Neil Mehta

And Benny is so good at stepping back and reminding everybody, including myself, especially myself, what we're trying to optimize for over the fullness of time. I mean, really, really astonishing. He is the most... clear and concise thinker one could have as a partner. So often I describe a situation that I'm thinking through.

And Benny is so good at stepping back and reminding everybody, including myself, especially myself, what we're trying to optimize for over the fullness of time. I mean, really, really astonishing. He is the most... clear and concise thinker one could have as a partner. So often I describe a situation that I'm thinking through.

We talk about everything at Green Oaks, down to when you walk in, what the lighting is in our office. We are micromanagers to the max. And when we talk about investments, Green Oaks, it's not atypical for us to sit around as a team and talk for three, four hours about a single company. And then we go home, we put our kids to bed, and then I'll call Benny and we'll talk from 9 p.m. to 1 a.m.

We talk about everything at Green Oaks, down to when you walk in, what the lighting is in our office. We are micromanagers to the max. And when we talk about investments, Green Oaks, it's not atypical for us to sit around as a team and talk for three, four hours about a single company. And then we go home, we put our kids to bed, and then I'll call Benny and we'll talk from 9 p.m. to 1 a.m.

in the morning. We almost do that every night. I've talked to Benny already four times this morning.

in the morning. We almost do that every night. I've talked to Benny already four times this morning.

We do do a lot of that. If you just step back for AI, we're investing in a bunch of different companies across a bunch of different industries. What is the common thread? And there's this great, I forget the name of the book, but it was about the Wright brothers figuring out how to get a plane in the air. And there's all these people chasing... trying to be the first one to flight.

We do do a lot of that. If you just step back for AI, we're investing in a bunch of different companies across a bunch of different industries. What is the common thread? And there's this great, I forget the name of the book, but it was about the Wright brothers figuring out how to get a plane in the air. And there's all these people chasing... trying to be the first one to flight.

Of course, you look at a bird and they flap their wings. I got to replicate that. And the Wright brothers are like, wait a second. The laws of aerodynamics are the laws of aerodynamics. You're not going to change the laws of aerodynamics. You just have to figure out how to make fixed wing flight work. And of course, they make it work.

Of course, you look at a bird and they flap their wings. I got to replicate that. And the Wright brothers are like, wait a second. The laws of aerodynamics are the laws of aerodynamics. You're not going to change the laws of aerodynamics. You just have to figure out how to make fixed wing flight work. And of course, they make it work.

And I think business is the same, which is like the laws of great businesses are the laws of great businesses. We know how this works. Did you delight customers? Do you break trade-offs to create something operational or technically really great? Do you have a competitive advantage? Is it a large market?

And I think business is the same, which is like the laws of great businesses are the laws of great businesses. We know how this works. Did you delight customers? Do you break trade-offs to create something operational or technically really great? Do you have a competitive advantage? Is it a large market?

Whenever we've screwed up at Green Oaks, it's usually been because we ignored the laws of crypto. No, no, you don't need a board in crypto. Don't worry about it. No, no, you don't need an auditor. It's not a thing that you need. You don't need to delight customers. It has nothing to do with customers. It has something to do with price movement and fund flows.

Whenever we've screwed up at Green Oaks, it's usually been because we ignored the laws of crypto. No, no, you don't need a board in crypto. Don't worry about it. No, no, you don't need an auditor. It's not a thing that you need. You don't need to delight customers. It has nothing to do with customers. It has something to do with price movement and fund flows.

You know, it makes sense to me, but I guess the laws are different this time. They never are. And so we absolutely look at evals and R1 benchmark and figure out, wow, DeepSeek figured out a way to deploy a model at the 35x reduction for input-output tokens on a comparative basis to OpenAI's reasoning models. That's a pretty impressive feat.

You know, it makes sense to me, but I guess the laws are different this time. They never are. And so we absolutely look at evals and R1 benchmark and figure out, wow, DeepSeek figured out a way to deploy a model at the 35x reduction for input-output tokens on a comparative basis to OpenAI's reasoning models. That's a pretty impressive feat.

What are the takeaways in terms of competitive advantage for OpenAI's model on a comparative basis to others? That's a really interesting question. But what we try not to do at GreenOaks is get excited about that development. and then deploy our time and effort, eventually invest a lot of capital just based on that.

What are the takeaways in terms of competitive advantage for OpenAI's model on a comparative basis to others? That's a really interesting question. But what we try not to do at GreenOaks is get excited about that development. and then deploy our time and effort, eventually invest a lot of capital just based on that.

We try to bring the abstraction level back up to what does this mean for customers? And then we work backwards from that. So when we talk about the model companies, our reaction has been like these large CapEx spends. My feeling has been, and by the way, I've been wrong.

We try to bring the abstraction level back up to what does this mean for customers? And then we work backwards from that. So when we talk about the model companies, our reaction has been like these large CapEx spends. My feeling has been, and by the way, I've been wrong.