How Marc Andreessen Actually Uses AI

0:00This is already probably the most democratic, you know, small d technology of all time in the sense of the very best AI in the world is fully available on the apps that anybody can download. This is just a completely different kind of computer that has these characteristics that are frankly more like a person, which is it's right most of the time. It occasionally gets things wrong. When it gets things wrong, it's able to self-critique. And you have to kind of work with it the way that you work with a person. You want to take advantage of the fact that it's creative, and then you want to be tolerant of the fact that it's not always correct.

0:28AI basically has snapped everything right back into the 20-mile square radius around where I sit to just an incredible degree. So I would say like almost 100 % of the actually interesting AI companies in the West are happening at sort of ground zero right here in Silicon Valley. There's a bakery owner somewhere using the same AI as Google CEO. And according to Marc Andreessen, the bakery owner is winning. The man who invented the modern web browser and built multi-billion dollar companies just revealed something remarkable. AI is spreading backwards to society. Individuals first, small businesses second, Fortune 500 companies third, government last.

1:03the exact opposite of how computers evolve from mainframes to smartphones. Mark says half a billion people already have the world's most sophisticated AI on their phones. So the question is, why are most using it to write emails while only some are using it to build empires? Today, we're sharing a conversation Mark Andreessen had with Mark Halperin on the show Next Up. They talk about the specific prompts that transform AI into a world-class advisor, why Silicon Valley just snapped back into a 20-mile radius after five years of dispersion, and the uncomfortable truth about America's AI race with China.

1:35We hope you enjoy.

1:40All right, next up, Mark Andreessen, innovator, creator, and damn successful businessman. Early on, he invented the Mosaic Internet Browser, co-founded Netscape. And since then, he has been the animating force and investor behind a lot of very successful companies, including some at the multiple billion dollar level. Co-founded his firm, Andreessen Horowitz, a general partner there. And they do a lot of stuff about a lot of stuff. And he knows a lot about a lot. Mark, welcome. Thank you, Mark. It's great to be here. Really happy to have you. So much about AI I want to talk to you about. So we're going to spend a lot of the time on that.

2:21Um, first off, uh, it's, it's, it's, it's, uh, it's tempting to say right now. And when I think about it, I think about where are we now and where are we going? It's tempting to say it's between like really smart, highly educated people who are adapting to it. And then people who just don't have the capacity to do that in their jobs easily. But what I'm finding is at people who do what I do, people who are well-educated, very privileged, there's a have and have nots there. I'm a baby using it. I'm not using it very sophisticatedly very often, but I am Einstein compared to some of my counterparts.

2:59And I'm wondering, is that how you see it? And what do you think differentiates those who understand how powerful it is even now versus those who seem oblivious to it? Yeah, so I think there's kind of a fake story and then there's a real story. So the fake story is kind of the classic, kind of Marxist story, which is, oh, only the rich people have it, only the fancy people have it, The big tech companies have it. Everybody else is going to be left out in the cold. That's not actually what's happening. There's a lot of data on this now that has been released by these companies to justify what I'm about to say.

3:31The real story is this is already probably the most democratic, small-D technology of all time in the sense of the very best AI in the world is fully available on the apps that anybody can download. Take your pick, ChachiPT, Klaue, Gemini, Grok, Mistral, any of these. by the way, DeepSea from China, you download any of these apps, you're getting state-of-the-art like the full, most sophisticated, powerful AI capability in the world. And, you know, the number of people already who downloaded these apps is north of a half billion on its way to a billion. And individual people are figuring out, you know, basically how to appropriate this in their lives.

4:08And what you see in the data is there's, you know, maybe what you'd expect, which is there are a slice of people who just use these new systems all the time, like literally all day for everything. And, you know, in a lot of cases, they're reporting that they're getting enormous benefits from that. And then there's a lot of people who are experimenting and trying to figure it out. And then there are people who are just, you know, not, you know, for whatever reason, not interested or not engaged. But I would say it's incredible out of the gate how distributed this technology already is. And then I could just say, like, I don't have that, you know, with all my resources and with all my connections, I don't have access to a better AI than the one that you just downloaded off the App Store.

4:41And so I think this is actually, like, just an incredible story of the most advanced technology in the world being available to everybody right out of the gate. You got to get people to use it to take advantage of it, right? So I've got a friend who's drafted a book and took it to his agent. And the agent said, it's 140 ,000 words. You got to cut it down to 70 ,000. And I said, if you try to do that by hand, even if you hire someone, it's going to take you months. I can do it in an hour. And I can say to AI, don't change the style. Don't change the tone. Don't cut anything that, you know, ruins the story.

5:14and he said, that's immoral. The publisher will be mad that I did it that way. It's not right. How should I answer that person? Yeah, so the exact same arguments emerged years ago with the introduction of computers, right? I actually just, I've been watching a lot of old science fiction movies with my kid and there's this famous science fiction movie, Tron from 1982, that had the first real computer graphics movies and it was disqualified for an Oscar for special effects because it used computers, right? And so there's this kind of long tradition of like whatever the new tool is, it's sort of illegitimate and there must be something wrong with it.

5:51In this particular case, I mean, there's no doubt there are people who feel the way that your friend feels. That issue was actually litigated in the last Hollywood strikes in the movie, on the film and TV side of the creative profession. The last round of strikes actually, it's actually funny, it started as streaming strikes and then AI hit and it became AI strikes. But the settlement with the studios, actually with the unions, was the following, which is, if you're a writer and you use AI, that's totally fine. What's not allowed is for the studio to basically use the AI and then claim it was a writer.

6:23But basically what the unions and studios in Hollywood decided was it's another tool. It's like the word processor. It's like the personal computer. It's like using a printer instead of writing out a manuscript by hand. And so I think there are people who feel like your friend does, but I think the world is already adapting very fast to using it. And frankly, one of those reasons you probably pointed out to your friend is like, I don't even know that anybody could tell anymore. Right. And so, you know, if you're going to have a moral prohibition on something that people can just do and nobody knows about, like, is that, you know, is that really going to work?

6:55And so I think those those sort of self-imposed barriers are probably going to collapse quite quickly. I hope so. Fortune put out a list of the top American companies using AI. Alphabet, Visa, JPMorgan Chase, the top three. Um, we talked about just now on an individual level, if you were the CEO of a company or advising a CEO, how important is it to be on that list? In other words, how important is it to get your company to adapt, whether it's for internal or consumer facing? How important is it right now? Yeah, so this goes back to actually where we started, which is, and in fairness, the old model of adapting actually computers when computers came out, the old model was the largest institutions get technology first and then everybody else gets it later.

7:40And so, you know, the way the computer rolled out was the government actually got mainframe computers first starting in the 1940s. And then big companies got computers, mainframes in the 1950s, 1960s. Small companies started to get computers, what were called mini computers at the time, the 1970s. And then we as individuals only got PCs in the 1980s. And so it took 40 years for basically technology to cascade down from the largest organizations in the world to small businesses and to the individual. This technology, AI, is going the opposite, which is like I said, the most sophisticated capabilities are available on the consumer app today.

8:16And then what we're finding is consumers are adapting the fastest, just individuals in their lives. Then small businesses are then adopted right after that because a small business typically is just a person who's making decisions for their own business. Very, very easy to do new things. Big companies are then following small companies. And so the companies on that list, obviously some of them are doing interesting things, but in general, big companies right now are pretty tied up in knots internally, kind of in all their processes and in all their legacy systems and all their organization and training, but their unions and all the other issues they have to deal with.

8:48you know, they're actually relatively slow to adopt compared to individuals and small businesses. And then government is the late adopter, right? And so governments, of course, are already trying to figure out, you know, kind of how to adapt to this technology, but they're not adopting it very fast because they can't because of all their rules and systems and bureaucracy. And so there's been a real inversion of how technology moves through our society that's really become, you know, AI is becoming a case study for it. And so the answer to your question is, I think big company CEOs and, you know, many of them are doing this, but I think they really have to force the issue.

9:18on this inside their companies because, you know, these big companies are just such now giant bureaucracies with so many rules that, like, by default, they'll smother new ideas. And so it's a real act of leadership to, you know, to get on a list like that and be able to, you know, actually stay kind of with pride that we're on the leading edge. So if your friend owned a bakery and he said, Mark, I want you to come in and help me figure out how to use AI, well, how could someone who owns a single storefront bakery use it now? Yeah, I mean, so, I mean, there's, you know, there's dozens of ways. You know, it obviously depends on what your business goals are.

9:52They're baker dozens of ways. They are, exactly. And so, yeah, I mean, look, the first thing you can do is just say, look, you know, do a performance review for me. Like, just feed in, you know, here's my staffing schedule. You know, what do you think of it? Give me a critique of it. You know, here's the last, you know, 100 emails we've got from customers. What are the patterns of that? You know, here's the copy for the ad that we're going to place in the local newspaper, you know, put up on Facebook or whatever. Like, you know, what do you think of this? you know, let it do a performance assessment.

10:20A lot of people find it very effective for personal coaching. And so, you know, the owner might use it that way or might, you know, ask the employees to use it that way. And then I think where the power really kicks in is, say, you're a, you know, you're a small business owner, you've got one bakery, now you want to have two bakeries, right? And you want to have a brand. And then maybe if that works, you're going to have five and then 50 and then 500. And then you're going to have, you know, packaged products going to supermarkets and so forth. And then there you basically turn the AI into a thought partner, right?

10:44And you basically say, okay, what, you know, what are the best ways to expand from a single outlet to multiple and turn this into a larger business? And the AI, because it's been trained on some large percentage of the total amount of human knowledge, it has within all the information on how Ray Kroc turned McDonald's from a single restaurant in McDonald's and how all these other entrepreneurs before actually did this. And so it can explain with you and help you figure out how to do this for your own business. And then the thing that you get into as you use it is basically, is just like, wow, it's like having the world's best coach, mentor, therapist, advisor, board member, but it's like infinitely patient.

11:23And so it's like, it's happy to have the conversation. It's happy to have the conversation 50 times. It's happy if you admit your insecurities, it'll coach you through them. It's happy if you run wild speculations that don't make any sense. It's happy to do all that at four in the morning. And so the people who are using it a lot are finding it's, it actually turns out to be very supportive in their real life. Could you say, here's the recipe for our best-selling cinnamon rolls? How can I make it better? Yeah, 100%. Yeah, you can say, and by the way, part of the art of AI is what questions to ask it.

11:54Because it turns out, it can answer many, many different questions. So you have to actually create it at best. Yeah, and so you can say, here's my current recipe. How might I improve it? You can also say, what's the best cinnamon roll recipe in the world? Word back verse from that. And then you can also say, look, I want to make the best run in the world, but I need to do it at a price. You know, what are the ways to cost optimize? All right, so the other thing you can ask it is you can ask it, what questions should I be asking? Right, and so you could plug in, I run a bakery, dot, dot, dot. What questions should I be asking?

12:25And you'll find it's actually a thought partner in helping you figure out what questions to ask. Right, brilliant. So the other day I said, I want to know every Republican who voted against any of the articles of impeachment against Bill Clinton. Give me the list of every Republican. And it came back and it listed some Democrats and mixed in there. And it said they were Democrats. And I said back, I only want Republicans. And it said, oh, sorry, I inadvertently included Democrats. Now, I could understand a human being, a junior researcher doing that, but how could AI make a mistake, have it pointed out to it and say, oh yeah, how could that be in the model?

13:03Yeah, so this gets technical and I'd be delighted to get deeply into the technical details. I'll try to resist. It's a new kind of computer. And the way to think about it is computers up until now have been what you might call like hyper literal, right? Where computers up until now, like they do math really fast, but like they do the same thing every single time. They exhibit no creativity whatsoever. And if you expect them to exhibit creativity, like, you know, they just can't do it. And then if they make a mistake, it's because the human programmer made a mistake. And that has made computers, you know, super useful for running like, you know, large math exercises, you know, doing a lot of things that computers do.

13:37but of course, computers have never been creative, right? Computers have never been able to write you poetry or work with you on your cinnamon bun recipe. Like it's just never even been a thing that we can think about. So it's never had kind of that human element of creativity to it. This is just a completely different kind of computer that has these characteristics that are frankly more like a person, which is it's right most of the time. It occasionally gets things wrong. When it gets things wrong, it's able to self-critique. And you have to kind of work with it the way that you work with a person.

14:05And so you have to basically figure out as you use it. Like you want to take advantage of the fact that it's creative and then you want to be tolerant of the fact that it's not always correct, just like you're working with a person. Now, having said that, when it makes easily avoidable mistakes, where it's just like makes boneheaded fact mistakes, you know, we call those hallucinations. The latest systems are much, much, much better at not doing that. They're much more accurate. And in particular, for anybody watching this, if you want to kind of see this in action, I'll just give an example. If you buy the full version of ChatGPT, there's a model called GPT Pro, GPT 5 Pro, which is the latest one.

14:38And then there's something called Deep Research, which is a switch that you turn on. And if you use GPT 5 Pro or Deep Research, you ask a question like that. Like at this point, I think it's, I wouldn't say it's bulletproof, but like it's really good as things fastly grounded. And literally you can watch it work and it'll literally go out on the internet and it'll like check all the authoritative sources. And it'll, you know, it'll go on converse.gov or whatever and check the voting records and verify that. And so I think that problem is being ironed out, you know, kind of as we speak. Yeah. For people watching, listening who haven't used it or not used it much, the thing you said about good prompts is so key.

15:11And that does differentiate the people really getting productive at it from not. And one of the things that you and I have talked about is it is hilarious. It can write, if you give it the right prompts, it can write stuff that is so funny. And where does that come from? How does it have the capacity to understand? Because humor involves metaphors and sophistication and humanity. How can it do that? Yeah, so this gets to this idea of how it's trained. So basically what these systems are is they're basically the accumulation of human knowledge over time. And by the way, most of the training data is just literally the internet, right?

15:52So one of the reasons that this is happening now and now 20 years ago is because the internet finally got big enough, but just the web finally got big enough to have like all this information in it, right? And so if you go out on the internet today, you know, you can find, you know, all kinds of material online that's, you know, it's, you can find like, you know, classic screenplays from, you know, the gold age of cinema 400 years ago. You know, you can find, you know, literally, you know, people joking with each other on social media all day long. You know, you can find, you know, professional comedians doing oral histories of, you know, how they did great comedy.

16:21So there's just like, there's incredible amounts of information that are online about comedy and about what's funny. And then all of that information is in the training data. So it's all kind of fed into the AI during the training process. And the AI basically processes it through like any other kind of data and comes out the other end and just basically is like, oh, you know, now that AI is a world-class expert in humor, right? And of course, you know, look, you could be an expert in humor and not actually be funny, right? And they're, you know, they're, I don't know, probably professors of comedy or something like that and psychologists who are very funny.

16:50But like, it just, it knows so much about what humor is. It knows so much about the pattern of jokes. It knows so much about what make people laugh. It has so many examples, you know, to be able to learn from. And, you know, the professional comedians will tell you there are patterns to comedy, right? You know, there are, you know, I worked with a professional comedian once on something and he said, you know, the key to it is specificity. Like you need to get, you need to just like nail the reference. You know, another comedian will say the key is, you know, timing or pacing or the callback to the previous jump or whatever, you know, the punchline, whatever it is.

17:18And so it just, it knows all that. And then it's just, you know, it's now so powerful. It's actually, yeah, quite honestly, especially at like two in the morning, I find these things historical. Yeah. All right, just a minute left before we take a break. What's your advice to an individual who's barely used it or hasn't used it? What's your advice to them to sort of how to get started? Yeah, so, I mean, by far the best way to do it is just download it and use it. And like I said, you know, there's several good, you know, Elon's got Brock, you know, which is now fantastic. By the way, it's actually interesting how these things are getting built into products now.

17:50So the new versions of X, formerly Twitter, actually, if you go to any post on X, there's a little GROC icon that looks like a little black hole icon in the upper right corner of the post. And if you click on it, it actually calls up the GROC AI to explain the post to you, right? And so it literally is like, it can explain to you that if there's some post on politics or something, you don't understand what's happening. You just have lost the thread of the topic. You just like block that button and you get in a dialogue with AI, a little AI window pops up and explains the post and you can ask it for more details.

18:22And so it's like built into that product. Google's actually built AI now into search. And so now when you do searches, it has this thing called AI mode and you bonk on it. And in addition to getting the 10 blue links from Google, you now get into an AI dialogue. And so you just start using those products or you just download one of these apps and start using it. And like I said, a really great question is like, okay, how do I use you? Or you can say, teach me works really well. You can say, you know, teach me how to use you in the best way. You know, teach me how to use you for my business, you know, for this project.

18:53And it'll, you know, these things love to talk and it'll happily sit there and shatter away and take you through it. Yeah, just my sort of analog to the way you just said it is ask it for what you want as specifically as possible. Don't hold back. Be really specific and it will do what you ask. All right. More with Mark Andreessen. That's next up. Stay tuned. If you're 64 years old or older, this is an important announcement. The Department of Justice recently sued three major Medicare brokers for claiming they were unbiased while allegedly pushing people into plans that got them the biggest kickbacks.

19:30It's true. So many insurance agents, they just can't be trusted. But you also cannot rely on the government to give you the best information that you need either. That's why I want you to know about something called Chapter. Chapter was started by people who went through all of this personally after their own parents were pushed into the wrong Medicare plan by an agent who was more focused on commissions than on good care. Chapter's mission is very simple. They want to give every American the honest, straightforward Medicare advice that they deserve. And here's what makes them different from everybody else.

20:02They're the only Medicare advisor that compares every plan nationwide, not just a few. That saves their clients an average of$1 ,100 a year. There's really no reason not to call. It's quick, it's easy, and they can review your options in under 20 minutes. If you're in the right plan already, they'll let you know that. But if there's a better plan, they'll help you make the switch. This could be the most important call you make this year. Dial pound 250 and say chapter Medicare to get peace of mind. Again, that's pound 250 and say chapter Medicare. All right, welcome back. We're here with Mark Andreessen still.

20:40Mark, are you of the school that says that the United States is in an existential struggle with China? Do you subscribe to that point of view? I was going to say, I hope that's not true, right? You know, I hope this is not going to walk all the way to the situation that we ended up in with the Soviet Union. I mean, you know, like you, I grew up in an era, I'm sure, see if you remember this, like, you know, growing up thinking there was a significant chance that like we're all going to die, you know, from nuclear war. And so I hope it doesn't get back to that level of intensity. But I do think there are a lot of historical parallels to what happened between the U.S.

21:15and the USSR in the 20th century, you know, happening right now. And, you know, you have, you know, two hegemonic superpowers that both have, you know, visions, you know, they both have visions of how society should be structured. and how the sort of global political systems are restructured. And, you know, obviously I think America's is better, you know, but they both have visions. And, you know, they're both working, you know, to that end. And then they both have international strengths and weaknesses militarily, technologically, economically, culturally. And, you know, there is that kind of geopolitical fight happening.