Sam Altman: Proof of Human in the Age of AI

Sam Altman (World/OpenAI) joins Cosmo Jiang (Pantera Capital) to discuss why proof of human is the necessary complement to powerful AI. 

Then Mason Nystrom turns the mic on Cosmo to break down the investment case for World.

The problem: we are rapidly entering a world of infinite AI-generated content and infinite AI identities. 

In that world, knowing what is truly human becomes one of the most valuable primitives on the internet. World is building the cryptographically secure, verifiable proof of human layer to make that possible.

Key Topics:

- Why Sam Altman believes World is especially urgent now: AI is advancing faster than expected and proof of human has never been more necessary

- AI as technology of abundance, World as technology of scarcity: why the two are complementary not competitive

- Pantera's investment thesis: blockchain as the right infrastructure for cryptographically secure, sybil-resistant proof of human

- Agents need to move money: why agentic payments are one of the things Sam is most personally excited about

- Proof of human as a substrate for layered credentials: driver's licenses, education, employment, all appended to a verified human identity

- Advertising as the first massive use case: a half trillion dollar industry that will pay a significant premium for verified human impressions

- Dating apps, government services, and enterprise partnerships: Tinder, DocuSign, Zoom, Okta, and Versa all announcing how they will work with World ID

- Human-directed agents: why proof of human accelerates agentic commerce rather than blocking it

- A potential $50 billion free cash flow protocol if verified humanness generates even $5 to $10 per person per year

01:27 Sam Altman Keynote at World’s Lift-Off Event
02:12 Why World Was Built
03:48 Blockchain x AI
04:08 World ID Use Cases
04:47 World's Founding Vision
05:10 World ID Trajectory
05:50 Why Pantera Invested in World
06:45 Sam Altman Saw This Coming
07:10 Proof of Human at Scale
10:12 The Value of Digital Identity
11:11 Agents Need a Human in the Loop
11:42 Scaling to Billions of Users
12:53 Enterprise Partnerships

The views expressed in the podcast are those of the individual personnel quoted and are not the views of Pantera Capital Partners LP or its affiliates ("Pantera"). The podcast is provided for informational purposes only to provide market commentary and for general educational purposes, and should not be relied upon as legal, business, investment, or tax advice. The podcast is not directed at nor intended for use by any investors or prospective investors and may not under any circumstances be relied upon when making a decision to invest. Please see additional important disclosures related to the content discussed in the podcast here.

Transcript

SPEAKER_00 0:00

AI has happened so much faster than I thought it was going to by now that the need for something like World feels especially strong. Clearly, the world is getting close to very powerful AI. But I'm sure many of you have started to have an experience more and more where you're like, Am I am I interacting with an AI or a person or how much of each and how do I know? I I had this like weird experience recently where I realized that using like Reddit or something like that, I couldn't tell anymore which comments were real and which were written by AI. I would love to be able to use a product like Reddit and really know who the people are. World is now on the way to being a real human network for the internet. We can increasingly tell, and we're seeing this now and now with what you all and others are building. People are really going to need this.

SPEAKER_02 0:42

Pantera Capital joined World for its liftoff event last week. We believe World is at an inflection point in its story as people begin to truly appreciate how important this technology is. We had the chance to sit down with Sam Altman for a conversation about the intersection of blockchain and AI. Later in the video, Mason Nicefront and I sit down to discuss some of the compelling use cases that led Pantera to invest in World and why we believe proof of human is so important.

SPEAKER_00 1:07

Clearly, the world is getting close to very powerful AI. And this is doing a lot of wonderful things. We're gonna see this play out in all of the ways we dreamed, many we didn't. But we are also heading to a world now where there's going to be more stuff generated by AI than by humans. And I think there's a ton of reason for optimism for this new world. Clearly we care more about humans than we care about AI. But I'm sure many of you have started to have an experience more and more uh where you're like, am I am I interacting with an AI or a person or how much of each and how do I know? And we are so hardwired to care about people uh that I'm not afraid for the future as long as we can tell. So years ago we started thinking about what we would need to have proof of human in a world with advanced AI in it, what it was going to be like when the internet had way more AI uh agents, you know, writing comments or making posts or on social networks or interacting on commerce uh flows or whatever else than it did people. And we really wanted to make sure that the world continued to be about people, for people, that people were still the primary thing that we all care about, which we knew was going to happen no matter what else happened with AI. We are so hardwired for that. We just wanted to figure out how to make it easier. Uh Alex and I and the team started thinking about ideas for how this could happen. Uh, and then AI has continued to go like this. And the world is now on the way to being a real human network for the internet. I don't think we know yet exactly all the ways that's going to be important, how that's gonna work, what people are going to do with that, but we can increasingly tell, and we're seeing this now and now with what you all and others are building. Um, people are really going to need this and really going to need to do this in a privacy-preserving way. So world ID is our effort at that, and it's been amazing to see the progress over the last year as people have adopted this and figured out how to integrate this into a new world of AI. And our goal is to make World ID be this new proof of human for the internet.

SPEAKER_02 3:13

Sam, thanks for joining us. Absolutely. We're excited to be investors in world and think you guys are doing really great things. There's always been a lot of advancements in AI over the last few years. And that is crowded out a little bit, all the fundamental developments we are seeing elsewhere. How's your view of how blockchain has evolved and how maybe that intersects with AI?

SPEAKER_00 3:31

I don't view it as crowding out. I think AI will like lift everything up. Uh, I don't know what all the most important pieces will be, but clearly agents are going to need to be able to pay each other and move money around in all sorts of ways. And that's one of the things that I'm personally most excited about. Obviously, everything we're doing at World.

SPEAKER_02 3:45

What are the use cases that you are most excited about World ID, like seeing World I being used in? And like where could it be the most impactful? I mean, you mentioned Reddit, that's a great example.

SPEAKER_00 3:54

I am most excited to see the kind of creative energy unfurl. I I don't know, like I have my own set list of things like that. I mentioned one. Uh clearly there's a lot of stuff around payments and how access to AI compute may go, but I'm most excited to just see what the creative world thinks of here.

SPEAKER_02 4:12

I feel like this concept of identity has become much more uh clear now than it was certainly many years ago when you guys first ideated this. How has that view of like world what world idea is trying to accomplish changed or maybe it hasn't changed?

SPEAKER_00 4:24

I think a lot of the core ideas have stayed exactly the same, but the exact ways in which this is going to be important, like you know, eight years ago we didn't know language models were gonna be the thing, for example. We knew that somehow you want to be able to like identify people um and that uh there was gonna be a lot of AI and there wasn't a lot of people in the world, but the ex the specifics of like where you're gonna most feel it breaking in the current system, that wasn't so obvious.

SPEAKER_02 4:48

Like, how fast do you think world editing can really catch on? It does feel like it is the moment for something like identity. A lot of people are talking about it. Just how fast and how large of a scale do you think this could be?

SPEAKER_00 4:57

I think the scale can be absolutely huge. How how fast, you know, I I assume it will continue to scale over a number of years to get to like a real significant fraction of the internet. Um, these things always almost always take longer than you think, but the the trajectory of it seems really great.

SPEAKER_02 5:11

Well, really excited to be part of this. I'm really excited about what we're trying to accomplish at World. And so thank you for for the time and the work. Really happy we could talk. Yeah, thanks, Ed.

SPEAKER_01 5:20

We've heard a lot about the builder perspective uh of world, but I think it'd be great to touch on the investor perspective. Uh Pantera, we are investors in Worlds. Uh maybe frame up why we're excited about World and you know what led, what calies that initial investment.

SPEAKER_02 5:36

Well, I I think let's start big picture and and and think about what the problem really is, problem state really is. We are rapidly entering a world where there's going to be infinite AI-generated content, infinite AI identities in a world that is increasingly digital. And in that world, we will want to know what is truly human or not. And the only way to do that is with some sort of cryptographic, uh cryptographically secure, verifiable, censorship-resistant layer called and proof of human. And it turns out that blockchain is actually a really good way to do that. Uh, and so that's why proof of human exists, and that's why world exists. And um if you know, this sounded kind of crazy a few years ago when they I mean, when they started eight years ago, it was truly crazy. Like no one could really see that happening. But Set Man Altman had the vision that he saw what open AI was probably building towards, and that actually proof of human is really necessary as a complementary technology to that.

SPEAKER_01 6:32

Right. AI is this technology of abundance, as you mentioned, and world provides this element of scarcity to actually verify human identity, this idea of proof of personhood. I think it's maybe uh sometimes hard to understand like how many ways that can impact the world. Uh talk about some of the use cases that really shows like the scale of how something like proof of personhood can be useful.

SPEAKER_02 6:54

So I'm a you know consumer internet tech investor by training. And so the the biggest ham that everyone's always talking about is advertising. All right. Um advertising what's driving driven like tremendous growth out of Google, Facebook, now Meta, and now you know Amazon and even OpenAI, really that's that's what's going to drive a lot of growth. Um, because you know, advertising is effectively in many ways, you can think of it as 20% of GDP. You know, uh retailers or you know, businesses are willing to spend up to 20% of revenue to get customers. So that's advertising. Uh today it's about uh, you know, the measurable advertising, digital advertising, a half trillion dollar industry. And when I think about what, you know, what the details of what advertisers buy is they buy impressions, right? They're trying to get impressions that are valuable that hopefully convert into some action they care about, which is you know, purchasing something. And um, advertising will pay willing to pay a lot for better targeting. And so you can imagine a world where if there's infinite AI bots, man, like crawling your website, like you don't want to pay for an impression that's going to a bot. You'd want to pay for an impression that's actually viewed by a human. And so you can imagine a CPM, you know, CPMs are something like 15, 20 bucks on the open web. I don't know, probably could be 30 bucks, 40 bucks if we knew it was a human, 50 bucks, right? And so you you immediately realize that wow, like humor for human could be applied towards one of the largest industries, it's half trillion dollar industry, and maybe it's five, 10% capture of that. You know, there are other use cases that are uh that haven't historically been monetized, but may become new use cases. You know, if you think about it from uh from dating apps, right? There uh the uh there's a dating app today that or dating app called Match.com, which is the largest runner uh runner of dating apps. Um they announced last year their first partnership with OpenAI around Tinder. Uh and uh in order so that you know kids can't get catfished, right? And that's you know, it's like kind of funny, haha, but actually it's a really serious issue. And um, it's actually really valuable, especially because people pay on these dating apps to meet other people. And so why not pay to make sure you're meeting a real human? Uh other use cases like government services, you can imagine, you know, in the future, most of our government services are probably going to be digital. That's probably even how we get distributed uh, you know, social security or other insurance benefits. And they need to be able to verify that they're not getting they're they're they're not getting the government's not being scanned to deliver it to real verifiable humans. And so there are just so many surface areas that this primitive um clearly addresses, but also potentially opens up.

SPEAKER_01 9:19

Right. And this primitive also has the potential to evolve. When we think about what makes you a digital human today, it's maybe your username, it's your email, it's your phone number, maybe like your government ID. But if you have this actual, you know, verified you human identity, you can start to tie it to other credentials. And so maybe as you think of kind of the grandest framing and scale, this can be, you know, a proof of person for for the world, uh, but that can expand to other types of credentials or ways that you know we think of ourselves as human.

SPEAKER_02 9:50

Oh, completely. I think uh the uh this this concept of identity is just becoming more and more important. And there are many ways to layer on different types of identity, right? And so proof of human is the is is is a great substrate on which you can apply other forms of verification or licensing. Like you will want to know that I have a driver's license, maybe, but maybe you can append that to your your proof of human. You will want to know that you're you know, some some employers will want to know that you're educated in XYZ place. Maybe you can append that credential to your proof of human. That way you know that you know this human is it, there is a real human. He's authenticated that he's authenticating that he indeed went to these schools. And you can verify that with, you know, with the school. And and so uh there they're just there there are clearly a lot of ways that we've thought about that that we've taken identity for granted in the physical world, right? Because we we see people, we we we speak to them in person. As we move more digital, this concept of identity and the value of a digital entity is only going to increase.

SPEAKER_01 10:48

Right. And same as we move towards a world where we're using agents on a more daily basis, the idea that there's actually even a human behind that agent becomes increasingly valuable.

SPEAKER_02 10:56

That's right. That's right. And by the way, that's it is not to stop agenti commerce in any way. It's in fact it really accelerates it because this way you know that, you know, an agent is truly being is tied to an authenticated human and know that you know that is something that well, there are some cases where you're okay with bots hanging around, but there are some cases where you want to know there's a human in the loop and that a human is directing the agent to do something, you can you can do that. Right.

SPEAKER_01 11:20

And maybe the last thing to talk about is scale. So world is being used by uh or it has 18 million verified humans today. How do you think of the potential scale this type of human network can get to and on what timeline?

SPEAKER_02 11:33

Oh, yeah. Well, I mean, there are 8 billion people in the world, and you know, about 6 billion of them are online in some form or fashion. You know, something like 4 billion of them have a smartphone. And so the the immediately addressable TAM is the 4 billion people in the world that have access to a smartphone. Um, I think you could see a world where, you know, today it's it's around 20 million users, so it's early days, but we could really see that scale to billions of people over time. And you can see why it'd be valuable for each person, you know, on the order of five, ten bucks a person in value a year is maybe attached to this ID. This could become a$50 billion free cash flow protocol because of how valuable, because of the revenue that it can generate. And that is a very, very valuable system.

SPEAKER_01 12:14

Yeah, absolutely. We're uh going to find out very quickly how the value uh of a human is going to accrue.

SPEAKER_02 12:21

Oh, I just think we're the time is now. Like we are such an inflection point in the recognition that we need digital identity and digital solutions. I mean, you're seeing every AI leader talk about it. And today on stage, we saw all these major enterprises come out and announce partnerships, starting with Tinder, DocuSign, Zoom, Okta, Vercel. All these companies realize they want to integrate with World ID. So we're seeing adoption come to life now.

SPEAKER_01 12:44

Cosmo, thank you for highlighting the investment case for world. Thanks for having me.