‍

{{big-quote}}

‍

What if you could prove your identity to anyone, anywhere, without sharing your personal details? Alex (David Alexander Scheer) from zkMe explains how zero-knowledge proofs could end the era of uploading ID documents to countless websites. But can a startup really take on tech giants like Google in the identity space?

‍

Alex shares his vision for revolutionising digital identity. From protecting medical records to transforming online advertising, zkMe's approach to data privacy could reshape how we think about personal information in the age of AI—all while letting people earn from their own data.

‍

‍

Transcript

Jonathan Nguyen (00:00)
Welcome back to another episode of the Unsensible Podcast where we talk to the lunatics who think they can defy the odds and build a startup to change the world. And today I am talking to Alex, who has a much longer name on his LinkedIn profile, but I'm gonna let him pitch himself in a second. I'm not gonna go too deep into it. Alex from zkMe over to you. Tell us what zkMe is all about. Go.

‍

Alex | zkMe (00:30)
Thanks so much, Jonathan, for having us. It's always a pleasure talking to you. First of all, it's David Alexander Scheer, a very long German name. I usually go by Alex. It's like the easiest in every language. Yeah, actually it's pronounced ZK-me, like the more American way. (It) stands for Zero Knowledge Me. It's the promise that you can prove something about yourself, the identity and all of the experience and all of the credentials that you have without revealing any private information. So it's the promise that you can prove and monetize your identity in full privacy and full decentralization.

‍

Jonathan Nguyen (01:09)
we met at a, I think it was in Japan, I think it was at a blockchain conference, But when I saw what you guys did, I immediately understood the applications and its importance, which is very rare for a blockchain slash crypto type project.

I'd love for you to unpack a little bit about where you see kind of the current model of how we prove ourselves is and how you guys address it.

‍

Alex | zkMe (01:42)
Yeah, so at the moment, first of all, identity is a completely underestimated vertical and need within industry. If you think about it, pretty much anything that you do is tied to some form of your identity. Even opening your phone is tied to the identity that is owned by Google or by Apple.

Tap your octopus card here in Hong Kong that also goes to like a federated identity that is owned by private company and in general the approach has always like always within the last 20 years been the following that you have a federated identity owner so someone that owns the identity in your name and has the ability to to monetize and ascribe some kind of value to that identity biggest issue is that this creates natural monopolies like you have companies like Google, most of their money through identity, through profiling, knowing who you are, and stifling a little bit of innovation. If Google is the only one that has access to your data, and especially in an era of AI, has the monopoly of monetizing that data, it makes it very difficult for smaller providers and up and coming startups to profit from that data as well.

So what we do is we say there's technology that allows you to self-sovereignly own your data. And then we use a mechanism in order to separate the private part of the data from the value of the data. And that allows us to very seamlessly monetize the value of the data without touching anything that might be sensitive, or might be covered by legal requirements. And that technology is called a zero-knowledge proof, which is a mathematical proof that a calculation has been done. And that's also where the comes from.

‍

Jonathan Nguyen (03:47)
one area aside from monetization is we have at the moment, a lot of data breaches we have, and you have to prove, you have to prove yourself and your identity on so many different websites. You sign, you get an email account, a trading account, bank account, any account now you have to prove that you're this person. And it always makes me uncomfortable to upload my details to these sites because it's not a question of will they be breached. It's just a question of when governments get breached with state level funding. So talk a little bit more about the implications for, security and being able to prove your identity without releasing all your information. Talk more about that.

‍

Alex | zkMe (04:22)
Yes.

Yeah, so that's one of the core value propositions that we offer. So monetization, interoperability and security. So in regards to security, the main differentiators that we encrypt and pseudonymize the data directly on data collection. So there's no replication, duplication of the underlying data that makes data so vulnerable.

One of the reasons why big state agents or big identity databases get hacked is because they represent one giant honeypot. Like no matter how much you protect it, if what you're protecting is so valuable, it makes an attack like profitable as well. So what we do is we encrypt every single data set singly with the same kind of protection that you would normally have in the world, so like a 16 or 32 bit encryption. Making it really unattractive for hackers to attack those data sets. Because even if they were to attack, maybe after a month of trying, they would only unlock one person's single credential. So that's the core approach that makes the solution more secure. And then the thing that is tradable, like the thing that changes hand in a high frequency, is a completely public and anonymized set of information. So for example, just the fact that I'm male and that I'm over 18 without telling you like what my name is exactly, where I'm exactly living, what my birthday is exactly.

‍

Jonathan Nguyen (06:18)
you would sell the solution in a B2B context. I can see how that, you know, a bank, a crypto exchange would get value from engaging with you guys. We've talked a little bit about, you know, the potential for a retail product or a direct product.

Can you talk about that publicly yet? What can you tell us?

‍

Alex | zkMe (06:43)
Sure, can. I mean, we're already hinting at some of the value proposition that we are expanding upon. By having a credential that is fully anonymized and that is owned by the individual. You unlock the potential to reuse that credential. So let's say you KYC it with bank A and then bank B comes to you and says like, know that you KYC it with bank A, why don't you come to me? I give you some special offer. And these kind of use cases only become available if there's full transparency and an open data economy that has the user in the center.

And we're currently building that infrastructure that allows both sides to have discoverability for the user to know, okay, with my kind of credentials, I can participate in these kinds of services. And for the service providers to say like, I have this kind of service and I'm targeting this kind of user. we are like zkMe will be the open infrastructure in middle that helps for the data economy to flow efficiently.

‍

Jonathan Nguyen (07:52)
So we've talked about some various applications. So you've just given one example, but what are some other examples? Because platforms are notoriously hard to sell, right? Because people don't understand what is a platform? What are you really selling? What are you really selling me? So make it really tangible. What are some use cases?

‍

Alex | zkMe (08:04)
Yes.

Yeah, in general, I think for identity there's only two plays at the moment. There's either a traffic play where you just say like I'm collecting a massive database of users that I can then monetize just because there are many users or there is the high value play where you target.

information or credentials identity segments that are of high value and within zkMe we've decided to go the high value plate and the first use case that we tackled and that is already live is the ability to verify yourself in a legal sense so KYC, anti-money laundering, these kind of due diligence requirements in zero knowledge so anonymously.

Mostly because it's naturally a segment that has a very high willingness to pay. And expanding into other identity verticals will follow the same logic. So we are thinking of which other vertical has a very high willingness to pay for data. And we're like full transparency, maybe this is Alpha, looking into healthcare as the next vertical. Because we believe this, like healthcare data is super valuable, has a lot of inefficiencies and just ripe for disruption.

‍

Jonathan Nguyen (09:37)
Yeah, and people whose data it is often doesn't get to profit from it.

‍

Alex | zkMe (09:42)
Yeah, there are some really extreme examples in the last couple of years of companies that collected healthcare data and then unknowingly to the user sold it for billions of dollars.

‍

Jonathan Nguyen (09:56)
23andMe comes to mind immediately. And there's, I think at the moment there's even talk of acquisition of 23andMe, because it's a bit of a distressed asset at the moment. And the buyer of 23andMe will potentially have access to all of that information as well. And we don't know who the buyer is gonna be, right?

‍

Alex | zkMe (09:59)
Yes, that's probably the most publicized example here.

And it's really scary data, right? Like if you are able to combine a name with a genetic marker, that's like among the most private data that you can have.

‍

Jonathan Nguyen (10:31)
Yeah. So healthcare, mean, I'm a massive, massive proponent of being able to own and monetize my own healthcare data. It's not as nefarious as it sounds. I think there's a lot of research, researchers looking for drug-like cures and genetic researchers who would benefit from it. And I'd like the option to be able to supply my data to them if it's beneficial to them.

‍

Alex | zkMe (11:00)
Absolutely, and you get paid for it and also you know that you're only sharing the anonymous part of your data. Like for most cases, like for most non-nefarious cases, actually only the statistical part of the data is really valuable. Like if I'm an insurance company and I'm thinking about which diseases I cover, what kind of medicines I cover, I only care about knowing what is the statistical distribution within a certain population of a certain genetic marker and not does Jonathan have this marker. So I believe there's a lot of potential for just making the whole economy more efficient if you have an efficient data layer that allows for the exchange of such anonymized data.

‍

Jonathan Nguyen (11:49)
So to get to make that happen, you need to end users on board. You need to get practitioners on board. You need to get other data sources. I mean, there's multiple parties that need to come onto this flywheel. And when they're on board, it's pretty much unstoppable. How do you get them on board?

‍

Alex | zkMe (12:11)
I knew this question was coming. It's as stupid as it sounds, like one step after another. You try to build connections. The way I see it is the user side is the strongest one, right? Like if you have a lot of users that becomes such an attractive value proposition to both the consumers and the data providers.

And that's what we are currently focusing on building up. So we know that the solution works. It has been up and running for a year. All of the bugs are cleaned out. We build our own protocol that is on top of the infrastructure. And now it's like it's the next proof that we have to provide is that this works at scale, not just 600,000 users, but maybe like 2 million users, 3 million users. So that's the next step.

‍

Jonathan Nguyen (12:52)
Mm.

Okay, let's talk, let's dig into that a little bit. I mean, all this infrastructure sits on a public ledger, or is it public private public? So are you reliant on, you know, the, you running on, what is it, Ethereum, Polygon?

‍

Alex | zkMe (13:15)
So at the moment we're running on Polygon. quick hint, we're moving to our own app chain to have the ability to be a little bit more independent from existing infrastructure providers. Because they have not been built for data in mind. They have only been built for security, throughput, and digital asset transactions in mind.

Jonathan Nguyen (13:28)
Yeah.

Yeah. how will that, what will have to happen? Do you have to spin up servers? What has to happen there to make that work?

Alex | zkMe (13:48)
Yes, so in order for any network to exist you need to have some form of consensus mechanism and in an optimal scenario it's like both globally and in numbers distributed and decentralized. So what we're planning to do and we're already in discussions is to have node operators in all major data jurisdictions to ensure that all data protection laws around the world are followed. So if you are, for example, a European user, we can guarantee that your data is only stored in Europe. Or if you are a Hong Kongese user, your data only stays in Hong Kong. And that's a process that will take a couple of months.

‍

Jonathan Nguyen (14:28)
This is...

And this is quite a niche question, but on that matter, some of these jurisdictions will require destruction of data at the user's request. How do you do that in a decentralized system?

‍

Alex | zkMe (14:45)
That's actually a super important question. There's many competitors that are ignoring that question, even though in my opinion it's like at least from a legal perspective one of the most important ones. Like permanence is highly illegal in Europe. Like as soon as you create a solution that makes data, like personal data, be stored forever, no matter how much you protect it for encryption, that's just purely illegal.

So what we do is a combination of two things. Every credential and every data set has a limited, has like a validity that is preset. And after the validity, the data storage just expires and gets deleted. And every credential has the ability to be self-managed by the data holder. So you can go into your identity wallet on app.zk.me and say the following credential I don't want to use anymore. just delete and then it'll stay in the system only as long as it has been confirmed by.

So if the user went into a with a genetics company to say this dataset is available for at least five years, then after those five years it will automatically be deleted.

‍

Jonathan Nguyen (16:01)
Yeah, that's interesting way of solving the problem. Because some of these things on the ledger, can't really go back and delete what's on the ledger.

‍

Alex | zkMe (16:11)
Yeah, and I think it's a big issue in general that many web3 companies don't have an eye out. So what we store on the public ledger are just pointers towards decentralized storage. So you would forever be able to see where something was stored, but the underlying storage will not be accessible or just not exist anymore.

‍

Jonathan Nguyen (16:36)
So then the next that kind of leads me on to the the next question is okay I'm interested in this I want to sign up I want to make money out of my own data.

How do you see people signing up for that on the user side? Is that going to be like you have to drive them to the ZK.me app or do you see that there's a tick box when they go to the doctor and say, hey, I sign up for this. I wanna make money out of this consultation.

‍

Alex | zkMe (17:17)
Both will be possible. Any user can go to app.zk.me and just try it out. But in general, most of the discoverability of identity solutions is going to be through the service provider. That's how we won most of our users so far and how we will continue to get most of our users. So for example, if you open a bank account...

It'll just say like your identity is protected through zkMe. Here's like a quick explainer. Verify your identity and then at the very end it says, please feel free to visit to learn what you can do with your identity, like this. And then stickiness factor is relatively low, but that's how you get, at least you have like a constant flow of new users.

‍

Jonathan Nguyen (18:04)
This also points to another factor I think is when you monetize and then when you're hosting nodes, is there going to be some kind of token that you're thinking about that obviously this is an exchange of value, right? So here is a use case where I can see a like legitimate use case for token exchange as opposed to just pump and dump.

Talk a little bit more about that. What are you thinking about in that kind of direction?

‍

Alex | zkMe (18:38)
Yeah, mean, nothing confirmed yet, but we do tokenize three different aspects of your identity. So we tokenize the data itself, store it under your wallet, you own it, you can control it, but that's just data. Then the second one is tokenizing the value of the data that I mentioned previously, the zero knowledge proofs. Those are technically NFTs that are minted onto your wallet and that you own, transferable but there's still tokens on your chain. And then as you mentioned the third aspect that we tokenize is the exchange of monetary value to that data. And yes it would make sense to have like a utility token that allows for different stakeholders to come together like an issuer like let's say in the future a genome company says my genome test is worth five dollars for every time it gets verified they could set that on chain as a parameter and then get five dollars equivalent of utility token from like a service provider that needs access to that. And then maybe that original issuer says, in order to be fair, I give like one dollar of those five dollars back to the user. So the user gets those tokens into their wallet. And then within no time you create an economy in which even though there is no centralized orchestrator, whoever has need for data and whoever has supply of data come together.

Without having to rely with a centralized orchestrator that sets the prices. So you have an open market that defines what the real value of a data set is.

‍

Jonathan Nguyen (20:24)
What are the downsides here?

‍

Alex | zkMe (20:26)
Downsides to launching a token. I mean there's always downsides like by building a network.

It's a big undertaking, right, that only works at scale. Just thinking at what I mentioned previously, there are at least 21 jurisdictions around the world that have data provenance requirements, meaning you need at least 21 different nodes around the world that are running your solution, like this network.

Also have like redundancy and so on, it gets to a relatively big number relatively quickly. And then obviously any data that includes individuals is like breadcrumbs, right? Like for a KYC solution, we're talking like maximum $1, $2 per user, per KYC.

‍

Jonathan Nguyen (20:58)
pretty fast.

‍

Alex | zkMe (21:16)
For genome data, we're probably talking like $50 per user, something like this. You will not be able to run a global network with anything less than a million users. So we're talking scale here.

‍

Jonathan Nguyen (21:31)
So it's a big job ahead of you.

‍

Alex | zkMe (21:33)
Yeah, but it's an exciting one and I really believe in what I'm building. So it doesn't feel like work. It just is, think, the right way, like how a startup should feel like.

‍

Jonathan Nguyen (21:47)
Is there another use case in a vertical that is unrelated to health that you see immediately?

‍

Alex | zkMe (21:55)
I mean the very first thing, and I'm surprised you didn't bring it up, is advertisement. I mean that's where most of the money is at the moment. But that requires even higher scale. But the pie is probably the biggest in advertisement.

‍

Jonathan Nguyen (22:11)
Pi, the Pi is enormous, the, it's, each transaction is, I guess, a much lower value transaction.

‍

Alex | zkMe (22:19)
Yes.

‍

Jonathan Nguyen (22:19)
And, but again, it's one of those things where an acquisition would be interesting because if, you know, one of the big advertisers turns around and says, you know what, we're not going to build this ourselves. We're just going to buy it. Google's like, you know, we'll do another AdSense style acquisition or AdMob style acquisition and we'll just take the lot. Thanks. We'll integrate it into. That's probably how it will play out.

‍

Alex | zkMe (22:46)
Yeah, I mean that's potential scenario. But I hope that no matter where this technology ends up, it really helps users to monetize their own data. Like there's so much potential and it would be a pity if a solution like ours wouldn't find a scale.

‍

Jonathan Nguyen (23:06)
anytime any technology that is really disruptive enters the market, there is a very high incentive for the incumbents to have it not work. So.

From your perspective, what do you see that the roadblocks that the incumbents would put in your way? And maybe not now, but maybe in a few years, what kind of roadblocks do you think they'll throw in there?

‍

Alex | zkMe (23:33)
Hmm, very good question. So the very first thing that comes in mind is just data sources, making data sources unavailable. As an infrastructure, you never have any humans in between, right? We're building a purely web3 solution, meaning it's fully programmatic. There's no humans ever looking at data. It's only code, like receiving data, anonymizing data, and then making data tradable.

meaning code requires a very high-performant data source. So if you're working with data source providers and just kept that and say like, it'll be forbidden to use oracles to bridge credentials from traditional internet to web3, that'll be a big blocker for solution like ours. There are ways around and I think it'll be impossible to completely block.

‍

Jonathan Nguyen (24:16)
Mm.

‍

Alex | zkMe (24:25)
One attack. And then obviously as an early stage startup just interrupting anything before it reaches scale. There are big blockers on the way to reach scale, like additional node operators getting visibility and so on. That's where big incumbent players could come in and still stop this.

‍

Jonathan Nguyen (24:50)
And if it all works out, like five years from now, 10 years from now, what does that look like?

‍

Alex | zkMe (25:00)
I think you'll have a very diverse picture, but I think the biggest vision that I have is that in five years you will not be verifying yourself with paper documents anymore. I think that's the most obvious thing that is currently in decline. More and more regulators, more and countries, more and more private companies fully lean into EIDs, electronic identifiers, for legal purposes.

And zkMe will be that interoperability layer that makes it such that no matter where you are, no matter what kind of use case you want to do, you'll be able to prove yourself legally with just one click.

‍

Jonathan Nguyen (25:43)
I have one last question for you. It's a rather obvious question that I'm surprised I didn't ask upfront.

So you're German sitting in Hong Kong.

Tell me how you got there. Why built this business in Hong Kong? What's going on there?

‍

Alex | zkMe (26:02)
I mean, it's a longer story, so let's see how much time you have. yeah, long story short, I used to work in automotive and worked as a consultant, worked specifically as a software consultant in the automotive industry. Got relocated to Shanghai, led a software development department in Shanghai until COVID hit.

And then during COVID, I just realized like this job is getting a little bit too repetitive, a little bit too boring. Why don't I try to do something on my

And just at that time a new law was taking shape called MiCA, which is a European regulation that is starting to be enforced on the 1st of January of 2025. That among other things states that any financial service that is provided by a smart contract and is not sufficiently decentralized will be deemed a centralized financial service.

with all of the implications like they will have to KYC every single counterparty, every single user. And knowing how DeFi works, I was like, if there's no identity solution in Web3, this is going to kill Web3 and crypto as we know it. So at that time I did a little bit of research. I was really fascinated by the concept of zero knowledge proofs, did like a deep dive.

and I realized like staying in China would not be sustainable for the company so I looked around a little bit and I've always loved Hong Kong and at exactly the right time Hong Kong stated we're going to be the the web3 capital and web3 hub of the world so I just said like okay let's jump on that bandwagon incorporate the company in Hong Kong and that's it like been here for the last two years and

‍

Jonathan Nguyen (27:47)
All the stars have aligned.

‍

Alex | zkMe (27:53)
really enjoying it. It's a great city.

‍

Jonathan Nguyen (27:56)
Awesome. Well, really great to chat Alex as always. And I have no more questions for you, which is surprising. We've been very succinct. So looking forward to seeing you. We're catching up in person. I think there's going to be an accelerator, a demo day. think we're worth going to be at.

‍

Alex | zkMe (28:09)
Yeah

Yes.

‍

Jonathan Nguyen (28:20)
And on that note, thank you very much and see you soon.

‍

Alex | zkMe (28:25)
Thank you so much, Jonathan, for having us again. See you soon.

‍