Adrian Aoun on healthcare as a product, Apple's AI problem, and why you should have kids now

**Auren Hoffman** (0:00)
Hello, data nerds. Welcome to Summation. My guest today is Adrian Aoun. Adrian helped founded Torch, a health data platform that was acquired by OpenAI earlier this year. He's also the co-founder of Seneca, which raised $60 million to build autonomous drones for wildfire suppression. Adrian previously founded Forward Health, and before that, he ran special projects for Larry Page at Google, where he founded Sidewalk Labs and helped build Google's AI division. Adrian, welcome to Summation.

**Adrian Aoun** (0:29)
Well, thank you for having me. I'm still getting used to this new Summation title, but I've been an avid listener since the very beginning, so I'm excited to be on.

**Auren Hoffman** (0:37)
You have a thing where you ranked all society's problems against this set of criteria. What is at the top of the list right now, and how has that framework changed?

**Adrian Aoun** (0:47)
Yeah. One of the things that we had to do, I spent a few years working as Larry Page's right-hand man, and we created Alphabet, and then my focus was like, let's go create a bunch of alphabet companies. We had this weird thing which is like, well, if that's your focus, where do you even start? What we did is we tried to say, what are the things that could be most positive for humanity multiplied by where we have some competitive advantage? We started really naively, and we just said, let's just take cuts of GDP. People have all these ontologies, construction is this and education is this. The problem of that is that it doesn't quite give you all the cuts that you want.
You're never going to say, oh, I should go work on AI because that's not one of the industries 20 years ago or whatnot. What's weird about a lot of technology is that it often starts as a vertical and then becomes a horizontal. Like AI 10 or 20 years ago. I started a company that Google bought in AI, and we were considered in the AI sector. Now saying something's in the AI sector is almost weird. You're like, isn't everything just AI, right? What we ended up doing is as we kind of went down this path, we looked actually a lot at what the nonprofits do. So think of like, if you've ever come across players, obviously like the Gates Foundation, et cetera, they get together at this cool conference called the Copenhagen Consensus. They get together and they try and decide like, what's better for humanity, right? Let's say you're in one bucket, for the same price you could cure all of cancer, or you could get internet to the next billion people. Like, which would you rather do? Turns out, actually the one that's more impactful, as crazy as it sounds, is getting internet to a billion people is more impactful than cancer. I think that when I looked at the numbers...

**Auren Hoffman** (2:31)
Because cancer will give people on average of three more years or something.

**Adrian Aoun** (2:34)
Exactly, that's the number. It was three more years, right? Whereas lifting a billion people out of poverty, which is in essence what giving them internet access does, is much, much, much broader. And so what we tried to do was we tried to take their ontology, which is what are the biggest problems in the world, not what are the biggest kind of existing industries in the world. Now, there's no perfect ontology here, but the idea is we were trying to get things like, well, can people create economic value for themselves? Can people build shelter and find spouses and have children and all these sorts of problems and get them near the top? And then we were trying to get towards the bottom things like, do you have an entertaining movie on a Saturday night? Which is important, but nowhere near as important as living longer. And so you can kind of think of it almost as we were trying to encode Maslow's hierarchy of needs in some form of mechanism.

**Auren Hoffman** (3:29)
But if you're going to do that, you're basically not helping Americans, essentially, because they're kind of at the top of the Maslow hierarchy.

**Adrian Aoun** (3:36)
Well, yeah, so that's interesting. So you're right, you would immediately go there. But then where you also go is you need some place where we have a competitive advantage to work on it. And so one of the reasons why most technology goes with the kind of high-end disruption that trickles down, right? It's like we build the iPhone, and now in India, you can buy a smartphone for $5, is because it's a hell of a lot easier to fund building an iPhone and getting it to scale in the United States, right, than in India. And obviously, that's where we kind of spent the majority of our the majority of our effort and time.