Sonya Huang on AI's 'not your weights, not your product' moment and why application companies are building their own labs

Speaker 1: So Yep. Yeah. That's accurate. Well, have a great rest of your day.

Speaker 2: Good to see you, Garret.

Speaker 1: You so much for the update and explaining everything. We'll talk to you soon. Cheers. Have a good one. We've been keeping Sonya Huang from Sequoia Capital waiting too long. She's a general partner. She's been on the show before.

Speaker 2: To just talk about Let's bring It'd be great to just talk about venture capital for a little bit. Back

Speaker 1: to AI. Back to 's it going? How's We

Speaker 8: haven't seen each other in a while. Congrats the acquisition.

Speaker 1: Far too long. You. Thank you. We're time. Yeah. So temperature check. What's going on with this AI thing? Is there anything there?

Speaker 8: Temperature check. Oh my gosh. There is absolutely something there. The numbers we're seeing from these companies Yeah. Never seen anything like it before.

Speaker 1: Interestingly, less of a power I mean there's power law companies, Anthropic OpenAI. There's lot of companies that are doing really great. But then there's also diffuse community of like NeoLabs and and NeoClouds and so many different application layer stuff where you're seeing just really solid business fundamentals that previously would take a decade to build up. You're seeing it in two to three years.

Speaker 8: Totally. The anthra like, Anthropic and OpenAI and x AI are growing at a pace that nobody has ever seen before. Yeah. But even if you remove them Yeah. This next cohort of companies Yeah. Open Evidence, Glean, Factory, they're growing at rates that we've just never seen before. Yeah. One of the most interesting things that's happening right now is we used to have this like separation in our heads of there's the foundation model companies and then there's the application companies. Companies. Mhmm. And I would say like one of the most interesting things that's happening now is that all the application companies are starting to build their own research capabilities, their own labs. Yeah. And that's kind of like this concept of democratized intelligence. Yeah. So, would say it's not only revenue that's not just accruing on only the first at the top two players. It's it's the production of intelligence itself seems like it's very much democratizing.

Speaker 2: Yeah. We were I don't know when Dylan was on. Maybe it was last week? Mhmm. But Dylan from Figma Yeah. We were talking to him. Was like, I want Figma to work on the problem of Basically, Slop is like has been this like, you know, it's been Slop has been getting better and better. Mhmm. But, you know, it's basically like you have a a new breakthrough and then for two months it's like, wow, it's so good now. And then you realize that it can only do like one style over and over and over. And I feel like, there's all these application layer companies that are in such a great position to work on some of these like fundamental problems that for better or worse, like the labs are not able to focus enough on. Right? Like maybe it's like only a billion dollar revenue opportunity. Right? And so if you're a lab and you're adding billions of dollars of revenue a month with your core business, why would you work on a problem like that? And so I think there's so many examples of that from from Figma to some of the other ones you mentioned.

Speaker 8: Mhmm. Totally. And I think if you look at it from the perspective of the startups, there's just this huge wave towards companies wanting to own their intelligence. And I think a smaller set of companies has been beating this drum for a long time. Like I'm on the board of Fireworks. Their tagline is own your intelligence. So they've been advertising this for a long time. Obviously, cursor went on the journey two years ago of starting to train their own models. But what's happening now is like both the giants and the ecosystem are starting to speak up and then the startups are actually getting extremely good at building their own research. So in terms of giants, you have people like Alex Karp talking about sovereign intelligence. He he has this phrase, own the means of production, which I freaking love. Saacia talking about your proprietary data. Jensen, championing open weight. So you have all these giants of the ecosystem speaking up of like, hey guys, you should own your intelligence. And then if you look at it from the perspective of the little guys, the startups who we are in the business of backing. Yeah. A couple of years ago, they were primarily looking at you know moving some of their intelligence towards these open weight models primarily as a cost rationalization exercise. The thing that's different now is like it is an existential and strategic imperative for them. And so, there's this phrase, you guys probably remember this from the crypto days, not your keys, not

Speaker 7: your crypto.

Speaker 8: Yeah. This idea of like if somebody else is custodying your weights for you, I don't are you like custodying your keys for you? It's not yours. Because something could happen to that brokerage. Something could happen there. And I I think like I I think the AI version of this meme is not your weights, your products. Sure. Because fundamentally, if you don't own the weights, if you're just making an API call, you don't have ownership, you don't have steerability, you don't the data file doesn't accrue to you. And so like people are kind of waking up to this. Yeah. And so if if your use case is like you want a coding agent, like for that, I'd say

Speaker 7: Yeah.

Speaker 8: Closed model, closed agents, that's fantastic.

Speaker 7: Yep.

Speaker 8: If it is like your core product, I think companies are increasingly waking up to like Yeah. Not my ways, not

Speaker 7: my products.

Speaker 1: Sell to your customers. It should Exactly. That makes a lot sense. Founder office hours coach me through. You're on the board of my hypothetical company. Software company, I'm using AI, 500,000,000 ARR, let's say. And I buy this thesis and I come to you Pretty

Speaker 2: confident example there, John.

Speaker 8: 500. Not bad.

Speaker 2: There we

Speaker 1: go. We're cooking. We're cooking. No. No. No. I mean, like, we have we have fully made it through. We're, you know, an AI winner. We're accelerating, growing. Yeah. Growth is great. But I come to you and I say, okay, I'm I'm all in. We're going to own our intelligence stack. Is there a moment where we need to have a conversation about what the talent budget will be? I mean, we saw these crazy MSL deals. And for a lot of these companies, they might be unicorns. They might even be decacorns. But they're not going to be able to staff a team of AI researchers and build a full neo lab doing like next generation fundamental research. Yeah. What does the team build out actually look like? Is it enough to take your your best software engineers and have them use coding agents to fine tune models for for your team? Or are you hiring entirely new disciplines? What is the what is the correct shape of Totally. An internal AI lab at like a successful scaled unicorn, decacorn software company?

Speaker 8: Yeah. So typically, you're not going to be pre training your own models. There are specific use cases where you actually need pre trained models. That's totally different thing.

Speaker 1: Okay.

Speaker 8: If what you are doing is trying to take off the shelf open weight models and then adapt them and make them really, really excellent for your domain, that's a much larger talent pool. And what we've seen is there's actually two flavors of talent that are really good at this. One is people that have done post training before.

Speaker 1: Okay.

Speaker 8: So, the post training teams at the labs are very, very large at this point.

Speaker 2: Sure.

Speaker 8: There's plenty of these people floating around. And then the second profile is actually interesting is just like engineer or I guess the like just generally smart person.

Speaker 1: Smart, difficult

Speaker 8: person. Pokemon. Because this stuff is actually not that hard and like part of the reason it's even possible for all these companies to have their own labs now is because the actual post training stack has matured.

Speaker 1: Sure.

Speaker 8: So it

Speaker 3: used to

Speaker 8: be that only OpenAI had and Anthropic had the infrastructure in house to be able to do things like post training, reinforcement learning especially. Yeah. But now you have companies like Fireworks that gives you the post training infrastructure. You have LangTrain that gives you the evals. You have Trajectory that helps with the continual learning. You have you have Mercur that helps you with the data factory stuff. And so like all these all these components now exist. And so if you as a generally smart person see the see the menu of opportunities, you can actually cobble together your own research stack in a way that wasn't possible a couple years ago. And so like to give you a sense, very, very small teams can get very far. The Harvey team has put out I think pretty extraordinary research. Yeah. They're they're they've they just found an entire RL environment last week. Their benchmark is state of the art for legal. Yeah. And their entire research team is seven people. So you can get very, very far. We're definitely not, you know, you're not competing with with Meta or OpenAI for size of talent budget here.

Speaker 1: What about one click down? I don't want to invest in building the team owning the stack entirely. There are neo labs that will show up and fine tune a model for me. Is that is that the domain of growth stage startups or is that product going to be more consumed by enterprises that maybe don't have the DNA to just move a bunch of amazing engineers over to do a post training stack and spin up and roll their own? But the NeoLab offers I'm thinking of like a thinking machines, a Tinker, like what happened with Ray Dalio's fund and and and how they were able to fine tune a model, get really good results. That feels like its own new market of like post training as a service, fine tuning as a But how does that how does that piece in? Is there a world where you don't necessarily have a relationship with Merkor but your third party does?

Speaker 8: Yep. Look, this is a this is a huge ecosystem and a huge market because I think everyone sees the opportunity of, you know, you obviously the closed model inference market will always be gigantic. Yeah. But I think the open model inference market is becoming very large And for companies to actually be able to make use of open weight models Mhmm. There is a maturation process. There is a hand holding process. There is an entire like, you know, come work with us. We will forward deploy people onto your staff Sure. To help you go on that journey. And so there's there's lots of options. Fireworks has a fantastic team for this. Mhmm. Merkor has a fantastic team for this. It kind of depends on the specific problem you have. Okay. So the specific problem you have is, hey, I I really want to do reinforcement learning on my online data. Fireworks is fantastic for this. If you're like, hey, I can't train on my customer data. I need to get my my model really really good for this specific domain. Can we create a bunch of synthetic data around this opportunity? Mercor is fantastic for it. Oh, So, kind depends so on the use case. Mhmm. Generally, I think that it's like really important. It's an and, like companies need to have extremely smart people in charge of this in house. This can't be something you outsource like your lunch menu outsourcing, right? This is so core and so you need to have smart people in charge deciding on your strategy, deciding on your technical roadmap, and then making judgment calls of which parts of the stack you want to outsource to others, which parts of the stack you want to lean on others for. Harvey's actually done a really good job of this. They lean on pretty much all of the five or six in your labs I just mentioned as their partners. But you need to intentionally own it and over time, just like we saw Cursor go on this journey, I think a lot of application companies will go on the same journey that Cursor did. Over time, you become more and more incompetent more competent in house and you bring more of that expertise in house.

Speaker 2: Sure. Sure. How are you processing the the current price war? Mhmm. You have sort of the lagging labs starting to compete more on price. Mhmm. You have leading labs trying to make sure that that they have competitive, you know, models at every Yeah. Part of the curve. But yeah. Anyways, what what's your take and and where does this go?

Speaker 8: Look, think Jevan's Paradox, not to be an annoying VC, but Jevan's Paradox is like freaking wonderful thing. Because what's happening is like, I see the data from all this stuff. Right? From oh, there we go. Bam. There we go. Look, Our companies, we see the margins going, their gross margins going up because at the same time as like AI usage is going way up, but their gross margins are also going up because they are the beneficiaries of intelligence getting cheaper and cheaper to meter. On the other hand, I see this from fireworks, see this from the model companies we're in business with, their cohorts are getting better and better and better. So it's not like they're facing price competition and their businesses are going into the gutter. They have phenomenal businesses where because they're able to provide intelligence at increasingly cheap prices, their businesses actually get better and better. So I actually think this is an everybody win situation.

Speaker 2: Yeah. Part of the part of the challenge with with I feel like acts trying to process the the price wars and and how open source is fitting into all of this is that when you have Anthropic and OpenAI still as private companies, people just don't have the visibility and they don't realize that even with great open source models and even pricing price cuts across different providers, you're still seeing that just massively accelerating revenue. And it's not like it's not like the biggest customers aren't aware that open source exists and it's good and they are using it in a bunch of different ways. So I think I think Yeah. It'll be very Yeah. It's an and. Yeah. And think it'll be very helpful when when more of these players are public just so that everybody has the same access to information.

Speaker 8: Yeah. Totally agree.

Speaker 1: Thank you so much for coming on the show.

Speaker 3: Let's do it

Speaker 8: again soon. Done. Good to see you guys.

Speaker 2: Yeah. Again soon. Great to see you.

Speaker 1: We'll talk to

Speaker 8: you soon.

Speaker 1: Goodbye. Let me tell you about public.com. Investing for those who take it seriously. They got stocks, options, bonds, crypto, treasuries