From selling startups to Google to backing multibillion‑dollar AI winners - Anish Acharya [a16Z]

I'll say a little bit of a cheeky thing, which is there are no distribution problems, only product problems. If I years ago the top price point a consumer would pay is $20 a month. You can just build great software and sell it for a lot of money. We're in a world where the kind of LLM wrapper point is not a relevant consideration anymore. Winning is underrated as something that will actually deliver a lot of value and sustenance in your career life. Today on Billions, I'm sitting down with Anish Assharia. He sold his first company to Google, his second to Credit Karma, then stayed and helped scale their US card business to nearly a billion dollars in annual revenue. In 2019, Anderson Auroitz made him a general partner. Since then, Israel's Series A Indial, which just hit a $17.3 billion valuation in October. Most vices have never operated anything. Anish, built, scaled, sold, and then learned how to pick. Anish, thanks a lot for being here. Thanks for having me. You have an incredible setup and I'm really excited to have this conversation. You started as an entrepreneur, two startups, two acquisition, Google and Credit Karma. What exactly did you learn from selling companies that you actually, contrary, learn from building them? Yeah, that's a great question. I think there's the success of being a founder, this year from your own companies. There's a success case for your company, then there's also just the day-to-day joy and pain of building it. I will say that there are times at which the ratio of joy to pain, independent of whether the outcome is successful is much higher and better. That's how it felt in 2008. Despite the fact that it was a financial crisis, it was very hard to raise money. Mobile was new, social was new, and there was just a lot of things for technologists to work with when building a startup. It was easy to get customers, people were enthusiastic about downloading new apps to their iPhone. It was a very, very fun time to be a founder. My second company in 2014, it just wasn't as fun. Distribution was hard. You had to spend a lot of time on marketing. There wasn't a big new technology for founders to play with. This technology had sort of settled around mobile. I think today is a lot like at 2008, when it actually is like every single day or week you wake up. There's a whole new set of technologies and capabilities to play with. I will say the day-to-day experience of a founder changes a lot depending on where we are in the macro. This feels like a very fun one. Since today feels like 2008 for you, do you want to go back to being a founder? Are you happy? I think my founder days are done. If I was to ever do it again, this would be the moment. Even if you look at it, what are the attributes of your day-to-day experience as a founder? What I mean by that is that if you have a distribution problem, one way to solve it is by being more ambitious on product. We're seeing incredible consumer enthusiasm, tons of organic downloads for some of the best products. Also, you have so much to play with from a technology primitive perspective. I think there is an opportunity to be ambitious on product that didn't exist three, four, five years ago, which is cool. Investors are really excited. Consumers are also paying for software, which is new. If you look at the top skis of Gemini, ChatGPT, Quad, they're all $200-$300 a month. I think Groc, heavy is $300 a month. This price point is unheard of. Five years ago, the top price point a consumer would pay is $20 a month. Now you've got people willing to pay directly, which means you don't have to build ad networks into all this indirect monetization. You can just build great software and sell it for a lot of money and be pretty happy. It's quite like, I love what you said about distribution. Because I feel like, I started my company in 2018 and I think in the last five years, everyone was talking about distribution. I think everyone wanted to become an influencer. Everyone wanted to have their audience because obviously you can definitely, when you have a large audience, sell faster, but with ChatGPT and all these products going from zero to, I don't know how many million users in just weeks, it's pretty insane. Why do you feel it's such a difference right now versus the past? Well, I think consumers are so enthusiastic about the new technology. They're just so excited about what these technologies can do. But even if you think of the fact that every technology we've had for the last 50 years has been shaped in a really specific way. It's made us better at math, better at doing quantitative exercises. We've built some interesting human and social products like social networks on top of those foundations. Now we have a new technology foundation that are fundamentally human and emotional in a different way. That's so much of our experience. Among the very rational things we do, the conscious mind, things we do day-to-day as humans, underlying that is often emotional, maybe even spiritual needs. I think we have a technology that addresses more of the human experience than we've ever had before. How can you not be so excited about that as a person? Of course, there's also many other things happening. There's a lot of press about the new technology. People are very curious about the new technology. But I think the biggest thing of it is just the nature of the technology is more human than any we've ever seen before. I agree. Since you were also like, I mean, you worked in 2008 and it was also a very exciting era for SaaS, for apps, etc. I feel now it's kind of like, do you feel like the order of magnitude has changed? How big can this be now? Do you feel? I mean, think of the speed, I think CHAT GPT is 900 million weekly active. The speed of that product to that scale is, it's an order of magnitude faster than the next fastest. If you look at the prices, they're an order of magnitude higher. If you just sort of squint at the technology surface and the things you can do, it's got to be two or three orders of magnitude more powerful than anything we've ever built. Mark said something recently which struck me which is like, forget about the internet, this is bigger than the wheel. I don't think that's the first statement. I really don't. I love that. For some of us who have seen Quad Code or Codex 5.2, I feel like I've seen God. It's the most powerful thing we've ever invented. I think in human history or certainly that I've seen. Yeah, it's insane. Talking about distribution because I think recently, Google published their numbers when it comes to the usage of Gemini. I think at the beginning of the year, I think like Chad GPT had around 85 to 87% of the total usage and now they're down to 60 ish something while Google was around zero and now is at 25 ish. What's your take when it comes to all this model popping up? Obviously you have Gemini, you have Claude Atoncho, you have OpenAI, etc. It's very interesting. There's maybe two notes. One is if you just look at Chad GPT, that's the noun and the verb. They have a place on most consumers phones in a way that even Gemini still doesn't. I think that's a very powerful position to be. Look with that said, if we look at the point versus the slope, if you look at the point, you could say, well, from a relative basis, Chad GPT is falling behind. If you look at the slope, all of these things are growing like crazy. On an absolute basis, they're all still growing. If I'm Sam Altman, I'm not unhappy with where they are. Look, it's real testament to both Google and Anthropic that they're succeeding and of course, Grock as well, because they're succeeding in different directions. If you look at where Anthropics succeeding, they've singularly been focused on code and everything downstream of coding agents. That seems to be working. If you look at Gemini, they've done a really nice job of integrating with the Google ecosystem. I think one of the most under-discussed technologies is the integration of search into nano-banana. Now, if you say, hey, give me an accurate image of this bottle of wine instead of just coming up with an image of a bottle of wine, it actually can reference the actual, actual real image, which is very useful in cases like product photography for e-commerce websites. Google is doing the thing that it does best. You know, very practical low-cost AI. Anthropic is focused on coding, Chad GPT in OAI. It feels like the most horizontal product of them all. Grock is doing a really good job of extending the X ecosystem. Yeah, that's very true. I'm actually wondering what you think of. There was this article about a thunder mode. That's kind of what happened at Google where it was code red. Then I think it's Sergei that came back to the office. How do you see it? Do you feel like, because I was talking the other day with Nikola, who is a former founder of Algolia and GP at Weissie now. For him, it was telling me, Google hasn't been innovating a lot in the recent years. If you take the last 10 years, 10, 15 years, nothing has really happened, changed or whatever. But now it feels like something is happening. Do you feel like such a big company can still be highly innovative and competitive when a new technology like what's happening right now appears? I think that what big company is who are capable often do well is using the new technology to extend their current products. They do a good job of saying, "Hey, let's do the existing markets in which were very dominant. Let's use this new technology to be even more dominant. So then Google search will be an even better search than it's ever been before. And Microsoft PowerPoint will be an even better PowerPoint than it's ever been before. So I don't know that their lead in those markets goes away, but I do think the new markets get created. Things like AI native image and video and audio, they're just not set up to be the winners in those new emerging markets. So I think usually when you have this new technology, it's sort of this wash of value. And then there are of course some very specific extreme winners and losers and everybody benefits. I think the shape of success for existing companies is usually in their existing markets. And do you feel like, 'cause you mentioned you are obviously like when you're a big company, you can leverage technology to improve what you were doing in the past, but what you were doing in the past might not be relevant. Because if we look at the usage typically of a chat GPT or a cloud, it has totally changed. So do you think the new generation will still use search? - It's a good question. If you think back to the historical examples, the browser came out and was a huge threat to operating systems, but we still use operating systems. A lot of people still pay for windows. Now maybe from a kind of economic value perspective search has been a lot more valuable than operating systems. Maybe 10X more valuable, but operating systems as a market I don't think has shrunk. So I think the same thing happens. The sort of models and language models as a front door to the internet. Maybe 10X more valuable than search. I don't know if search goes away though. - Okay, interesting. (laughing) And the-- - And you know what you're going about Google and Sergey and everything else. I don't have any inside information, but I will say that it feels like this moment in time where all the technology is new, which I think gives a lot of founders energy versus working on the distribution and scaling problems with no new technology, which I think, you know, for somebody who's already made it, maybe is a little bit too boring. - Yeah, it's too repetitive. (laughing) - Yeah, yeah, it's just, I mean, they're just not getting energy from it. The way you get energy from waking up every day and being like, oh my god, there's a new, you know, open AI model, there's a new Google model, there's a new capability of video models like every day is Christmas. - And I think like talking about models, you know, when Chad GPT came out, 'cause I mean, for me, like obviously with my company, I started using like Chad GPT at, I mean, it was not even Chad GPT, it was way before like we were just like leveraging AI, like to create text, et cetera, et cetera. And it was not quite there yet, but when Chad GPT came out, everything becomes like a lot more relevant, a lot more useful, and they were like a whole shift in the AI market. So at that point, everyone was saying, okay, like the biggest winners are gonna be like the large language models. And if you are just like a LLM rapper, essentially, you're dead. - Yeah. - What's your view on that? - Yeah, I don't think that's a relevant consideration anymore. So if we look back in history to late 2022, what was unclear at the time? So Chad GPT was November 22, what was unclear at the time was, would one company have the sort of best, you know, best in class model on an ongoing basis? So would OpenAI always be one to two generations ahead of everything and everyone? Was it the sort of compounding network effects style product where they'd be one winner? In a world where there would be one winner, I think that consideration is very relevant, because then you have to look at the market and say, well, either we have to train our own foundation model that competes with OpenAI does, and we've all the sort of disadvantages of being sub-scale, or we actually have to build in their ecosystem, and they can take as much of our gross margin as they want, because they're the sole supplier of intelligence. That's not what happened at all. Instead, if you look at what's happened, many models are cutting edge, and because you almost have this dis-economy of scale, where you have distillation where models can train on the outputs of other models, you really can't stay ahead for more than a few weeks. So instead, we actually have this sort of huge supply of foundation models. Now, I still think for the 20% of use cases, the cutting edge models are better than everything else. So maybe for coding, Opus 45 is a bit better than everything else that's in market than Gemini 3, and maybe for multimodality, something like GPT Image 1, is a little bit ahead of the others, or maybe Nanobanana. But for the 80%, I think there are substitutes. And I think most businesses and most buyers of the API just need the 80%. So we're in a world where the kind of LLM wrapper point is not a relevant consideration anymore. And with the switching cost being extremely low, because I mean, super easy. You probably work the same way as I do. You have like a chat GPT Gemini, everything's open on the other tab. And eventually, so what's kind of your view? Do you feel like people will use potentially like 304 models and eventually stick to one, or do you think people would keep like a certain model for a certain task? Yeah, I do. I mean, it's sort of like if you have a team of people, and they all are, in a sense, if you have five people, they could all do a basic set of things, pretty capably, right? They can organize an event, they can send an email, maybe they can program, they can represent you well in a meeting, but then they all have their specialization. Maybe one of them is really good at Closier customer who just doesn't want to sign the deal. And one of them is really good at culture and getting the best out of the team. Like everybody has their specializations, which is why, even though for the 80% all those people on your team are sort of substitutes, for the 20% you need them all, I think we're going to need and rely on all of the models. But I think an important question though that's implied in yours is, okay, if these companies have infinite budgets and they sort of own the models, which gives them some economic leverage, are they going to move upstream and build the apps as well? And I think that, look, there are some areas in which they are going to build apps, and that is going to be a threat to apps companies. But I think there are many areas in which app companies are advantaged. You know, one of the big ones, and I think cursors are great example of this, Kriya is a great example of this, any product where you benefit from being multi-model. So when you actually use a creative tool, you don't want to just use nano-banana, you want to have access to open AI, nano-banana, cling, all of them, when you name it. So using a single interface to access all the models is powerful, and Google is never going to actually provide you with an interface to open AI's models. So there are examples like that, or even a network effect product, Wabi is a great example of this. You know, it's an app where you can create many apps and also consume many apps. That's a classic app platform feedback loop that has a sort of compounding effect. So even if Facebook wakes up in a year and says, "Oh my God, we need to replicate this." Sure, you can replicate the app, but you can't replicate the network just as it's trivial to replicate Instagram, but impossible to replicate Instagram's network. - Yeah, I agree. And I think it's interesting, like, 'cause you know, you mentioned the fact that aggregator of different models, because even if it's a wrapper, it can aggregate just like a different model. It can be like extremely helpful, and on specific use case definitely have like a huge engagement and traction and be helpful for the user. But do you see like other companies? So if we take, for example, I don't know, like, lovable, when lovable got started, and I think it's a great company, it's impressive what they've been doing, et cetera. But then when you see like, clothe codes. And so what's kind of like your view? Do you feel like this is typically an area where, like what AI is doing, like what the big models are doing can be like a threat to these companies, or how do you see it? - I admire lovable. I think it done a nice job. You know, companies like Repplet have done an exceptional job, cloud code, obviously they're doing an incredible job. I think they're pointed at different parts of the market. And it's easy to zoom out and say, oh, they're overlapping. You know, to me, a great example of this is legal. You know, if you look around and say, well, you know, Harvey is dominating in legal, and they are doing an amazing job, there's never gonna be another legal A. A company like that's insane. Like if you think of the legal industry, that is infrastructure for capitalism, right? All of capitalism runs on legal and law. So you cannot just have a single winner. It doesn't really, it's like, say software is a single category. So I think there is more specialization than we appreciate. There's room for many of these companies to succeed. And then look, I think the other thing that's under discussed, and by the way, this benefits, cloud code, but also lovable, and also cursor, and also CREA is that you get this data exhaust, the RLHF, which is people put in prompts, and then they get output, they respond to the output, they're getting all of this feedback, and you saw that with cursor releasing their composer one model. So they then are able to train their own model after the kind of unique exhaust from their user base, which is sort of another interesting advantage of aggregators. - Yeah, it's, yeah, very true. And, because you talk about prompting, so it made me think of something like, you know, at some points, there was this kind of trend of prompt engineering. And I remember when Chagy Pt was out, I would spend hours on Reddit and do this super long prompting to make sure, but eventually, you know, when I was looking at it, I was like, okay, this is not a feature, this is like a bug, you know, like models eventually, they're not gonna need you to prompt. So I feel like way less prompt engineer job offer, you know, everywhere. So do you feel like there are things like this that we currently doing with all these tools that will deserve it? appear as models get better and maybe you could share like some business opportunities that you would see from it. Yeah, it's interesting. I'll tell you a funny story. So I got my first computer in the 80s and then the late 80s, you couldn't really even buy games, many games. And the way that you would play game is at least my parents would take me to the library. You would check out a book. The book would have the source code of a game. You would go home, you'd type all the source code in and like have it help you if you made a mistake. And then you would run it. And that's how you would actually play the game, which seems insane, right? I mean, you never do that for call of duty or grab that or something like that. And even back then for a really simple game, it was a lot of work. The same thing I think is true of some of this, like, you know, the workflow from 23, if you even look at this, there was a very viral prompt from Halloween, which is it was this is on TikTok and it was a sort of take a photo of you. And then it's like, you want to bet in the early 2000s and there's a scary guy coming in with a knife and it was this Halloween trend. But if you then look at the comments, people would say, okay, how do I do this? And there would be this three page prompt, which is crazy, right? So now if you look at a company like Wabi, they're taking what would otherwise be shared as a prompt and saying, hey, we're just going to internalize that prompt to a mini app. So I think things like mini apps are interesting containers for what might otherwise be a prompt. And those are areas that I think there'll be a lot of opportunity and growth. Yeah, definitely. No, it's interesting. And yeah, going back maybe to your entrepreneurial journey and after I want to also discuss, you know, like the investment, I mean, you know, like after Credit Karma, like when they acquired like your company, you decided to stay. And I think you were running like the US car business and you ran it to like close to a billion in revenue. Yeah. I think a lot of founders when they sell their company, you know, they usually like live or you know, they're just wait for the or now and go like what made you stay? And what did you see in that company that got you excited? I mean, my personal framework is if you're learning and winning, you should stay. And by the way, even if you're just learning or just winning, maybe you should stay. But if you're learning and winning, you should definitely stay. And I think this sort of it's underdiscussed how much time because of the time value compounds when you're inside of an organization. So I was learning a ton. If you look back at 2015, it was a very tough time to build a pure play consumer company. The main part sort of Google Facebook, Apple were highly dominant. They were actively deep platforming companies that they deemed as a threat. So it's just a really tough time to build a new core consumer company. Meanwhile, in consumer fintech, it was this incredible sort of renaissance moment where all of a sudden there is new technology, new sort of legislation, and a new set of founders who said, hey, let's let make it really easy for people to save money on their credit card bills and better understand their credit. And I will tell you, it's funny, I'm like, I remember when I was building some of my social companies, I tried to explain to my parents what they were with the products men. And you would always do this very complex explanation of, well, if you look back in human history, people would connect by sitting around the kitchen table, playing a board game. And therefore, this is a social gaming product that's really about human connection. It's just like, okay, what is this? Whereas when you're helping with their money, it's like, look, people just need more money and they need help with their finances and when we're helping them. And everyone's like, oh, that makes sense. That's a great thing to work on. So I think it was both spiritually satisfying as well as fun because it was genuinely new. And there was no Google Facebook Apple. It was just the banks you were competing with or no one. So I learned a ton and then the company was, was, you know, incredibly dominant in its market over 100 million users, you know, 45 million active when I was there. It was a big, big company. So I had a ton of fun there. And, you know, it really did set me up for in recent Horowitz. I wouldn't be here if not for that. No, that's that's great. And you say like, when you learn and you grow, like, it's usually the best place to be at. What are the things, you know, that, that you learn there, that's right now are quite helpful in the way you address and see the whole market. Yeah. I mean, I'll tell you one big insight. This is both from credit karma and consumer fintech is that paternalism kills products. Okay. So let me tell you what I mean by that, which is especially in finance, but in many parts of software or consumer thinking, we want this set of things for the consumer that they may not want for themselves, you know, like, look at America, America is irrationally optimistic in this incredible way about everything, right? Everybody is certain they're going to be rich. Everybody loves spending and hate saving. You know, everybody sort of glorifies or there's more of a glorification of like having a big life versus having like a cautious, careful life. And we can criticize some of that, but I think a lot of the magic of this country is from that irrational optimism. It like gets willed into existence. Either way, if you build a product that is dependent on people not drinking their morning coffee, you know, it's just a bad assumption. It's judgmental. It's a sort of decision in a direction that consumers won't make. And yet so many fintech founders are like, Hey, people are just irresponsible. So let's make this product that really helps them, you know, understand their own irresponsible, like nobody wants that. You know, it's just like a guilt trip on steroids. No, I want to like feel good about the decisions that I've made. I want to feel informed about the decisions that I'm going to make. You know, if I'm going to spend money anyway, I want to do it in a smart way, but I don't want to be told not to spend it. And I think there is, you know, there's a thousand fintech companies that have lived and died and not one assumption. And it's something that chronic harm agar really right. So you feel basically like if you essentially like try to sell against the wind, you will fail. And even though like, because I think I have the more or less the same view, you know, I've seen and I've met like a lot of people who have like, I mean, their interest at heart is quite nice. So typically you would see like all these us that are, they're going, yeah, you know, like they're going against, for example, like the TikTok, the Instagram and they say, like, stop doom scrolling. Yeah, of course doom scrolling is bad, you know, and it's like, yeah, and they want to help you like learn stuff. Yes, of course, learning is great. And they have like these gardens. But I don't, I don't see how like companies like this can succeed when you go against, you know, like the human basic foundation, which is however like people, people try to, you know, like every time they have an effort or whatever, if you simplify it, it works. If you try to make it like harder, it's never going to work. I'll give you a great example of this. You know, so I think it was in the 1970s, there was a bunch of competing studies about human obesity and what cost it. And essentially the best thinking out of Europe was that, hey, it's sugar. And the best thinking out of America was, hey, it's fat. And there's also some commercial interests in both directions. So I think the Europeans were right. And it turns out that we as a collective population of Americans just got fatter and fatter until we know we all real, the problem was, but even upon realization, I think trying to tell people, hey, don't eat sugar, don't do this, don't do that. Like we got so extreme, we went to this whole body positivity movement and then ozemic. You know, so you sort of like, you, no matter how many times you tell people to sort of operate in a way that is against either human nature or cultural tailwinds, they just don't do it. You often need a new technology breakthrough. And it's amazing how much that's technology. I think it's going to transform our society. So I think just being attuned to what is happening culturally, by the way, it goes in the other direction as well. Like sometimes if you look at location sharing 10 years ago, many people predicted that would fail, especially a journalist, love to say, oh, it's so the privacy, it's so sensitive and safety and people don't want to share their locations. And that was just the whole generation, my generation, thinking like this is so crazy. And now if you look forward seven, eight years, you know, people share their locations with everyone, find friends with their friends, with their exes, with their, I was like, why? But they do it. So if you were building a product that, and you were early to that sort of cultural change, you could have built a really successful product. And a few people like life 360 did or snap. I mean, if you were one of these big displeievers, you, you know, you sort of missed it. So it goes in both directions. And talking about like sharing a location, I think, I'm not sure if you read this, but the kind of like new where the French president was because of his bodyguard on on Strava, you know, we're like running in. When he was, when he wasn't supposed to be at the specific place, so they were kind of like a scandal over this. So yeah, it's, it's quite, quite interesting. And you know, like, because you mentioned, obviously, like, you know, that some technology or like at least some, you know, like new trends. So if we take TikTok or like Instagram new apps, who are like revolutionizing, like the way people interact, communicate, consume content, et cetera, TikTok has been criticized a lot for, you know, like the endless dooms, doomscalling that people can can can do. But as you said, you know, you can't go against human nature with AI. What do you feel are like the, yeah, the risk overall with with what's, you know, like with all the models and the, the chat version like that feels as you mentioned earlier, very like human like. I actually think of it in the other direction. I think, you know, if for all the anxiety about social networks, and we can have a separate discussion about whether, you know, the sort of reports on the impact on society are accurate or not, but actually think that social products had some unintended consequences. is perhaps that I don't see in AI products today. If you look at the AI products today, people are using them as a way, one to improve themselves. So they're using them a lot to sort of reflect to be able to iterate, to have a safe space to discuss uncomfortable topics. Two, I think that the impact of AI on loneliness is substantial. The founder of replica Eugenia is a really impressive person and she'll tell you stories all day long about people that are in these really tough situations, personally, because they just don't have anyone to talk to. And yeah, me and you and I, we have this sort of embarrassment of social riches, right? You are like, oh my God, I want to respond to all my texts because all of these people want to hang out with me. Like your calendar is always booked, but that's just not the experience of a ton of people. You think about positive impact on senior citizens and folks like that. Like there's just so much benefit to be delivered to the customer here and we're seeing the signs of that. So I think that all of the folks that are sort of critical of the technology maybe are not in a place where they're experiencing those human benefits, which I think are really substantial. Yeah, I agree. And I think like what's really surprised me when Chad G.P.T. and I started using it like more and more, it's like people were obviously criticizing it, et cetera. And then it's like you look at health and you look at the way you know like doctors would interact with their patients. And typically in France, like if you want to become a doctor, it's very, very complex. Like the first years are extremely complex, but you only do mass, physics, a bit of like chemistry, biochemistry, et cetera, but it has nothing to do with empathy, with communication, and so you end up with a lot of like doctors who are just like kind of like heartless. And I understand it might be hard for some people, et cetera, but when you look at the tests that they've done with, it was basically a doctor behind the computer or Chad G.P.T. answering or like another LLM. And you could see that on the empathy score, it graded like much higher than actual human beings. So to your point, I think like AI has basically like endless patients, which I think for some people is very reassuring. - Yes, what could be more human than patients? You know, and by the way, I think it's not one or the other. I think AI will also create more patient doctors. And the reason for that more apathetic doctors is because it's able to do more of the work that they don't want to be doing so that they can do the work that they do want to be doing. You know, we see lots of examples of this. There are parts of every company, many companies that are just sort of emotionally draining for the people that do those jobs. So, you know, let's say you do collections. You work at the bank, you know, and you really aspire to be in sales, but you start in collections. You phone people all day long saying, where's my money? Where's my money? Like that's a pretty tough job to do day in, day out. And now when AI can do that work, you suddenly free all those people up to do work that they feel good about, that they can contribute more to. So I think it's not just about efficiency, it's also about increasing the NPS of the average person's work. And that's happening already. - Yeah, I agree. To your point of, 'cause for a long time, like there was this saying online that was, AI will not replace job, it will replace people who don't use AI, etc. But, and it was kind of like the trend to reuse that sentence, but the reality is like, it's gonna replace jobs. Like, let's face it, like eventually, like some jobs are gonna disappear. Like for jobs, you know, who obviously like sometimes are not like the best job in the world, where people usually are bored by doing them, etc., doing very repetitive tasks, etc. Like, what do you feel is gonna, do you feel like the, yeah, I'm curious, sorry, I'm gonna rephrase, but it's like, do you feel like the, these people are gonna be able to find new job or new activities that are gonna be very interesting for them, or do you feel like the gap between the very rich, we can do like a lot more with AI or whatever, and the four are, the gap is gonna increase. Like, what's the, what's your view on that? - I don't know, I think everybody, okay. So first of all, I think jobs and tasks are not the same thing. Carpathy said this and he's right, which is automation of tasks is not mean automation of jobs, because most jobs involve a set of tasks, maybe that AI can do, but also a ton of judgment, human interaction, I mean, the AI is never gonna take your client to a stake dinner. So there's just so much, there's exception handling. So there's so many aspects to every job that can't be automated, that even if you have task automation, which has been happening, by the way, for a long time, I don't think that means widespread job losses at all. In a two, I think that there's sort of two ways to look at the technology. You could say, okay, there's a 24% efficiency increase, so we'll have 20% less jobs, or you could say 20% efficiency increase means we work four days a week. And so far, the way that the sort of technology is impacting work, it feels more like the latter, more of the four days a week, because we still need you to exercise the judgment, handle exceptions, take the client out for the stake dinner. So I think there's a much more optimistic view than the one that's being discussed in at least the mainstream media. - Yeah, yeah, I agree. And I think like, it's funny for me, because I've been in the sales space for about like 10 years, and I started as, so by trade, I'm a chemical engineer, which has nothing to do, but I started like a sales automation agency. And eventually like every single year, I would see like sales are gonna be replaced by automated like males. Then it was sales gonna be replaced. And every day, like every year, we have like this new thing about a part of the sales rep jobs that's supposed to be automated, dead, et cetera, et cetera. But in the end, I 100% agree, it's just like the job is evolving, it's exciting for people, and you still need that connection. - And by the way, the latest version of that is AI is gonna replace all engineers and programmers, you know? Like no. The whole history of computer science is one in which we increase the level of abstraction. And it's funny because the machine language programmers were all suspicious of the assembly language programmers who were all suspicious of the C programmers, who were very judgmental of the C++ programmers. And Java was a total joke for everyone who knew C++, right, like so on and so forth, that all of the existing folks are looking at people that are coding with cloud code saying, "Well, that's not the real way." I think vibe coding, the kind of vibe prefix has done a disservice to what's really happening, which is just a new abstraction layer. Computer science has never mattered more. - Yeah, no, I agree. And I think like it's, as you mentioned, you know, like back in the days, you know, when you would use, and you would start coding, like whether in C++ or C or even like use other layers, sometimes, you know, you had to go back down to, you know, decompose the code, go to the assembly, if you wanted something to work. And I think it's exactly the same. Like code is just another layer. Sometimes when you're gonna use like a new, I don't know, framework or whatever, you might have to go down to understand what's happening. But it's, yeah, yeah, yeah, yeah, it's very, very true. - Yes, yes, but you're still gonna need engineers, like you need people to understand the framework and how things are built. - Well, I think, and I think you're gonna need more than ever, because I think what's now happening is so much of work. Like there's information ages, industrial age. We say we live in the information age, yet even at Google, how much of your work day to day, if you're a PM, or even if you're an engineer, is information age versus industrial age, right? How much time do you spend in one-on-ones with your manager, writing status reports for your VP, having discussions and debates and disagreements with other people internally. Those are all industrial age tasks. Those are all gonna actually get assisted by software to minimum. So I think we're going to need more software engineers than ever. And look, we can fact check this in a year and see how the field has grown. I'd be struck if it hasn't grown significantly. (laughs) - Yeah, that's, no, that's true. I agree. And when it comes to like, yeah, like engineers and in the startup you invest in, how do you see the shift? Because you talk to many, many founders, what do you feel is the adoption of tools like cursor, of Cloud Code, et cetera, in the company? - It's 100%. - Yeah, I mean, how can you be a founder and not be using these things? Yes, yeah, it's 100%. And I will say the founders these days are they're more technical than the founders from five years ago, which isn't a critique of those founders, but it's a different sort of shape of founder than we were seeing previously. And to just build up on this thing where obviously you mentioned vibe coding. So some people are able to code a lot faster, some apps, et cetera. A use case of Cloud Code typically in our company, it's like our product managers, they are not able to copy paste, like the code base locally. And they can just start coding like proper features that way, they can show it to the dev team, validate the code with them. And if it's fine, they do the code review and then it goes live to production. But we talk a lot about the SaaS era kind of being over. Some people were saying, "Okay, if you can code pretty much anything you want, "you take a notion, it's a doc where you have documents," et cetera, what do you think about? Do you feel like companies are gonna create a lot more tools internally? Or do you feel like that the SaaS era is still there and it's continued to grow and it will still expand over time? - So this is such a great topic. So one, I think that the point about software, SaaS software can be replicated has always been true. You know, five years, forget about AI. Like five years ago, you could recreate notion, you could recreate Salesforce. Maybe it wouldn't be as fast as it would be through Cloud Code, but it's not like those products were, you know, that wasn't self-driving cars five years ago. So why did those product, well like why is the sort of effect you're predicting or potentially predicting not already taken place? It's because those products have a different kind of positive feedback loop and that is reference selling in the enterprise, which is in the enterprise, you wanna buy the one that is the correct one that all your peers are using, that you're not gonna get fired for buying. So there's such a powerful effect of just reference selling in the enterprise and that drives a sort of, compounding network effect like outcome for SaaS companies. That's as true today as it was five years ago. I know with that said, I think there's going to be a lot more software out there and there's gonna be potentially some business model changes, but look, if you have a SaaS product with reference selling in the enterprise or a consumer network product, your modes are as strong as ever. Yeah, and do you feel like, 'cause I was wondering like some enterprise have been known for kind of like building their own tools, but the issue, what I see also, it's like the, the issue is not so much like the building, it's also like the maintaining and all the use case and being able to like update features as you go, because the truth is like, let's say you can build like, because for me, you know, it's even more on economical standpoint. So let's say like a, I use CloudCode and I replicate like a notion, let's take notion, I love notion, but for them. So I replicate like a notion, but now I want to maintain my code base. So if I need an engineer or someone who's like spend time to replicate my code base, it's like, you're yearly salary of someone, you know, that you're spending on just like one or two tools that you could spend like, let's say, even $10,000 per year. It's just doesn't make sense. Yes. Well, also you then like to take the notion case, you have a hundred people in your company using notion, they're using it in different ways. People have built many apps on the notion platform, you know, they're using the notion API programmatically. They're just some, and then the frequency of updates to the existing notion is high. So how do you do the migration? There's just so many issues that are totally independent of writing the code. So I don't know that that much has changed from five years ago, but I know, you know, with this, you see a lot on Twitter about well with CloudCode, it's the other fast. Like, I don't think it is, you know. Yeah, no, I agree, I agree. And talking about like your investment and the company you've invested in, it's funny because I was talking with Harry Stabings like the other day and Harry like missed Dill with Alex and you, my name is to lead like the series A, I think it was in May 2020. Today, you know, like Dill is worth 17 billion. So what did you see, you know, like in the first meeting with Alex? Yeah, there's, I mean, there's two things that I always think about. So the first is sometimes you meet somebody and you've met these people, you are one of these people, you know, which is they give you this sense of inevitability, which means you just, when you meet them, you're like, whatever they say is going to happen, is going to happen, whether I'm a part of it or not. You know, there's this feeling of momentum and hustle and intensity and commitment. And when you see it, you don't see it that often. When you see it, I think you just have to find a way to be a part of it, whether it's as an employee, as an investor, as a co-founder, you know, whatever. So one, Alex and his whole team, Shuo have that in spades, you know, they have this intensity where you just know that what they say is going to happen, is going to happen one way or another. But I think the other thing is that they were not the number one actually by revenue scale in the category at the time. Another company called Papaya was, but the difference for Alex and Shuo was that they were taking this software first, in for a first approach. So many of the other companies, and you know, who have all done a nice job as well, who have not gone to the same scale, they really didn't vertically integrate and build all of the necessary global infrastructure. Part of that is software, part of that is compliance. And Alex and Shuo were focused on that from day one. So the bet at the time was a combination of them, how exceptionally well as founders, but also that the software first approach would win. And that's what we've actually seen happen. I think for many of the companies that took a non-software approach or a light software approach, they've struggled to scale and to get to the kind of adjacent products that you need to to support a multi-billion dollar run rate. Yeah, definitely. And also what I love about, like, what I love about deal, like we've been customers for many years, like almost since inception. So I've been giving feedback from time to time to Alex. But it's, first I think is an amazing founder, because you know, like he still replies to pretty much like everyone on LinkedIn. - It's terrifying. - I don't know how he manages, but you know, I'm really impressed. And also I think like, you know, deal, they also had like a really good timing. Because these companies, you know, would do like, we can help you like basically like hire people in different countries. Like that would only, I mean, the target markets, you know, would be mainly for big corporation, what offices like in many different countries. But with SaaS companies, with tech companies, the market getting bigger and bigger with remote work becoming like kind of like the new norm. We've seen people hiring overseas like more and more. And I think like for a deal, they're also benefiting, you know, from the fact that the market is expanding. What's for you? Right now, a market that is actually expanding thanks to this time, like a technological breakthrough. - Yeah, well, so on deal, yes, absolutely. There's been, and if you look at it, all of the companies in that category benefited from that tailwind, but the difference in execution is who benefited the most and that was deal. So I think that kind of shows you what is the kind of core benefit that everybody gets from just showing up versus what is the kind of incremental benefit from doing all the really compelling things that they've done. And then look, I think there is a secular trend. I don't even know if it's remote work so much as cross-border, payroll and hiring. I just think we are moving to a world that's more software-led, software is default global, right? You know, when you launch Slack, you don't launch it in just America. You're like obviously Slack is gonna be Slack for the world. And as there's more software in the world, there's more global products, more global teams, and deal has done a nice job of being ahead of that. - Yeah, I think that there's a lot of interesting trends that are happening right now. To me, the biggest one is I do believe coding, there's sort of two opportunities in coding. One is that coding and coding agents are upstream of all knowledge work. 'Cause if you think about coding in the narrow way, it's like, hey, it's a way to make software. If you think about it in the broad way, almost any problem can be expressed in the language of code. So there's no reason that like sure, codex is great for engineers, but there's no reason that codex can't also solve problems for analysts and marketers and salespeople. So I think on the B2B side, the idea that coding, which has made tremendous exponential progress in a year, is going to unlock progress in all knowledge work. That's like one trend that I believe is happening. I also think on the consumer side, the idea that consumers can create their own software. Now, trivially, like we've called this the YouTube moment for software, if you think about YouTube 20 years ago, it's like, hey, we have lots of video and lots of television and it's high production quality. And it wasn't clear that we needed more and 20 years later, YouTube's a $550 billion enterprise, it would be one of the biggest companies in the world if it was independent. I think the same being is going to happen for software. People want to make software. And for the first time, they can, they can distribute it and they can consume it. And but sometimes it's going to be important software. Sometimes it's going to be totally trivial. It's going to be software for a bachelor party weekend, software for a joke, software for a prompt. So I think we have this sort of seriousness about software that we had about video and television 20 years ago. And now it's like, no, I just took a video on my phone. It's going to be like, no, I just made an app on my phone, same energy. I've used in any company that actually allow you to create very specific apps, but directly on your phone. Because there's a lot of desktop apps and that are very like SaaS oriented and you create apps. But for phones and applications, I haven't seen that many, but maybe you have. Yes. I'm so glad you asked. Wabi, W-A-B-I, is the company that founders extraordinary. The product is super well done. And in fact, if you add Wabi on X with your app idea, they will make it for you and send it to you. Oh, OK. That's-- I love this good. It's a very simple product. It's like it's prompt to program, prompt to mini-app. And then you can also consume it, share it, it supports multiplayer, all the integrations. Anyway, I don't want to sell too hard, but that is a product. It's a good product. That's your question. Yeah. That's nice. And yeah, I think you were saying also that you were looking at when you invest at like a weird product that works, especially in consumer AI. So what's kind of like the what are like the weirdest products that you've seen and that deliver it actually like a good customer experience? Oh, man. I mean, so many of these products are sort of unusual. Look, I think everything to me companionship is such an interesting category, the kind of AI friends market. And I feel like it's underdeveloped relative to how big it's going to be someday. And look, I actually think that startups are really advantage, because if you're at Google or Apple, like they don't want products that can surprise you in that way. Guess what? Like, even experiences, sometimes uncomfortable, sometimes it's persuasion, disagreement, anger, sexuality. Like these are all parts of people, which means they're all going to be parts of these companion products. And if you're at Google PM or a Google attorney, like, my god, that is the last thing that you want to work on. So I just think this is an area where startups are uniquely advantage. And yeah, some of the products are pretty weird. I think they're going to get weirder. And I think that's a good thing, because guess what? If the internet is taught us nothing else, it's that some people are pretty weird. And maybe we all are in our own unique way. And a lot of what's made it powerful is sort of finding our tribe of fellow weirdos. Yeah, that's interesting. And I don't know if you've rewatched recently like the movie "Her" with Scarlett Johansson. Yeah, I watched it again, you know, like a few months ago. And I was like, that guy, you know, like, if he could have a bet, you know, on polymarket or whatever, like, for a trend, he would have been a billionaire. Yes, yeah, 100%, 100%. And do you see, like, because we talk a bit about, like, companionship, which I think is extremely, like, important, especially, like, I mean, there are, like, a lot of people suffering from depression, loneliness, et cetera, et cetera. But you also have, like, elderly people, you know, like, who basically, like, need assistance, like, the amount of people who just die because they fail and no one came to see them in, like, seven days, 10 days, et cetera. It's just, like, something that definitely can be solved. So how do you see, like, robots and kind of, like, AI working together? And the future of that market? Yeah, I mean, I, so I'm not an expert on robotics. So I haven't studied that very carefully. I think it's going to happen. I will tell you, I think, though, you know, to this whole point of, like, a good question might be why haven't companions worked? Companionship, products work at scale. And then how do we actually get these into the hands of senior citizens? And going back to our earlier conversation, you know, senior citizen, somebody who, you know, our parents, you know, my parents have got older now, like, they've got a big sense of pride. So if you tell them, hey, I'm going to give you an AI friend or an AI worse, like, an AI, like, nurse, like, they don't want that, you know? So there's this whole concept of, I'll add a contextual companion, which is you have to give somebody this plausible deniability, you know, which is why for, you know, senior citizen, they have people come to, you know, come in like play chess with them. Because in their mind, they're like, well, yeah, I'm just playing chess with this person. Of course, it's about connection, friendship, being seen. So I think a lot of this, the trick to getting these technologies deployed into markets like that is going to be creating a kind of pretense. So you don't feel like this weird, sad person who has to have an AI friend. So you're like, no, I just played chess with this dude. And like, you know, we make jokes once in a while, right? Something like that. - So basically, like what you're saying is like, you need to have like, first like a very specific use case to enter like kind of in people's life, to afterwards like build up and that like more feature. - 100%. - Yes. - And what do you think, what do you think that that pretext is going to be? - It'll be different for different people, you know? I mean, this is what is the pretext of any like club or friend group, you know? Some people like to play chess, some people like to talk about World War II, some people like to tell old stories, some people like to hear old stories. So, you know, I think it'll be different for every person. - We're just, we're almost running out of time. So I'm just going to wrap up with the final question. You've been a founder, a general manager of a billion dollar business. Now you're a GP at A16Z. When does it actually cost to win at that level? In time, health, relationships that people, you know, outside like this room and the conversation we're having, don't always understand. - Yeah. I think that's a little bit of a false dichotomy, you know? I actually think that there's this dog chasing the car thing where we talk about retirement and society, like it's some kind of a destination. And then a lot of people get there and they're not very happy. So I think the best kind of work is the work that you can contribute the most to, you can be successful in. And it sort of intertwines with the rest of your life in a natural way. Of course there are trade-offs, right? There's no balancers trade-offs, but the trade-offs can be made in either direction on a given day. So I would kind of come back to what can you contribute the most to that will lead to success in winning. Winning is underrated as something that will actually deliver a lot of sort of value and sustenance in your career in life. And I think pairing that with a really awesome rest of your life is the secret to success. - Really love it. Anish, thanks a lot for your time. Where can people like follow your updates and also your investment to get excited about the new companies? - Yeah, no, no, no, follow me on X, please. It's @ilscience. ILLSCIENCE. It's my DJ name from 40 years ago. So forgive me. - Nice. - You can also get me at Anish at A16Z if you've got a picture or anything else. We read them all. - Awesome. Thanks a lot, Anish. Have a great day. - All right, guys, good to hang with you, man. Take care. Stay safe. (upbeat music) (upbeat music)