Pod Insights Equity Podcast Sam Altman’s eye-scanning startup wants to prove humanity in the age of AI bots About this episode Ever wonder if you’re talking to a real person online or just another bot? As bots increasingly outnumber humans online, leading to an explosion of deepfakes and AI-driven fraud, one company has a solution straight out of sci-fi: scanning your iris to verify your identity.
Today on TechCrunch's Equity podcast, Rebecca Bellan spoke with Adrian Ludwig, Chief Security Officer and Chief Architect at Tools for Humanity, the company behind World’s eye-scanning Orbs appearing around the globe. Bellan and Ludwig discuss building privacy-first identity verification, the open-source approach to biometric tech, and why proving humanity matters now more than ever.
Listen to the full episode to hear:
How zero-knowledge proofs verify that you're over 18 (or human) without revealing your location, browsing history, or other identifying information.
Why Tools for Humanity open-sourced the entire Orb — from hardware to firmware and software.
How Match Group, Event Pop, and other major platforms are partnering up to combat bot abuse.
Subscribe to Equity on Apple Podcasts, Overcast, Spotify and all the casts. You also can follow Equity on X and Threads, at @EquityPod.
Learn more about your ad choices. Visit megaphone.fm/adchoices
Transcript 0:00 Speaker 00 This episode of Equity is brought to you by Schneider Electric. Schneider Electric, advancing energy tech.
0:05 Speaker 01 Hello and welcome back to Equity, TechCrunch's flagship podcast about the business of startups. I'm Rebecca Beland, and this is the episode where we bring on industry experts to help us explore a trend in the tech world and dive deep. I don't know about you all, but lately, whenever I interact with someone online, I'm wondering, is this even a real person or is it a bot? Some people call this the dead internet theory, the idea that most of the web is already controlled by automation and algorithms. As bots increasingly outnumber humans online, leading to an explosion of deepfakes and AI-driven fraud, how can we prove humanity? One company is trying to solve this in a way that's straight out of sci-fi, scanning your iris to verify your identity. Joining us today is Adrian Ludwig, Chief Security Officer and Chief Architect at Tools for Humanity,
0:49 Speaker 01 the company behind World and those eye-scanning orbs that you may have started noticing appearing around the globe.
1:01 Speaker 01 Adrian, welcome to the show.
1:03 Speaker 02 Thanks, happy to be here.
1:04 Speaker 01 So you have a storied history. I kind of like that this company, you've got a dual role of Chief Security and Chief Architect, because that tells me how deeply embedded security is into the product. And you have nearly 30 years in computer security and building internet services. Can you just give us a little bit of your background? I know you were at Android for a hot minute.
1:25 Speaker 02 Yeah, a very hot minute. Pretty early on, actually, I was reached out and began working on Android when we had just a couple of devices and maybe a few million users. And by the time I left, we had over 2 billion users and hundreds of devices and manufacturers around the world. I worked on building Android security. I also worked at Adobe for several years, building out the security capabilities there for the web. And then at a company called Atlassian, where we were trying to help enterprises transition to cloud, building that security. So it's been quite a journey, that's for sure.
1:53 Speaker 01 Yeah, cool. Before we dive further in, where are you, because you sang at Lassie and made me think of Australia, where are you calling in from today?
2:00 Speaker 02 I'm in San Francisco today, but certainly with World, we are all over the place. So it's been a year of, I've been in a lot of time in airports and traveling around the world to meet with our customers and users.
2:10 Speaker 01 Yeah, so probably a lot of time dealing with those, what is it called at the airport security where they make you, where you can like scan your face to blitz through?
2:17 Speaker 02 You know, at this point, they scan your face at the airport. I was in Universal Studios in Japan not too long ago. You go through a biometric check there when you're going in. I think these things that you were referring to as sci-fi earlier, that's just day to day at this point in the way that people are interacting with the world and technology.
2:34 Speaker 01 Okay, so Tools for Humanity is building World, which is a project within Tools for Humanity. And for those who don't know, Tools for Humanity is co-founded by Alex Blanier and Sam Altman. So they built the tech and infrastructure. You have nearly 17,000 active orbs around the globe right now. Can you give us again, a little bit of a background about what the orb actually is and how it solves this problem of trying to verify humanity in this age of AI bots everywhere?
3:02 Speaker 02 Yeah, I mean, the core observation that was made at the beginning of the company was if you rely only on online signals, then AI is getting to a point where it can replicate everything that a real person would do. It can look like you, it can sound like you, it can, I don't want to use the word think because it's not quite what's happening, but it can appear to think like you. And so all of the ways that we have to evaluate whether someone is a real person, if you only use online information, they break down with the sophistication that we're seeing in AI. And so the observation was you need to take something from the real world to ground those interactions in the digital world in reality.
3:40 Speaker 02 And so the orb is a camera. It's a camera that takes pictures in visible light of your face. It takes pictures in infrared also of your face and of your eye. And then it uses that to understand whether this is a real person and then uses that to give them an identity that they can use in a private way and interacting online. So you can use your social network or your game and you can know that that is a person behind the character that you're interacting with or the identity that you're interacting with.
4:08 Speaker 01 Okay, so who are some of your current customers that are using this technology and implementing? And I know we've heard rumors that Reddit might be a customer. You're welcome to confirm or deny on the record, but who else?
4:19 Speaker 02 I mean, we're definitely talking with all of the big players. A couple of years ago, when we would go talk to a large social network, they would say, hmm, not a problem that I'm worried about.
4:27 Speaker 01 They're like, we like the bots. It makes us look like, it makes us look like we've got more engagement than we do.
4:32 Speaker 02 At this point, they're all trying to figure out how to navigate this. I would say almost any of the big names that you could think of are ones that we're having conversations with. There are several that we're at a point where we're doing experiments with and beginning rollout. Some of the biggest are not in the pure play, social networking. They're more focused on a specific use case. So Match Group, which makes Tinder is one that we have a partnership with and are in the process of prototyping and then rolling out production in the Tinder product, which is an online dating application.
5:00 Speaker 01 And that's in Japan specifically, or is that all over, starting in Japan?
5:04 Speaker 02 There are some regulatory reasons in Japan where they need to have some information about how old someone is. That's something that we're seeing that's being developed in other countries as well.
5:12 Speaker 01 Yeah, I mean, there's a lot of that, right? There's a lot of suddenly you need to prove your age. That's happening with, as you said, social media companies like Facebook are doing that as well.
5:20 Speaker 02 Yeah, we've seen that in Australia has started to do rules around that. The United Kingdom, the United States has rules for certain types of content. At the same time, we're also seeing more and more instances of companies that are using traditional IDs, you know, whether it be your passport or your driver's license, and then having trouble keeping that information private. And so one of the most important attributes of our system is that it's designed so that the user's data is held only by the user. And so the verification process doesn't require that a particular application, whether a game or social network, actually hold on to your data in order to know that you're a real human. We thought about that from the beginning in a way that most other service providers didn't.
5:57 Speaker 01 Yeah, I feel like because this tech is so new, there's a few reasons, but I feel like you're maybe going to be held to a much higher standard of security and ensuring the security of people's biometric data than, you know, as you say, Universal Studios just scanning my face so I can get in to ride a ride, right? You know, you're seeing across the internet that people have to give their IDs now to use Facebook. And they're freaking out about it. They don't want to give Facebook their ID. They don't want all of this kind of sensitive data on the internet. So what Orb is potentially providing or what World is providing through Orb is a potential solution. But then again, it still feels creepy, right? We're just like, oh, okay, let me like walk up to this gigantic orb. Let something backed by Sam Altman, who does not have necessarily the best track record for safety and for trust at this moment.
6:42 Speaker 01 Have my face and my eyes get like what? Yeah. Like, how do you assure people?
6:46 Speaker 02 Our goal is to not be held to a higher standard, but really to set a higher standard. The reality is that as the world has been progressing and as these new technologies have been coming on board, we as for a global population have just accepted that they're going to be crappy. We've accepted that governments are going to have the data and they're going to do whatever it is that governments do. What we have done with the World Project here to through humanity is say, that's not acceptable. You need to have the utmost privacy. You need to have the utmost security and it needs to be something that's verifiable. There's a phrase in Web three around decentralization. Our belief is the technology that we're building needs to be trustworthy to everyone. So it needs to be founded in mathematics. It needs to be based in mathematics and it can't rely on
7:30 Speaker 02 a significant amount of trust in any one party. For example, the orb, when it takes pictures, analyzes those pictures on the orb. It doesn't send them anywhere, right? It doesn't upload them to a server. It doesn't keep a copy of them. It analyzes them in a way that has been entirely open sourced. All of the hardware in the orb has been open sourced. All of the analysis is open source. And so anybody in the world can understand how it is that that works. And it makes an evaluation of whether this is a real person or not. And then provides the package of images to the user. And the user gets to use them how they want to use them. They can use them to go to a particular website or they can go to a particular service. Or they can choose not to do anything with it if that's what they'd like to do. But every time they use it, that's an interaction between them and whatever service interacting with.
8:14 Speaker 02 And again, World as an organization, Tools for Humanity as an organization, doesn't see any of those interactions. Cause we don't have a copy of the image. We couldn't see those interactions by design. And that style of interaction that is decentralized and allowed to be held by the user interacting with the services they want is fundamentally different from how previous generations of identity and security have been implemented. And the reason we did it that way. It's because those old ones are not good enough. What we've done is build the technology, open source the technology, and then made it available. So, for example, there's something that we've built called anonymous multi-party compute. It gets into the weeds, but the critical thing about it is it solves the problem of having
8:58 Speaker 02 a centralized service that's holding a bunch of biometric information, which most other identity solutions end up having that. We don't have it at all. There is no central mechanism.
9:07 Speaker 01 So is there no single point of failure?
9:10 Speaker 02 No single point of failure, no giant database of all of the images, no ability for anybody to misuse that data, because that's how all of the existing systems have been built. And not only do we think it's important that our system's not built that way, but we also think it's important that future systems not be built that way. And that's why we've open sourced it. If somebody else wants to build a good identity system in the future, we would encourage them to either use the one that we've built or use the technology that we've built and just go off and do it on their own.
9:35 Speaker 01 So you open sourced this, and this happened, I think, just last week, right? That you open sourced?
9:39 Speaker 02 We open sourced the most recent version of the orb last week. We have been open sourcing elements of the project literally every week since I've been here over the last two years.
9:48 Speaker 01 So when you say open source, though, because that can often mean a lot of things, is it just the documentation or is it the actual source code firmware, like the gad files, GitHub repo? Like, if I were, you know, someone who could do this, can I recreate everything from hardware to software on my own?
10:02 Speaker 02 Yes, you could recreate everything from hardware to software. So we're one of the few companies that's open sourcing all of our hardware. So the specifications of which chips are being used inside of the orb, the specification of what code is running on those chips, there are some very specific parts that we don't control. So, for example, there may be a modem inside of that device that functions just like a regular modem, but the modem manufacturer doesn't open source their part. But for the most part, it's all been open source. And certainly, everything that we control, we have made up.
10:30 Speaker 01 Is there any license or contribution policy?
10:32 Speaker 02 We've tried to make it as royalty free as possible. Our goal is not to make money on it. Our goal is to make it possible for anybody to use it.
10:38 Speaker 01 Okay. Have you published any third-party audits already? Is this-
10:40 Speaker 02 Third-party source code. So we can't always make it completely accessible. I don't think there's any fees associated with it, but there may be some constraints around how they can be used.
10:49 Speaker 01 Okay. And is the project actually being like used or forked by independent researchers or anybody else at this moment?
10:54 Speaker 02 There are some parts of AMPC that are being used by other parties, but it's not something that we keep track of.
11:00 Speaker 01 Okay.
11:00 Speaker 02 There's certainly nothing that we would do to constrain it. We would love to have other parties being involved.
11:04 Speaker 01 Okay.
11:04 Speaker 02 In terms of, I heard you ask a question about audits. We've done dozens of audits and we've made those open as well. Happy to point people to it. We've done that from a security standpoint, from a privacy standpoint. The claims that we make, we want them to be as trustworthy and open to inspection as possible.
11:19 Speaker 00 What moves our world forward? Energy. History has been shaped by it. Revolutions won with it and progress dependent on it. For 190 years, Schneider Electric has been a leading force in this story, creating the technologies that harness energy and transform the competitive landscape. And today, in the dawning era of energy and industrial intelligence, Schneider is helping customers build more efficient, resilient and sustainable operations across every sector. From data centers to homes, from buildings to industry. Schneider Electric, your energy technology partner. Discover more at se.com.
11:55 Speaker 01 Okay. And so obviously it's good to hear that if I come to an orb and I get my iris scanned, like that biometric data is just being like, yes, human. And then I get a unique identifier. Is this done on the blockchain? Or does it go into a wallet? Or how do I get to control that identifier?
12:12 Speaker 02 All of the above. So you generate the key yourself. The hash of that key is then confirmed to be held by the party who went to an orb. So we check to make sure that the key is associated with you as an individual. It's written into the credential that the orb generates. And then you're able to use that credential in the future. Now, there is a record that is currently being written to the blockchain that that key has been registered. And the reason that we did that is that there's a highly visible, highly auditable record of those transactions. We want as much transparency as possible. It could be done without that, but we wanted it to be more transparent. So the registration of a specific person is something that gets written to blockchain, which is why you can go and you can look and see that we've had over 17 million users that have gone and been verified. You can see sort of the overall growth of the project and the level of confidence that's happening there.
12:57 Speaker 02 And you can begin to get more visibility into how that's working.
13:01 Speaker 01 So I assume that like over time, you're going to want to improve whatever algorithms are verifying humanity. What are your training data sets like? If you're not keeping, you know, everybody's data to be able to use that in your training data set, what are you using? And is that open sourced as well?
13:15 Speaker 02 The training data is not open source because the training data is held by the individual that has volunteered it and they do it in confidence and ask us not to make it open. It's an interesting question of whether there are open training sets that could be used or it could be used to validate the accuracy of it. There are things like the NIST training set for biometrics that are out there and could be used. We use an independent process for bringing in training data. So we basically pay people to participate and interact with them in that context. So they know that what they're doing is participating in a training process.
13:45 Speaker 01 And are you paying them cash or are you paying them crypto? Because I know that there's a crypto element to all of this, that there's like the world coin, that if you do the scan, you get free crypto essentially.
13:55 Speaker 02 The specifics of any one of those interactions is going to be different. So there probably are some instances that are crypto. I also know there are definitely instances that we're paying in what gets called fiat currency because it's going to be different parts of the world, et cetera. There are also elements that are really important to keep in mind for biometric training. Like you need to have the same person involved in submitting over a period of time because we want to confirm, for example, that it still works with the same person if they come back at a different time. So it's fundamentally different from the way that people interact with the orb right now. For the most part, when someone goes to an orb, they go a single time at the verification point. And then in the future, they're not going back to an orb. This is not a very good way to do training of models because we don't have good ground truth that way,
14:36 Speaker 02 which is by design, but it's just, again, very different from trying to train on a data set of, for example, a huge collection of images captured by the same number of people who have a registered account, which is the model that's been used for other large scale facial recognition training, for example.
14:51 Speaker 01 Okay, so obviously there's an element to this that it's like, I have to go to the DMV to get a new driver's license, right? Like I have to literally go to a location to find a location if I want to be scanned and registered in this way.
15:04 Speaker 02 Which makes sense because remember the root problem is everything that's online can be picked with AI. And so you need to have an offline interaction. So the DMV historically has provided that in the context of a driver's license. And here the orb provides that in the context of getting a verified world ID.
15:19 Speaker 01 Yeah, but it also provides kind of like some friction, right? For people actually going to do this, right? So what's Toolstream Humanity's plan here? I know you've already raised, what, like $135 million? Are you planning to raise more and increase production of these orbs to make them more readily available around the world?
15:36 Speaker 02 Yeah, I mean, we're in the process right now. You mentioned we have thousands of orbs that are already available. We're also in the process of increasing our manufacturing of those orbs. We have manufacturing going on in Europe and then also in Texas and the United States. And we're expecting, you know, thousands of additional orbs per month coming out by the end of this year, which is standing up the Texas facility right now. So yes, is the short answer. The value to users comes through both access to services that already exist, but also future services. And so having a conversation with the big social networks, talking to gaming companies, some of the areas that I think are most interesting are very tightly connected to commerce. So for example, people are really frustrated by how challenging it is to buy tickets to
16:20 Speaker 02 go to a concert and the fact that those tickets get quickly purchased by bots and automation. And so, for example, in Thailand, just in the last week or two, there's a company called Event Pop, which does promotions of large concerts in Thailand, where they've announced they're going to start using World ID as part of their process of doing verification. And there's a few different ways that could happen, right? They could say the first tranche of tickets are available to people who have a World ID, or they could say people who have been to a concert before and been verified at an orb at that location are given accelerated access. And there's a lot of ways that this could be used to help improve the fan experience when they're going to a real event. And that could be for tickets, it could be for reservations for restaurants. There's a lot of different types of sort of commerce oriented problems that we're seeing
17:05 Speaker 02 right now in the world related to bots and automation.
17:07 Speaker 01 You know, one thing this reminds me of, yes, it's frustrating as someone who wants to buy a ticket to a concert that I can't get it because the bots already picked up a bunch of them, but increasingly, we're going to have agents do all of this for us, right? So I think you and I spoke about this before, about how there'll be kind of like a future layer to this where it also can use these unique identifiers to be able to tell if an agent is acting on behalf of a human. Is that right? Or did I butcher that?
17:31 Speaker 02 I think that makes perfect sense. There's no opposition to the idea that someone is acting on behalf of you, but ultimately, who is it acting on behalf of? And you want the bots to play by the same types of rules that a human might play. For example, if I were a ticket scalper, I can buy two tickets, but I then have to use them. So I can't really do that other thing, which has never been authorized and was never appropriate. It was just always easy to do and abusive. I certainly can't buy hundreds of tickets by delegating to dozens of agents. And so those types of basic elements of the social contract that exist in the physical world, figuring out how to have them exist in the digital world is what we're trying to enable.
18:08 Speaker 01 You said that there's a big use case for this in commerce, right? I can also imagine there's a big use case for this within governments, which is useful and also sometimes scary, right? If in fact this is open source and anyone can recreate it, including governments, is it out of your hands or can you still kind of prevent abuse and surveillance?
18:28 Speaker 02 So the way that we've designed the system is intended to make it very hard to abuse. And this is a fundamental part of trustworthiness of it is if it can be abused, then it's not a good system. So that's something that we've had to think a lot about. When you go to the org, for example, it doesn't retain any images. It doesn't upload them to a central location. It deletes them. The key that a user uses to interact with services is a key that they generated themselves. It's not one that they've given. We use what are called zero knowledge proofs, where you don't give away information about your identity in order to confirm one fact. So if a site says you need to be age 18, you can produce a zero knowledge proof that just says this person is a real person
19:09 Speaker 02 who is in control of their world ID that was registered at an orb and they are over age 18. It doesn't also say where they're located, doesn't also say where they live. It doesn't also say what other sites they've been to. So there's a lot of work that we've done to try to minimize the possibility of.
19:26 Speaker 01 And that's not something that like a bad actor could potentially manipulate on their own?
19:30 Speaker 02 No, no, we've, we've been very conscious of trying to make sure that that's not possible. Now. Could they take a different system and introduce those kinds of things? Sure, right? That's what the existing systems are, is systems that are designed to do tracking, et cetera. And so our hope and the sort of foundational idea for World is that if it's able to provide enough value to users and do so in a private way, then it can help a lot of the services that we already have know less about the people that they're interacting with, but also know that they are real people. The current process is you build a website, it gets to a large enough scale that it begins to suffer from abuse. And then you begin to collect a lot of data about the users in order to differentiate the abuse from the not abuse.
20:11 Speaker 02 And so we think having a private mechanism to do that sorting between real user and an AI eliminates the need to have as much information to tracking. So our hope is that over time that World ID actually makes not just the users of its own services more private, but really all of the internet more private and more resistant to, as you mentioned, abusive collection of data either by the individual party or by adjunct to that, like law enforcement or surveillance state.
20:38 Speaker 01 Or just, you know, data brokers or social media companies or anyone that wants to advertise to you at all. So I imagine that they would probably have a big problem with a type of technology that gave people more control over their own data. Like, the world runs on advertising, right?
20:51 Speaker 02 Governments have a complicated role, and the big companies also have a complicated role, where they want to provide services and do it in as private a way as possible. There's a separate role where they, in some instances, want to have monitoring and visibility. But for the most part, our belief is that mission of providing high-value services and doing so in as safe a way as possible is the primary mission of most of these organizations. And that's what they're going to be interested in investing in. And if that can be more and more successful, I think we can end up with a more private and also more successful population.
21:20 Speaker 01 And when you say, do you mean governments, like they want more privacy? Or when you say these organizations?
21:25 Speaker 02 Yeah, I think the reality is governments have a complex role, in the same way that large tech providers have a complex role. For the most part, they would like to have there be less risk for their populations when they're interacting online.
21:37 Speaker 01 Sure, yeah. But definitely not corporations. Like, I feel pretty strongly that the way that we interact online is constantly being tracked so that companies can get as much data on us as possible, so that it can then advertise us for stuff. Like, I can imagine a world in which they would not be super pleased with this type of internet that you lay out, which doesn't bother me, but...
21:59 Speaker 02 I think the conversations that we're having, I think they suggest that even the large advertising-supported organizations recognize that having a little bit less data, if they're still able to accomplish their job, is a good thing. And they also realize that the quality of interaction that people have on those services is the most important part of those services. If you can't go to a social network and truly be social with real people, you can't sell ads on that network, right? It has no utility. There's not going to be people there.
22:28 Speaker 01 Yeah, who's looking at the ad? Is it just a bot?
22:29 Speaker 02 I think all of these organizations see that as a bigger risk than having better privacy and having a little bit less information about people. So I think the tables have turned on that.
22:39 Speaker 01 Yeah, that's certainly something that I've spoken to Reddit specifically about how it's a platform for humans. That's why people love it because it's authentic. But if it's still anonymous, right? And so being able to still use that platform and know that you're getting advice from real people and not just a bot, I think that when those researchers came out and said that they had been on the Change My View subreddit, completely just creating fake AI personas, Reddit was like, all right, we got to clamp down on this because it's going to ruin our entire platform.
23:06 Speaker 02 And that's something that two years ago, people wouldn't have seen that, right? They wouldn't have had that same reaction. But now I think the reality is even the companies that we are all most cynical about are beginning to realize that they have businesses that are at risk. And the business is not sending ads to bots.
23:22 Speaker 01 Right.
23:23 Speaker 02 The business is making people want to be there and being able to service those people and having the relationships and the interactions they want to have online with their friends, with their colleagues, with their families.
23:32 Speaker 01 Well, the business is ads. Let's not beat around the bush. The business is serving ads to human eyes, not bot eyes. But I'm sure you've seen, this is a controversial topic. People are freaked out. People don't trust big tech or Silicon Valley generally. Anything else that you want people in this space to know? I mean, keeping in mind that there's a lot of founders and investors who listen to this podcast.
23:53 Speaker 02 I mean, I think we're at a seminal moment. Obviously, AI is something that the whole world is thinking about and focusing on. And there's a lot of benefit that can come to it. But I think we shouldn't lose sight of the fact that ultimately it's designed to be and intended to be a tool for people. And right now, we lack the ability to know who we're interacting with or what we're interacting with online. And that's something that, like, it is just weird.
24:17 Speaker 01 Yeah.
24:18 Speaker 02 That we've accepted that and not challenged it. And I think it's important that we make that change, that we have the ability to tell what you're interacting with online, whether it's a person or not.
24:27 Speaker 01 Lastly, if anyone were to be incentivized to go find an orb and put their face in front of it, what do they, I know we talked a little bit about this, but like, what do they get for the, what's their prize? Do they get some world coin? Like, what exactly happens?
24:39 Speaker 02 This is such a fun question for me because why do you go online?
24:43 Speaker 01 Because everything is there.
24:45 Speaker 02 Right. For some people, it's like, oh, I get to play Pokemon Go. For another people, it's like, oh, I get to see Netflix. For other people, it's like, it's to go on a social network. For other people, it's TikTok. Like, for every one of those people, it's a different reason. So it is certainly the case that you get access to all of those things when you go online, but which one is motivating for you is going to be a little bit different. What we see with world ID right now is there are some services that already exist, right? There is world coin, you mentioned it. So there are some connections to the Web three ecosystem and the ability to do financial transactions. There are some games, there are some social networks. If you're, you know, going to be doing online dating, then having increased confidence that the people you're interacting with are actually people before you go on that first real world date is pretty interesting. But I think what motivates you is going to be a little bit different.
25:29 Speaker 02 I think at its core, when you started the podcast, you mentioned, like, every day you're now interacting with something online and you don't know if it's real or not. If you think that's a real problem and that it's affecting you personally, then doing something to make that better is how I think about world ID.
25:45 Speaker 01 Be the change you want to see in the world.
25:47 Speaker 02 Yep.
25:47 Speaker 01 Yeah, cool. Okay, well, I think we're just about out of time. Where can our listeners connect with you online?
25:52 Speaker 02 Me personally?
25:53 Speaker 01 You personally or world or whatever?
25:56 Speaker 02 I mean, we are world.org. Unfortunately, I'm not a very social person, so it's not easy to find me.
26:01 Speaker 01 Yeah, fair enough.
26:02 Speaker 02 I'm not hiding, I'm just not out there.
26:04 Speaker 01 More power to you. Well, yeah, thanks again for joining us on the show and walking us through this very interesting project. To our listeners, you can find me on X and on Blue Sky. You can find Equity at Equity Pod on those same platforms. Talk to you next time.
26:22 Speaker 03 Equity is hosted by TechCrunch senior reporters and produced by Teresa Locan Solo with editing by Kel. Subscribe wherever you get your podcasts and see what's next at techcrunch.com/events. Thanks so much for listening and we'll talk to you next time.
Previous From SB 243 to ChatGPT: Why it's ‘not cool' to be cautious about AI Next OpenAI wants to power your browser, and that could be a security nightmare Page updated October 12, 2026 · transcript and summary are machine-generated