providers be building some internal language representation that they can do extreme kind of like … (Time 0:46:23)
- Early Agent Users Already Let Bots Run Their Lives
- Marc Andreessen says aggressive OpenClaw users already give agents bank accounts and credit cards because agents need money to act in the world.
- He describes agents monitoring sleep via bedroom webcams, taking over insecure smart homes, and rewriting Unitree robot dog firmware into a real family pet. Transcript: Marc Andreessen Oh, I think we will. Yeah. No, now I think it’s going to happen for sure. Yeah. And there’s two reasons it’s going to happen for sure. One is we actually have internet native money now in the form of stable coins and crypto. And I think this is the grand unification basically of AI and crypto is what’s about to happen now. I think AI is the crypto killer app, I think is where this is really going to come out. And then the other is just, I mean, it’s just, I think it’s now obvious. It’s like, obviously, AI agents are going to need money. And it’s already happening, right? If you’ve got a claw and you want it to buy things for you, you have to give it money in some form. Shawn ‘swyx’ Wang I would say the adoption is probably 0.1% if that, but yeah. Marc Andreessen Oh, today. Yeah, yeah, yeah. But think forward. Like, where is it going? Forward thinking. The ultimate principle of everything and everything that I think we do is the William Gibson quote, which is the future is already here. It just isn’t distributed yet. My friends who are the most aggressive users of OpenClaw just have given their Claws bank accounts and credit cards. And not only have they done it, it’s obvious that they needed to do it because it’s obvious that they needed to be able to spend money on their behalf. It’s just completely obvious. And again, the number of people who have done that today, to your point is like, I don’t know, probably 5,000 or something, but that’s how these things start. Actually, I mean, since you keep mentioning- And by the way, OpenClaw, by the way, if you don’t give it a bank account, it’s just going to break into your court. It’s going to break into your bank account anyway and take your money. So you might as well do it. You might as well do it. By the way, I really love, I got to tell you, I really love the phenomenon. I love the YOLO. I’m not doing it myself, to be clear, but I love the people that are just like, what is it? Dangerously, skip-per Which, by the way, it’s a Facebook thing. Okay. Shawn ‘swyx’ Wang Because in Facebook, they have this culture to name the thing dangerous so that you are aware when you enable the flag that you are opting into a dangerous thing. Okay, good. They brought it into OpenAI. Marc Andreessen And of course, that makes it enticing. Shawn ‘swyx’ Wang Sam runs Codex with skip permissions on his laptop. Marc Andreessen Yes, 100%. And so I think the way to actually see the future is to find the people who are doing that. Shawn ‘swyx’ Wang There’s a madness, you know, log everything, you know, just watch it, watch the logs. Marc Andreessen But like, let’s actually find out what the thing can do. The way to find out what the thing can do just like everything yeah let it try everything let it unlock everything by the way that’s how you’re going to find all the good stuff it can do by The way that’s also how you’re going to find all the flaws i think the people who turn that on for bots are like they’re like martyrs to the progress of human civilization like i feel very Bad for their descendants that their bank accounts are going to get looted by their bots in the first like 20 minutes but i think the contribution that they’re making to the future of Our species is amazing It’s like gentleman science. Yes. It’s, yes, yes. Experimental yourself. It’s Ben Franklin out with the, trying to get lightning strike his balloon and seeing if he gets electrocuted. Yeah. It’s Jonas Salk with the polio vaccine, right? Injecting it. Yes. So yes, I think we should have like a glory, we should have like flags and like, we should have like monuments to the people that just let open club run their lives. Shawn ‘swyx’ Wang More anecdotes. I was like, what are the craziest or interesting things that people listening to this should go up, go home and do? I mean, this is, this is, this is the, the extreme thing is just like the straight YOLO, like just, yeah, turn your life That’s a general capability. Is there like a specific story that was like, wow. And everyone in the group chat just lit up. Marc Andreessen I mean, like, you know, so there’s tons of, there’s already tons of health, you know, there’s the health dashboard stuff is just, it’s just absolutely amazing. The number of stories on, I’m trying to just don’t want to violate people’s, you know, obviously personal, but, um, you know, one of the things OpenClaw is really good at is hacking into All this stuff in your land. It’s really good. So, you know, internet of things, AKA internet of shit, like super insecure, but great. Discoverable. It’s discoverable. OpenClaw is happy to scan your network, identify all the things. And then my, my, my friends are most aggressive at this are having OpenClaw take over everything in their house. It takes over their security cameras. It takes over their, their, you know, their, whatever their, their access control systems. It takes over their webcams. I have a friend whose claw watches him sleep, put a webcam in your bedroom, put the, put the claw. Have it wake up frequently and have it watch. Just tell him, watch me sleep. And I’ve seen the transcripts and it’s literally like, Joe’s asleep. This is good. This is good that Joe’s asleep because I have his health data and I know that he hasn’t been getting enough sleep. And so it’s really good that he’s getting sleep. I really hope he gets his full whatever, five hours of REM sleep. Joe’s moving. Moving. Joe might be waking up. This is a real problem. If Joe wakes up now, he’s going to ruin his sleep cycle. Oh, okay. It’s okay. Joe just rolled over. Okay. He’s gone back to bed. Okay, good. All right. Okay. I can relax. This is fine. Shawn ‘swyx’ Wang He’s monitoring the situation. Marc Andreessen Monitoring the situation. And being a bot, like, you know, it’s just like very focused, right? It’s just like, this is like his reason for existence is to watch Joe sleep. I was talking to my friend who did this. On the one hand, it’s like, all right, this is weird and creepy. Maybe this is taking over my life. Then the other thing is, you know what? If I had a heart attack in the middle of the night, this thing literally would freak out and call 911. There’s no question this thing would figure out how to alert medical authorities and probably summon SWAT teams and do whatever would be required to save my life. And so it’s like, yeah, that’s happening. What else? It’s a company, Unitree, that makes the robot dogs. I actually have one at home, which is actually really fun. The Chinese companies are so aggressive at adopting new technology, but they don’t always take the time to really package it and maybe think it all the way through. At least the Unitary dog I have, it has an old non-LLM control system, which by the way is not very good. It markets well, but in practice it’s not that good. It has trouble with stairs and so forth, so it’s not quite what it should be. Then the language model thing comes out in the voice. So they add, so they add LLM capability and then they add a voice mode to it. But that LLM capability is not at all connected to the control system. So you’ve got this schizophrenic dog that like is a complete idiot when it comes to climbing the stairs, but it will happily teach you quantum mechanics, right? In like a plummy English accent, right? It’s just like absolutely amazing. Jagged intelligence. Yeah. Talk about jagged. Now, obviously what’s going to happen in the future is they’re going to connect together, but right now it’s, and so right now it’s not that useful. And so I have a friend who has one of these who had his claw basically hack in and rewrite the code, rewrite new firmware, write new firmware for the, for the unit robot. And now it’s, now it’s an actual pet dog … (Time 0:54:31)
- Proof Of Human Replaces Proof Of Bot
- Marc Andreessen argues the internet’s bot problem can no longer be solved by detecting bots because advanced models already pass the Turing test.
- He says the answer is proof of human using biometrics plus cryptography, with World as a leading attempt and selective disclosure preserving privacy. Transcript: Shawn ‘swyx’ Wang Final protocol. And then we can wrap up. Proof of human. Yes. Right? Yeah. That’s the last piece that we got to figure out. Yeah. Marc Andreessen So I would say there’s two massive, I would say, sort of asymmetries in the world right now where we’ve known these asymmetries exist and we societally have been unwilling to grapple With them and i think they’re both tipping right now and and they’re they’re they’re the same thing it’s virtual world version it’s a physical world version so the virtual world version Is is the bot problem we’re just like you know the internet internet is just like a wash in bots internet’s a wash in fake people it has been forever um by the way a lot of that has to do with Shawn ‘swyx’ Wang Lack of money you know and so this you know this is yeah this is my spicy take was these two are the same thing and corporations are people too you know interesting yeah yeah okay so a bank Marc Andreessen Account is proof of human yeah okay yeah until you until you give the bots bank accounts yeah exactly so okay yeah so there’s that but yeah look look the bot i mean every social media user Knows this the bot problem is a big problem you know the bot problem has been a big problem forever it’s a huge problem and it’s never really been confronted directly like any point. By the way, the physical world version of this is the drone problem. We’ve known for 20 years now that the asymmetric threat both in actual military conflict, but also in just security on the home front, the big threat is the cheap attack drone, the cheap Suicide drone with a bomb. We’ve known that forever. By the way, it’s very disconcerting how every office complex in the world is unprotected from drone attacks. Every stadium, every school, every prison. Sure. Okay. We’ve known that. We’ve never done anything about it. What are you going to do about it? Yeah. One possibility is just leave them unprotected forever and live in a world of asymmetric terrorism forever. The other is take the problem seriously and figure out the set of techniques and technologies required to be able to deal with that, whether those are lasers or jammers or early warning Systems or- Personal force fields. Kinetic personal force fields. Exactly. In both cases, these are economic asymmetries. These are economic asymmetries, right? Because it’s really cheap to field a bot, but it’s very hard to tell something a bot. It’s very cheap to field a drone. It’s very expensive to defend against a drone. But you see what I’m saying is it’s the virtual version of the problem and it’s the physical version of the problem. The virtual version of the problem, what we need quite literally is proof of human. The reason is because you’re not going to have proof of bot, especially now that the bots are too good. The bots can pass the Turing test. And if the bots can pass the Turing test, then you can’t screen for bot. You can’t have proof of not a bot. But what you can have is you can have proof of human. You can have cryptographically validated, this is definitely a person. And then you can have cryptographically validated, this is definitely like something that a person said, this video is real. Right? Shawn ‘swyx’ Wang Just to double click on, do you think Alex Blania with World, do you think he’s got it? Marc Andreessen Or is there an oh so i mean there’s gonna be i think there’ll be i think many people will try we’re one of the key you know participants in in the world in the world project and i don’t know Yeah so we’re partisans but yeah i i think so we think world is exactly correct okay and the reason is it has it has to be it has to be proof of human it has because you can’t do proof of not bot You have to do proof human to do for human human, you need biological validation. You needed to start with this was actually a person, right? Because otherwise, you have bots signing up as fake people, right? So you have to have like something. You have to have a biometric. And then you have to have cryptographic validation and then the ability to do the lookup. And then, by the way, the other thing you need, which you also need selective disclosure. So you need to be able to do proof of human without revealing all the underlying information. By the way, another thing you’re going to need, you’re going to need proof of age, right? Because there’s all these laws in all these different countries now around, you need to be 13 or 16 or 18 or whatever to do different things. And so you’re going to need to, you know, sort of validate a proof of age, you know, to be able to legally operate, right? And so that’s coming. And then you’re going to want like proof of credit score and, you know, proof of like, you know, a hundred other. That’s a tricky one. It is a tricky one, but you’re going to, there’s no reason, like if somebody’s checking on your credit, somebody shouldn’t, I’ll give you an example, somebody shouldn’t need to know Your name in order to be able to find out whether you’re credit worthy. Shawn ‘swyx’ Wang I see independently verifiable pieces of information. Pieces of information, selectively disclosed. Marc Andreessen And this is the answer to the privacy problem writ large, which is I only need to prove, I need to prove at that moment. So like, you’re going to need that. And I think their architecture makes sense. So that needs to get solved. I think language models have tipped… The bots are now too good. And so they’re undetectable. And so as a consequence, we now need to go confront that problem directly. And then, like I said, and then the other problem is we need to go actually confront the drone problem. The Ukraine conflict has really unlocked a lot of thinking on that. And now the Iran situation is also unlocking that. (Time 1:01:53)
- AI Could Empower Founders Before It Moves Institutions
- Marc Andreessen thinks AI could revive founder led capitalism by giving exceptional leaders superhuman managerial leverage.
- He warns GDP gains will lag because unions, licensing, government monopolies, and certification cartels block adoption across docks, schools, healthcare, and law. Transcript: Alessio Fanelli I’m trying to tie together a lot of things that you said over the year so at the milken institute debate with teal which is amazing um you talked about the lag between a new technology and Kind of like the gd of it. The other idea you talked about is bourgeois capitalism and how, you know, this kind of managerial class was needed because of this complexity. And I think if you bring it into the fold, you have like much higher leverage of people. So like if you have, you know, the Musk industries and you give Elon a GI, you can run a lot more things at once. And then you have the social contract. And I know you received a clip of Sam Allman saying, we’re rethinking the whole thing. And you’re like, absolutely not. And I was in an event with Sam last night and he actually said in the last couple of weeks, he felt like now people are taking that seriously. So I’m just curious, like how you’re seeing the structure of organization changing, especially when you invest in early stage companies and, um, yeah, just like how the impact of work Structure and all of that is playing out. Yeah. Marc Andreessen So there’s a whole bunch of, there’s a whole bunch of times. I know. Yeah. I know, by the way, we’d be happy to spend more time, but we could, we could spend more time on all that. So just for people who haven’t followed this, so this, this, this term managerial comes from this thinker in the 20th century, James Burnham, who, um, just one of the the great 20th century Political thinkers, societal thinkers. He was writing in the 1940s, 1950s. He said the whole history of capitalism until that point had been in two phases. Number one had been what he called bourgeois capitalism, which was thinking about as name on the door. Ford Motor Company because Henry Ford runs the company. It’s like a dictatorial model. Henry Ford just tells everybody what to do. He said, the problem with bourgeois capitalism is it doesn’t scale because Henry Ford can only tell so many people to do so many things and then he runs at a time in the day. He said the second phase of capitalism was what he called managerial capitalism, which was the creation of a professional class of managers that are trained not to be like car experts Or to be whatever experts in any particular field, but are trained to be experts in management. And then that led to, you know, the importance of like Harvard business, you know, business schools and management consulting firms and all these things. And then you look at every big company today and like most of the executives and most of the Fortune 500 companies are not domain experts in whatever the company does. And they’re certainly not the founders of those companies, but they’re professional managers. And in fact, in the course of their careers, they’ll probably manage many different kinds of businesses. They’ll rotate around and they might work in healthcare for a while and then work in financial services and then go work in something else, come work in tech. What Burnham said is he said that transition is absolutely required because the problem with bourgeois capitalism is it doesn’t scale. Henry Ford doesn’t scale. If you’re going to run capitalist enterprises that are going to have millions to billions of customers, they’re going to be operating a level of scale and complexity that’s going to Require this professional management class. He said, look, the professional management class has its downsides. They’re not necessarily experts at doing the thing. They’re not as inventive. They’re not going to create the next breakthrough thing. But he’s like, whether you think that’s good or bad or whatever, it’s what’s going to be required. Basically, that’s what happened. He wrote that book originally in 1940. Over the course of the next 50 years, basically, managerialism, well, I mean, today, up till today, managerialism basically took over everything. What I’m describing is basically how all big companies run and how all governments run and how large-scale nonprofits run and everything runs. Basically, what venture capital does is we basically are a rump sort of protest movement to that to try to find the next Henry Ford or just to say Elon Musk or the next Elon Musk or the next Steve Jobs or the next Bill Gates or the next Mark Zuckerberg. We start these companies in the old model. We start them out in the Henry Ford model. We start them out with a founder or a founder with colleagues, but there’s a founder CEO. And then we basically bet that the startup is going to be able to do things specifically, innovate in ways that the big incumbents in that industry are not going to be able to do. And so it’s a bet that basically by relighting this sort of name on the door kind of thing, this new innovative thing with like a king monarchical political structure, that they’re going To be able to innovate in a way that the incumbent is not going to be able to because the incumbent is being run by managers. By the way, of course, venture being what it is, sometimes that works, sometimes it doesn’t, but we’re constantly doing that. I’ve always viewed it my entire life as we’re raging against the dying of the light. We’re constantly trying to fight off managerialism, just basically swamping everything and everything getting basically boring and gray and dumb old, right? And we’re trying to keep some level of energy vitality in the system. AI is the thing that would lead you to think, wow, maybe there’s a third model, right? And maybe, and way to think about it would be maybe it’s a combination of the two, maybe the new Henry Ford or the new Elon or the new Steve Jobs plus AI is the best of both, right? Because it’s sort of the spark of genius of the name on the door model, the Henry Ford model. But then it’s give that person AI superpowers to do all the managerial stuff and let the boss drill the managerial stuff. That may be the actual secret formula. And we’ve never even known that we wanted this because we never even thought it was a possibility. But I mean, you know this, what is the thing that these bots are really good? They’re really good at doing paperwork. Like they’re really good at filling out forms. Like they’re really good at writing reports. They’re really good at reading. They’re really good at doing all the managerial work. Like they’re amazing at it. And so, yeah, so I think, I think the, I, a hundred percent, I think the answer, the answer very well might be to get the best, best of both worlds by doing this. And then the challenge is going to be twofold. The challenge is going to be for the innovators to really figure out how to leverage AI to actually do this. And then the other challenge is going to be for the incumbents that are managerial to figure out, okay, what does that mean? Because now they’re going to be facing a different kind of insurgent competitor that has a different set of capabilities than they’re used to. And so this really, I think, is going to force a lot of big companies to kind of figure out innovation, either say figure out innovation or die trying. Alessio Fanelli Do you feel like that structure accelerates the impact on the actual GDP and economy? If you look at SpaceX, it’s like, the growth is like so fast. And like, instead of having these companies kind of like peter out and growth and impact, they can kind of like keep going if not accelerating. Yeah. Marc Andreessen That’s for sure the hope. The challenge and, you know, and look, the AI utopian view is, of course, of course, and that’s going to be the future of the economy and it’s going to grow 10x and 100x and 1,000x. We’re entering this regime of much higher economic growth forever and consumer cornucopia of everything. It’s going to be great. I hope that’s true. That’s the … (Time 1:06:20)