Stephen: This is Stephen Sargeant live at ECC. This has been such an exciting conference. A lot of talk about wallets, tokenization, real-world assets, regulations, and compliance is such a hot topic. As you can see, we've been blessed with this beautiful background, this backdrop of amazing scenery. We're having the best conversations on this couch right here, talking to the movers, the shakers, the builders of crypto compliance, talking to all those that touch the unique aspects of blockchain analytics and compliance risk, and some of the best risk and transaction monitoring you can find in France.

We're live at ECC with the Around The Coin podcast. We switched venues. We're going to the Zama Villa to talk to Jeremy Donato, the COO of Zama. We go deep into FHE encryption and the privacy. We go deep into the confidentiality layer that the financial institutions dipping their toes into digital assets will have to deal with, especially around FHE and encryption.

We talk about operational aspects. We talk about the balance between the transparency of the blockchain, but the confidentiality needed for some of the top institutions to transition to digital assets. Zama is building an ecosystem around confidentiality as they are the layer that bridges traditional finance and DeFi, and we are here for it at ECC and the Around The Coin podcast.

Jeremy: Jeremy. Hi. I'm, I'm Jeremy. Uh, uh, Bradley Severo Donato, like a super long, complicated surname. Uh, I am originally from the US. I've been living in Paris for, I guess six years now, so quite a while. Uh, I'm the COO at Zama, which means I look after pretty much all the non-technical side of the business, so everything from kind of, uh, marketing, HR, finance- Mm-hmm

legal, all of that fun stuff, depending on the day. Yeah. And I've been in crypto for as long as I've been at Zama, which means not for very long. Uh, and before that I was working at EdTech. And I'm also a writer as well, so a little bit of a Renaissance man in my mind.

Stephen: You, you mentioned edtech. I remember when I was first- Yeah

on the website of Zama, it gave these really cool instructional videos and understanding. Yep. I felt like that you leveraged your edtech background- Yeah, that's right. Yeah. ... to create something like that, and I was wondering, like, what are your thoughts about that? Should more protocols have something where people can easily understand what the- Yeah

how it works in practice- Yeah ... versus getting into all the technicalities and primitives and all these things that people don't really have a good grasp of?

Jeremy: Yeah, I mean, I think just those buzzwords alone are enough to scare people, right? It's like, what's a primitive? Yeah. Like, so that, I think, is, um...

Sometimes we, we're, we get stuck in this kind of like crypto echo chamber. I mean, you see it happening on Twitter or- Yeah ... X as it's- Yeah ... called now. You see it happening, like, when we talk to each other. Um, and I think that's one of the biggest barriers to adoption is just, like, understanding the space.

So anytime we can do something which, like, breaks it down into really, like, simple concepts, I think this is really, really important. Yeah. I mean, if I can explain it to my mom- Right ... that's, like, the level I think we should be operating at most of the time.

Stephen: Can you explain Zama as if you were explaining it to your mom?

Jeremy: Yeah, well- ...

Stephen: I,

Jeremy: I mean, that, it already becomes complicated- Yeah. ... you know, because it's like when you're talking about a, a, a company who, uh, five years ago had technology which really wasn't usable, you know? Right. So the way I would explain it back then was we're working on something called FHE, fully homomorphic encryption, which means that you can encrypt something, something, right- Yeah.

end to end. So normally we talk about things like, uh, personal data, um, something you wanna keep secret, quote, unquote. Yeah. I don't like to use the word secret because it's kind of a nuanced word- Yeah ... but for a really simple audience, it makes sense. Um, generally we're talking about keeping data encrypted.

Encryption itself is, um, also a nuanced subject- Yeah ... because there's lots of ways to encrypt something, right? Right. Um, FHE does it in a way which is privacy preserving, essentially. Um, and when we talk about- Privacy, we're talking about confidentiality. Yeah. So keeping things confidential, not necessarily 100% fully, uh, black boxed.

And that's an important nuance because when you-- when it- when we come to things like, uh, privacy, uh, in terms of compliance, you wanna be able to unlock the data at certain points to give regulators access to the data.

Stephen: And I think you came up with that kind of confidentiality- Yeah ... layer, which is important- Yeah

'cause when you start s- using the word privacy around regulated financial institutions- Yeah, people get scared. Yeah ... but when you say confidential, they're like, "Yeah, we don't want all of our- Yeah ... embedders seeing these transactions." Yeah. Can you go into a little bit about the different aspects of privacy and how important-- We're in Europe.

We're in Paris. Yeah. And this conversation is a lot deeper than a lot of North American conferences- Yeah ... because they really value their data sovereignty and the privacy and confidentiality. Can you talk a little bit, maybe focus on this region, why it's so important- Yeah, sure ... to the institutions that you're serving?

Jeremy: Yeah. I think you've touched on something really interesting there, is that it depends on who you ask what they value, right? And what confidentiality means. If you ask a regulator, uh, it means one thing. If you ask a financial institution, it means something else. And those two languages, let's call them that, don't necessarily mesh.

And I think that's where Zama's solution becomes really interesting, um, because we enable something called programmable confidentiality, which means... All right, let's say we want, on a very basic level, we want to encrypt, uh, a transaction from the point of view of the end users on both sides of the spectrum, right?

You do wanna have this point in the middle where selectively someone, either a financial institution or a regulator, can decrypt the data, read it selectively- Right ... and get the information they need to do what they need to do, whether that's run their ledger or comply with some sort of regulation. In Europe, that's particularly important because, I mean, on a very basic level, we have GDPR- Right

Stephen: for

Jeremy: the right concerns. So the-- Again, that goes both ways because FHE can help you comply with GDPR, but It can also be kind of scary when you're thinking, "Well, everything's completely confidential." Right. "How can we know that we're complying with this regu-" Like, those sorts of, like, scary ideas, right? And then you get into this space where people start thinking about, well, if it's, if everything is hidden, everything's private, then it's gonna be used for nefarious purposes.

Stephen: Right.

Jeremy: Right? So that's why this programmable confidentiality is so important.

Stephen: Do you think that was the missing piece? It was either privacy for everything, and then regulators were always asking, "Well, what happens if we need to see a transaction?" Yeah. "Or we need to see what's happening?" Yeah. "Or you're not dealing with North Korean hackers?"

Yeah. Was that the concern, that it was either all privacy or no privacy at all?

Jeremy: Uh, yeah, I think so. There is a fine line between trying to meet the market demand for confidentiality and at the same time, um, making sure that you're protecting, protecting against bad actors, right? We cannot pretend like bad actors don't exist and that the technology could not fall into the wrong hands, but that's true of any technology.

Yeah. You know? It's just that we're at a good inflection point, I think es- particularly in Europe, where more and more consumers demand a level of confidentiality. Um, I think financial institutions, Layer 1 and Layer 2 chains know that.

Stephen: Yeah.

Jeremy: So they know they need to, to implement some sort of confidentiality solution.

It's just about how do we do that in a way which also preserves the right of regulators and institutions to have limited access to that data. Could you discuss some of the regulated case studies for

Stephen: regulated entities and what they would be using Zama for? I think, you know, it's been recently we've seen a lot about tokenized assets- Yeah

and, you know, a lot about privacy, especially here in Europe.

Jeremy: Yeah.

Stephen: Can you give us some use cases of what companies, what kind of institutions are using Zama, and what's the benefit of them using FHE through Zama?

Jeremy: Yeah. Well, we just announced a partnership with, uh, T. Rowe Price, uh, recently, which, uh, T. Rowe Price is, is, uh, essentially part of Apex Group, uh, so they have something like over 3 trillion in assets.

Um, and the amazing thing about that, apart from this number, which sounds really impressive- Yeah. And

Stephen: it is

Jeremy: And it is, yeah Is that Zama will become the default confidentiality layer for that protocol. So that means they'll be able to process transactions using Zama's tech on their, on their protocol.

Stephen: Yeah.

Jeremy: Um, so that enables you to do things which before maybe weren't exactly feasible on chain. Uh, so if I'm thinking of any sort of financial transaction, if you wanna do that on chain, it's a little bit complicated if everybody has access to see all of the transactions- Yeah ... right? Simple example, if I'm gonna pay my employees on chain, I don't want other employees to know how much I'm paying, right?

Like, hey, you're opening up a whole, a whole like, uh- Yeah ... Pandora's box of- Close

Stephen: your ears over there. Close your ears over there, guys. You're all getting paid the same amount.

Jeremy: Yeah,

Stephen: yeah.

Jeremy: You know, it's like I don't want someone to know how much I make. I assume other people don't wanna know either. Those sorts of use cases now become possible thanks to FHG and thanks to Zama's tech.

So yeah, salaries are a good use case. I think, um, any sort of like paying suppliers.

Stephen: Right.

Jeremy: Another interesting use case is paying influencers- Hmm ... which I hadn't really thought about before until quite recently when we had to pay influencers. Yeah. And, um, of course, every influencer has a sort of like a different, um, a different pay rate- Right

of course, a different rate card. Uh, you don't necessarily want influencer A to know that you've paid influencer B, uh, I don't know, 5,000 more- Right ... or whatever it is, right? So it allows you to, to do business in a way which is a little bit Um, more transparent and a little bit fairer.

Stephen: You talked a little about T.REX, and they've come up with the- Yeah

standards for tokenized assets. Yes. Where do you play in that kind of ecosystem of tokenized assets? Yeah. And now there's a standard that they're pushing for more members to get involved with-

Jeremy: Yeah ... and

Stephen: use when it comes to putting any asset on chain.

Jeremy: So yeah, I mean, Zama is, um... I- i- in one hand, we're kind of agnostic when it comes to these sorts of things because we wanna work with as many players as possible to enable these sorts of use cases that we just spoke about.

But at the same time, we, we want to champion, uh, partners like T.REX to push these sort of standards forward, because once the industry becomes more standardized, it allows players like Zama to really thrive, right? So it's a win-win for us, as well as for partners like T.REX. Can

Stephen: you talk about some of your other partners and clients and what you're doing with them?

'Cause I'm assuming they're using your technology to do, um, a, a number of different features, a number of different operations.

Jeremy: Yeah, I mean, any sort of FHE provider that you probably have heard about, those sorts of solutions that are on the market now- Yeah ... like, like Inco, like Phoenix, they're all using Zama's technology on the back end.

So yeah, all of those guys are using Zama's technology on, on the back end. Not to sound, like, too boastful about Zama- Yeah ... but we don't have a strong competitor at the moment because, uh, it's taken five years to get to this point where FHE is usable in, in, like, kind of mainstream production. Um, and we saw that recently with the launch of the Zama Token, and what we, what we discovered is that, uh, Zama's protocol is actually even faster than Ethereum now, which is really cool.

So, like, the bottleneck during the auction process was not on the Zama side, it was actually on Ethereum's side. Hmm. So that was really good because it means that all those kind of things people used to say about FHE, that FHE is too slow in practice, it's never gonna work- ... those sorts of, like, those objections are, like, out the window now-

which is really cool.

Stephen: Can you tell me about the launch? It was a huge launch, a lot of traction. Yeah. Tell me what were some of the learnings that you had from launching?

Jeremy: Oh,

Stephen: yeah, a

Jeremy: lot.

Stephen: I think tokens is one of those things you launch- Yeah ... everything on paper seems w- like one thing- Absolutely, yeah ... and the market tells a different story or shows you different interesting- Yeah

use cases. Once you put it out there, the market uses it the way it wants to.

Jeremy: Yeah.

Stephen: So any findings from the launch?

Jeremy: Well, even before you get to, like, go-to-market phase of the token launch, there were a lot of learnings. For many of us in Zama, it's like, it's our first time launching a token, you know? Yeah.

And I think it's one of those things where no matter how much you read about it, you're just not ready until- It's time to go, right? So there was, as I touched on a little bit earlier, there was the whole influencer marketing phase. I'm used to that with other projects, but it's a little bit different in the kind of token launch phase because you have to figure out, first of all, which influencers to work with, which audiences you wanna reach, what that looks like from a retail point of view.

Stephen: Yeah.

Jeremy: At the same time, we also were trying to drive institutional adoption of the token through our investors and through family offices and things like that. So we're running like parallel routes, which was interesting from a marketing point of view because a lot of token projects are launched to one audience or the other- Right

not both, right? Right. So that was probably lesson number one, which was just how to market the token. Lesson number two was around the, uh, KYC and KYB process. I was not ready for that at all. Right. I mean, we were ready from a, uh, from, from a compliance point of view. Like, we knew what we had to do, what we had to collect and things like that, but I was not ready for it from a kind of customer support point of view.

Stephen: Right.

Jeremy: Because suddenly we had like all of these, uh, retail investors wanting to sign up, like-

Stephen: Wanting to know how come it hasn't doubled in- Yeah, yeah, yeah ... 32 hours and like-

Jeremy: Yeah, and to-

Stephen: Where's my money? Yeah,

Jeremy: and to like even to just take part in the auction- Right ... it's like, well, you've gotta submit some basic KYC documents, passport, proof of address, things like that.

But I thought this was

Stephen: privacy

Jeremy: focused. Yeah, yeah, yeah. Yeah. So we had that. I think we processed like around 20,000, uh, KYC applications, I guess we can call them that, uh, within about 72 hours. So it was like all hands on deck just to get that done. So I was fielding messages on X, people reaching out to me on email, like LinkedIn messages, Instagram messages.

Right. Like, people found me on all of my platforms, um- That was, it, it became kind of fun.

Stephen: Yeah.

Jeremy: Because, like, people are really happy when their KYC got approved.

Stephen: And you get a chance to educate them about- Yeah ... hey, this is why it took so long. Yeah. By the way, this is what we're working on. Yeah. So there's a lot of touch points with people.

They feel like they're part of the process.

Jeremy: Yeah. Yeah, and I think it was cool for people to get a message personally from someone in the leadership team- Right ... of the project. That, I think, doesn't happen on a lot of projects. A lot of times support is outsourced, and I think that's kind of sad. Like, we should be championing our own products to our own users, you know?

Stephen: Yeah.

Jeremy: Um, so I got called everything from, like, the bald god of crypto. Yeah,

Stephen: that's what you guys want to put on your new LinkedIn headline.

Jeremy: Yeah. Yeah, yeah. But then there was a flip side of it, because of course you have some KYC you just cannot approve, right? That's not gonna go through. Yeah, yeah, yeah.

And so then you also get, like, you end up with death threats and, like, crazy, like, "I'm gonna come to Paris and find you." Yeah. So it really runs the gamut of responses.

Stephen: Which

Jeremy: is a realistic...

Stephen: Which is possible now with what we've seen happen in Paris- Yeah, it's pretty scary ... and around the world is- It's so

Jeremy: scary

Stephen: these people have access to dark web. There's a lot of doxing going on in the industry.

Jeremy: Yeah.

Stephen: Yeah. I'm curious from your landscape and from what you've seen, you've raised at a billion-dollar valuation- Yeah ... which has to be cool. Yeah. And now you're in the operation lens.

Jeremy: Yeah.

Stephen: Where do you deploy that capital?

How do you deploy it? Yeah. Resource. What are the resources you're looking for?

Jeremy: Yeah.

Stephen: And how to make sure that you keep your team lean, because- Yeah ... you know, you've seen the opposite of when companies get a lot of fundraising. Oh, absolutely. Yeah. How do you balance all that with, like, the CFO, the CEO-

Jeremy: Yeah

Stephen: and your internal teams?

Jeremy: Yeah. When I joined Zama, I think we had maybe 10 people, and we have 120 now. Wow. So it was, has grown a lot. I'm always super cognizant to try to keep Zama as lean as possible, but at the same time make sure that we have the right skill set in-house.

Stephen: Right.

Jeremy: And of course, like, over the last few years, that skill set has really changed.

Like, in the beginning, Zama's focus was on, um, machine learning and AI use cases, um, because, uh, it's still obviously a hot topic. Yeah. But that's kind of where we thought the go-to-market would be in the beginning. It's, over time, we saw that actually there's a huge need for privacy preserving solutions like FHE on chain.

Stephen: Right.

Jeremy: And so we have migrated towards that use case in particular. So that has meant hiring a whole team of people who are experts in that domain. Right. And then you get into, well, you also need compliance people. Yeah. You also need a legal team. You also need, uh, people with experience in GTM, uh, specifically with crypto projects.

Right. So it's like you need all of these people.

Stephen: How is that FHE space? Can we double-click on that? Yeah, yeah.

Jeremy: Like,

Stephen: is there a lot of experienced professionals in this? A very niche part of a- It's

Jeremy: super

Stephen: niche ... niche of a niche of a niche, right? Yeah. Crypto privacy, and then a certain type of privacy layer.

Yeah. That, and then, you know, the narrative around confidentiality. Yeah. You have to bring in people that are gonna end up being your end

Jeremy: Yeah, I mean, it's FHE in particular, it's, it's relatively easy to find good people- Yeah ... because there are so few of them. It's like they all know each other, you know? Yeah. I think something like 50% of all the FHE researchers in the world work for Zama. Zama. Yeah. At one point we did, like a few years ago, we did a kind of, uh, FHE landscape, and we found that there were, like, 200 researchers who publish on FHE frequently, um, in the whole world.

Right. And so, like, yeah, most of them work at Zama. Um, and the rest, it's, it's easy to find. And Zama has also spearheaded something called fhe.org, which is the community of researchers- Right ... around FHE, and we host a conference every year in a different city to bring all of these 200-ish people, uh, together to present what they've done in the previous year.

And that's really cool because you see everything from really theoretical stuff to stuff which is now been deployed at Google- Right ... at Intel, uh, through Zama, like lots of different things. That's really cool to see. Um, and it-- I have, like, this sense of pride about it because although I'm not a researcher myself, I've been in this space, the FHE space, since it was like a baby.

Stephen: Right.

Jeremy: You know? So that's cool.

Stephen: Do you feel it was a big bet? Zama was making a big bet on FHE, and now it feels- Yeah ... like the industry's finally come to you based on that vision.

Jeremy: Yes and no. Could it- We still have a lot of work to do.

Stephen: Is it more about educating then? 'Cause then your- Yeah ... end customers need to be educated about what F- Yeah

FHE is, why it might be a better confidentially layer. Yeah. And where do you play into, like, things like the Canton Network where-

Jeremy: Yeah ...

Stephen: you know, that you're seeing institutional adoption there because of- Yeah ... privacy, but it's not quite the same privacy that in confidentially that you're providing. Yeah.

Where do you see, how do you interact with, you know?

Jeremy: Yeah. I, I think, um, Canton is a, a similar solution to Zama- Yeah ... but for a slightly different audience.

Stephen: Mm.

Jeremy: I try not to use the word competitor too much- Yeah ... because I think we're doing something different- Yeah, sure ... but similar, you know? Yeah. Um, and we have friends there, so we don't wanna you

Stephen: know.

Jeremy: But yeah, you're right that it's, it's, it's, um, it's a competitor in some sense, right? On the other hand, I think that, uh, to go back to your, to previous question- On the one hand, we wanna do some education still to make people know a little bit more about FHE because it is, it's still a burgeoning technology.

But at the same time, like if, if you're using Zama's tech, the FHE is actually hidden away.

Stephen: Mm.

Jeremy: You-- We designed everything that we do such that even developers don't have to know FHE in order- Right ... to build on Zama-

Stephen: That's

Jeremy: it ... which is really cool. It's like you just have to know Solidity, and then you can just integrate Zama in your Solidity code.

Right. So that's really cool. And then the underlying technology that, so like the kind of base layer of Zama- Yeah ... is built in Rust, and Rust is like a super flexible- Yeah ... easy to learn language. Like even I can code in Rust, and I'm not technical at all. So that's cool because I've been able to follow the Zama tech, uh, the Zama documentation and build like basic, um, basic apps using Rust, which is pretty cool.

So I can build like super easy, uh, yeah, privacy preserving apps myself.

Stephen: How do you attract developers? I think the, you know, you talked about influencer marketing. I think a lot of- Yeah ... tactics from protocols was to, you know, give a lot of tokens- Yeah ... get a lot of developers, and then they jump onto the next project- Yeah

and next project, and your- Yeah ... ecosystem dwindles as, as soon as they can- Yeah ... get rid of those tokens.

Jeremy: Yeah.

Stephen: How do you balance out, you know, giving incentives to your-

Jeremy: Yeah ...

Stephen: your network, building an ecosystem, but also maintaining and retaining them and keeping them- Yeah ... you know, incentivized to contribute to the ecosystem- Yeah

that you've built?

Jeremy: Yeah. Look, I don't wanna be too critical of other projects- No ... but I do think that the airdrop meta is kind of dead at this point. Yeah. It, yeah, you can attract developers, you can attract influencers for like a hot minute.

Stephen: Yeah.

Jeremy: And then- I

Stephen: think that's a perfect term, a hot minute. Yeah, yeah.

Jeremy: And then they like,

Stephen: yeah-

Jeremy: They're

Stephen: on to the next one,

Jeremy: right? They're on to the next thing, you know? Yeah. That's not the kind of community we want to develop. Um, it's not the, it's not the ethos of Zama. Even like internally, we have really, really low turnover rate, um, of employees because people really care about what we're doing.

People buy into the mission of Zama before they buy into the tech. Yeah. That's how it was for me, and I think that's how it is for like most of the people who work at Zama. We care about privacy. We care about confidentiality. Oftentimes because many of us have-- we've been through a data breach, uh, we've had something happen to us which has felt kind of icky when it comes- Yeah

to personal data, and so we wanted to do something about that, and Zama is a way to do that on an institutional level, which feels really good, um-

Stephen: And it doesn't hurt that you put your employees up in this nice villa instead of- I mean, that helps, yeah ... some cheap Airbnb and, you know- Yeah ... they got this beautiful view.

Yeah. They walk down to the beach. That doesn't hurt as well- Yeah ... taking care of your people at a big conference like this. I

Jeremy: think so, yeah. But even the way that we reward- Yeah ... our team and reward developers, it's more incentive-based rather than like, "Here's some tokens-" Yeah ... and like, "Thank you very much," you know?

Like

Stephen: Talk to me about AI, 'cause AI- Yeah ... has played a huge role. Yeah. Doesn't matter what you're building in blockchain.

Jeremy: Absolutely.

Stephen: Heard very little of it in this conversation. Maybe I was only there on the regulation day, where- Yeah ... it was more about tokenization. Yeah,

Jeremy: yeah.

Stephen: Tell me how you're implementing AI, maybe personally or within

Jeremy: Zama.

No, it's a great question. Yeah. So I was a really early user of ChatGPT- Yeah ... just because I like, I like any kind of like deep tech burgeoning thing.

Stephen: Yeah.

Jeremy: So I was like, "How can I use this to just optimize my life?"

Stephen: Yeah.

Jeremy: You know, that was my initial interest in ChatGPT specifically. Uh, now, of course, I'm using lots of other solutions, Claude and so on.

Yeah. At Zama in particular, um, to kind of piggyback on your previous question about, um, what does it look like to keep a company like Zama really lean? Well, AI is one way that we can do that. It's like, how do we optimize our existing team? So it's not about getting rid of people. Yeah. It's like, how do we empower our team to use AI to do their jobs better?

So we started that with, um, how to use AI in our finance- Mm ... marketing, and kind of HR processes because that's sort of a light touch way to do it without touching the code.

Stephen: Yeah. So- Without getting in, you know, those regulated entities don't like to see- Yeah ... too much AI used- No, no, no ... on what they're, what you're building.

Yeah.

Jeremy: No, that's right. Uh, so just internally, like one of the first things we did was to implement a, a kind of a, a help bot for the operations stuff. So it's like, okay, you don't wanna bother the HR team with your question about, you know, vacation days. You can just ask the bot.

Stephen: Yeah.

Jeremy: And what I thought was really cool about that was like when I saw our office manager using it, um, like nothing against her, she's the oldest person in the company.

She's like the mom of the company- Yeah ... you know? Uh, which is really cool. And when I saw that she was embracing it, then I said, "Okay, we've hit on something really good." Yeah. Uh, like Catherine's using it, it must be like-

Stephen: The, the Catherine test. Yeah,

Jeremy: the Catherine test. Yeah. And that means it's super user-friendly.

Uh, it's doing what it needs to do. And from there, we've been able to deploy it like into other aspects of the organization. So now we have this massive push on how do we use AI to do our jobs better and to make sure we're like maximizing our time. So for me, the test is what used to take me maybe eight hours should now take me six or ideally even four, you know?

Right. And that means I have like four more hours to, to do this, to this, this kind of thing- Yeah ... or to, to meet people, to, I don't know, like to sell. To like whatever, you know? Or even if it's just to free up my day to, you know, like work on other projects, that's also okay. Like, as long as we're getting done what we need to get done That's what it's all about, right?

Like, we don't wanna be this kind of like traditional company where you have to clock in at 8:30, clock out at 6:00 PM, wear a suit every day. Like- Yeah ... I started out my career on Wall Street, and that's how it was. Yeah. You know? So that was a great learning experience at, like age 21, but it's not what I want for Azama now, right?

Stephen: If, if a regulated entity- Mm-hmm ... is watching this podcast and- Yeah.

Jeremy: We love you,

Stephen: but... If they're confused or any entity- Yeah ... that you're, you know, selling to is watching this podcast. They're, they don't understand FHE. Yeah. They understand the need for confidentiality. What, what would be your, like explanation to them in like one minute of like, well, how you can service them as an, within Azama ecosystem?

Jeremy: Yeah. Well, I think that, that more and more if you wanna, if you want to move any sort of service on chain, you need to have a level of confidentiality embedded. It's going, it's going to be asked either from the regulator or from the financial institution, or if nothing else, from the consumer, right? Yeah.

Like, there are multiple people, multiple entities now asking for some layer of confidentiality. So there's a market demand for it now. What that looks like in practice is, in a sense, up to you as an organization. You can use a technology like Azama to provide end-to-end encryption without a layer of complexity.

Stephen: Right.

Jeremy: By which I mean your existing team of developers can start to play with Azama right now without having to learn FHE.

Stephen: Right.

Jeremy: So they can go on our website, they can download our documentation, and just play around with really simple use cases. And like one of our really easy and very first use cases that we put on our website was you've got a photo, you don't want everyone to see the photo.

Yeah. You wanna apply like some sort of filter on the photo without the kind of filter technology- Yeah ... let's say it's Instagram, being able to see the photo. How do you do that? So you give it the photo without being able to actually see the photo. It can apply the filter.

Stephen: Right.

Jeremy: And then you get the kind of encrypted photo back.

You can decrypt it on your side.

Stephen: Which was

Jeremy: really- And-

Stephen: That was a cool exercise to go through. Yeah. Well,

Jeremy: you've got this like-

Stephen: The sending, uh, crypto transactions- Yeah ... also very

Jeremy: interesting.

Stephen: Yeah. Yeah.

Jeremy: It's cool because you can just do it. Right. Like, it doesn't require any kind of like technical explanation.

Right. So yeah, for me, it's like if you're interested in, in privacy or confidentiality, oh, just go on Azama's website- Yeah ... and play with these, like really simple use cases. And then the, the door's open for like the use cases that are more complicated in nature, right?

Stephen: And as you said, you can program that confidentiality, which is important, right?

Yes. It's not just a one-click confidentiality- Yeah ... or no confidentiality. No.

Jeremy: Yeah.

Stephen: You have to make it very unique to the use cases that you're servicing.

Jeremy: Yeah.

Stephen: Usually my last question is around the future of the industry.

Jeremy: Yeah.

Stephen: But I want to ask you, how are you future-proofing against things like quantum computing?

Yeah. 'Cause I think that's just as important- Yeah ... of where the future lies, and probably- Yeah ... has the biggest impact- Yeah ... that we're gonna see in this industry.

Jeremy: So the really cool thing is that FHE, not just on this FHE, but FHE in general, is quantum-proof. Uh, so which is really cool. So- Was

Stephen: that by design or it just happened to be that way as they were building

Jeremy: the

Stephen: technology?

Jeremy: Yeah. I think it, it happened to-- Well, I guess it's a bit of both. Yeah. Like, uh, the original inventors of FHE stumbled upon this cryptographic solution, which happened to be quantum resistant.

Stephen: Right.

Jeremy: Um, but now the job is to keep it quantum resistant, right? That's right. So, uh, I mean, with any sort of technology like, uh, like Zama's solution, y-you're relying not just on one thing.

Like FHE is the core of it, but there's, there's a bit of it which is NPC, there's a bit which is... Yeah, lots of different sorts of cryptographic solutions. So when you bring all these things together, you've gotta make sure that they work in such a way that you're maintaining the security of the system.

Right. Right? I, I think Google released like some sort of report yesterday saying that by 2029, they think that quantum computing can break Bitcoin, can break Ethereum. Right. Which is pretty scary. Like, that's around the corner. Right. You know? So you, you are gonna need solutions like Zama, which at least are encrypting the data, uh, that sit on top of, uh, these kind of L, L1- Layer

layers, right? Like layer ones. Um, it doesn't fix the fact that Ethereum itself could be broken, but it gives you already some guidance- Right ... towards what we need to do. So for me, it's like, yeah, the future of the industry is to become quantum resistant. Yeah. Because otherwise, like we're all in trouble.

And to me, that's not even about crypto money. It's even about, like, the data which is held on chain. You know, like, it's one thing to talk about, "Well, I'm gonna lose, like, however many thousands or millions- Yeah ... or whatever, like, of, of token if Ethereum is suddenly, or Bitcoin is suddenly not, uh, no longer, like, quantum resistant, or if it can, if it can be broken by a quantum- Yeah

computer. Um, but even beyond that, it's like, yeah, but what about all the data which is on chain?

Stephen: Yeah.

Jeremy: You know, like, personally identifiable data, company data, payroll data.

Stephen: All the things that we said we needed to put on chain are at risk- Yes. Yes ... when we really didn't need to put them on chain. Yes. Well, but now we put them on chain, they're at risk.

Jeremy: Yeah, but don't-- I don't know about you, but I see the same thing happening with AI right now. Yeah. It's like-

Stephen: Everyone's just feeding these algorithms- Yes ... and- It's

Jeremy: crazy ...

Stephen: it's like, yeah. The, the agents- You know? ... will be great at-

Jeremy: Yeah ...

Stephen: GTM until they give your information over- Yeah ... to the North Korean hackers, and now- Yeah

they know your customer list.

Jeremy: Yeah.

Stephen: They know your technology, your code.

Jeremy: Yeah.

Stephen: It's scary.

Jeremy: And I'm doing it, too. It's like, you know- ... but we're, we're all doing it. Right. But, um, I don't know. It's like, it's a little bit scary. Like, you know, it's like, I'm gonna give, uh, ChatGPT my bank statement and ask it to analyze my spending habits.

Yeah. It's like, should I really be doing that, you know? Yeah. Or should I be using, like, a different solution? And I, I feel like, in a lot of ways, today's Two big technologies, which in my opinion are on-chain, like in general, on-chain stuff and LLMs in general.

Stephen: Yeah.

Jeremy: Sometimes the two overlap and sometimes the two the most completely different technologies, which I think is good and bad.

It's a different topic, but- Yeah ... you know, there is like room for overlap there. Um, the point being that we're giving both of these two, like, buckets of technology, lots and lots of data, and we're not really too concerned about what happens to it. Yeah. And I think that's really scary. Um-

Stephen: Just like we weren't concerned about quantum computing- Yeah

were we? But now we're gon- you know, we're not gonna feel it right now. Now we're starting to think about it. But when we gave the, when we gave the AI all of our personal information- Yeah ... to create a budget, and now we're getting advertisements for everything we buy, we're gonna wonder wh- why did we do that?

Yeah.

Jeremy: Yeah, yeah.

Stephen: And I don't think we're as privacy concerned. I always ask the question to privacy experts like yourself is-

Jeremy: Yeah ...

Stephen: are we that worried about privacy when we're giving all of our information to save 10% at Bed Bath & Beyond? What are your thought- what are your thoughts about society and privacy?

Yeah. I think that's a great way to end the conversation.

Jeremy: Yeah. Well, I mean, you just have to open your Instagram or Facebook, you know, to understand that we have given away, like, almost everything- Yeah ... at this point. Uh, you know, it's, it's difficult because I'm also a super commercially minded person. So it's like, you know, I can see why companies want our information- Right

and what they can do with it. But you can do, like, responsible advertising with data using technology like FHE, right? Yeah. Again, it be- it comes back to that, like, selective programmability, selective confidentiality. You know, so long as, as an organization can get what they want from the data to be able to serve you those ads, they don't actually have to know anything about who it is that those- Right

that, that data belongs to. It's like, okay, they just know that it's Jeremy who, I don't know, likes to shop at Saint Laurent. They don't have to know that it's, like, actually me, Jeremy. Yeah. You know? It could just be like- Yeah ... this random 40-year-old guy who lives in Paris, right? Like, that's where I think you need to be able to figure out which data is actually absolutely necessary for doing what businesses wanna do with our data, and which bus- bits of it are just, like, we're giving away for free for no reason a- whatsoever.

Awesome. That's what's scary.

Stephen: Awesome. I

Jeremy: think

Stephen: leaving people in fear at the end of this podcast- Yeah, sorry ... is a great, is a great way to end, actually. No, yeah, yeah. Get them to look at, actually try out the technology for themselves. Yeah, absolutely. Jeremy, thank you so much- No, thank you ... for today's conversation.

‍