The Twenty Minute VC (20VC): Venture Capital | Startup Funding | The Pitch
00:19:38 · Mike Cannon-Brookes
[00:19:38] S0: But I'm not commenting on any of the social issues or other issues in that area. What I'm saying is from technological perspective, for someone to say, I'm not going to use a piece of technology that is open. Right? If I'm saying I'm not gonna use this piece of open source code, I'm like, but I have the code. Well, if I... If you choose not to use a new choice, that's totally fine. That's your choice. But for an enterprise to say, I'm not gonna use a cheaper alternative that'll give me better results when I know what's in it, I think it's gonna be a very hard long term choice. And I suspect people are gonna take those open weight models, which they are already doing, and fine tune and train them and call them their own. Most of the people who are training their own model, large company x claims they built amazing model for purpose y, and it's got some wacky name. We've seen many, many times. Almost every single time, that is... If it's not Nemotron, which it's usually not, it is a Chinese model that they've trained or added their own fine tuning specialization on top of, and then they call it their own model. And I'm like, hang on. That's bullshit. That is the same thing. Right? You've added to it, you've tweaked it. You haven't trained it from scratch.
00:20:37 · Harry Stebbings
[00:20:37] S1: That was actually gonna be my question, which is, you know, we've seen Harvey, and I think it's Ramp, talk about the benefits that they've seen from building their own models. You very clearly said last year that that wasn't a strategic focus for you and that you weren't gonna go down that route. Now with the benefit of hindsight, has your thinking changed when you see all of these companies come out and talk about the benefits of building their own models? Have you been validated? How how has that thought process changed?
00:21:04 · Mike Cannon-Brookes
[00:21:04] S0: It's a great question. The jury is really out on the benefits of building specialized models at the moment. There are definitely cases where specialized models are doing really well. Most of the people building specialized models, I would argue, are not in those cases, though. There are certainly... If you have a lot of proprietary data, if you have very specialized applications, there are definitely approaches to that. Right? So if you're doing, you know, certain robotics, for example, and world model, there's a lot of reasons why people are actually doing it that makes sense to me. There's a lot of others that I don't think will stand the test of time. They may have at the time of shipping a price performance benefit for a specific application. True. Will that price performance hold up over time? I'm not sure it will as generalized models get better and better and better for a lot of applications, firstly. Secondly, a lot of people who say they've trained their own model, I think they're being a little bit fast and loose with words as far as I'm concerned. We could say we've trained a bunch of models at Atlassian. I say that's bullshit. We have a bunch of very specialized open weight models that we fine tune for particular purposes that give great price performance for our customers that I don't claim to be our own model, I would say. Like, that's where I think the claim is often vastly overweighted. Most of these people are taking a Chinese model generally, applying some domain specific intelligence. You see they train them for, like, a million bucks, and I'm like, there's no way you train this for a million bucks. You took someone's billion dollar model and tailored it. If you wanna think about it generically for a particular application, that's totally fine. But that's not really training your own model, I would say. Right? If you're not starting with raw data, if that's their definition, then we do that a bunch in a bunch of different places. But it's not my definition of building a model from scratch.
00:22:47 · Harry Stebbings
[00:22:47] S1: What if it played out, would prove that thesis and your conviction to be wrong?
00:22:53 · Mike Cannon-Brookes
[00:22:53] S0: They could continue to show exponential gains over and above using the latest foundational model on a sort of price per watt basis. So we are seeing many more applications. And it's very hard to answer these questions generically because it depends on the the the the target application actually and the time to answer or or solve a problem or a question, the cost, and the period of time you're in. If you took a... Look at an agentic harness, it is a way of applying a set of prompts and configurations, if you wanna think about it in that way, to a certain type of model and reapplying it many, many times. If wanna you think about it as looping, so that you can get much more value out of a given class of model. And, yes, if you apply better models, you get better results from a lot of those harnesses. And harness design is really, really important nowadays, but it's not model progress. These specialized models have to prove that they can stay ahead of the generified model for a given price performance ratio. I don't think I see many of them doing that. Certainly not the vast majority of them doing that because you would need to have a continued specialized dataset and a problem that doesn't get generified away. Many people who I think had startups, hopefully none of whom you've invested in, that were building their own models two or three years ago, I think now the financial models have just basically smashed what they do. Right? Like, the quality of the generic model has improved far faster than their ability to stay ahead with whatever specialization they were doing. So I think I would need to see the reverse of that.
00:24:19 · Harry Stebbings, Mike Cannon-Brookes
[00:24:19] S1: Lucky, none of my companies have. So great relief. The question that I have on cost is, Mike, it's actually a joke we have with founders that I interview because I interview a lot of, like, very fast growing company CEOs, and they're all private. And I go, how do you think about the cost trade off, frontier versus open, where maybe, you know, there's a little bit of a detriment in performance? And they go, you're kidding me, dude. You've seen how much money I've raised. And they're always like, can you take that bit out? But I give a shit about cost. No. You do. I mean, you're a public... How do you think about that without the luxury of give a shit on private, and I've raised a bunch of money? [00:24:57] S0: I mean, I love all these private companies that have raised a bunch of money and don't have to care about cost because they're not building the discipline to care about cost. Some of them do. But over time, discipline is really important. Anything you do in life, about discipline at some level. Right? [00:25:09] S1: But what if your product is two to 3% off theirs because they have the ill discipline to always do frontier?