Okay. Yeah, my name is Alex. I'm lead of cryptography at distributed lab and RMO protocol. Yeah, I started doing cryptography 10 years ago. Uh I like finished at like master degree in cyber security and then switch it like to applying cryptography in real applications.
Shortly that's it. We are trying to help different projects like Raro and other ones uh to use zero knowledge proof and cryptography in their solutions. Uh how Raro started we started as like as identity aggregator for verifiable credentials. Uh it started two years two years ago. Uh then we found out that you can bring some real documents and real web two and web one credentials into web3 world.
So for example, you can take a passport, scan all data from that and generate appropriate proofs that some data is included into the passport but without reveal and sensitive information. For example, with this approach, you can prove that you're Ukrainian, you're adult Ukrainian and for example vote. Uh but all the data you don't want to share uh will be hidden. All proofs are generated locally. So you are not delegating any data to some I don't know trusted executors or trusted provers.
So you are generating gross 16 proof locally and sharing only the final proof with needed application. Uh yeah then we launched several solutions that allow you to organize a digital protest with this technology. I mentioned you can participate in some pools uh and uh yeah prove that you are eligible to participate but without again extracting some sensitive information. Uh right now we are working on like uh we are trying to extend this concept with uh identity data with like probabilistic identity data. Right now with our technologies you can scan your face and creates the proof of proximity for example or proof of likeness that you are the same person that controls a particular passport.
Uh again uh we are touching like the KML a bit. So we need like to create a provable environment for machine learning and AI protocols uh that like you should be able to verify everything uh that the particular AI or model doing. Um yeah that's that the design of RO protocol. Yeah privacy should be right. Uh so why I'm interested in this technology because yeah of course I started with Bitcoin but the next technology that I was wondering is uh Mona and Mimblewimble then and it's like an application that are combining cryptography uh to like extract the needed properties for people.
Uh with such applications you can pay to anyone without like any restrictions with like hidden your personality and like keeping your human rights. Uh we are trying like to use the same technologies but for more social for like more social oriented applications. Yeah. probably that's that's how uh mass and cryptography can solve some political problems, some social problems. Uh because I believe this this should be like a core component.
Uh yeah, privacy should be a core component. So you cannot operate in current world totally openly. That's the reason why a lot of companies cannot use USDC in current form because you cannot open your relations uh and your interactions with your clients with other businesses. uh you need to keep the privacy but at the same time extend that with eligibility. So if you are allowed to perform some applica if you're allowed to perform some operation uh in hidden form uh but with proven your eligibility you should yeah you should be able to do that.
Um the technology should be hidden like from people as blockchain like a radio waves nobody knows how exactly they are working yeah a majority of people but we are using them every day I believe the privacy should be the same. So when you're using some specific application you should be protected but at the same time you don't need like to investigate how zero knowledge proofs work how different how r signatures work how like different like confidential stuff work. So um the biggest challenge is to combine um a good user experience with hard technologies under the hood. Uh what our challenge for example that some statements require a lot of computational resources and for example proven different statements can take several minutes for on client devices and definitely that's not good for user experience. uh if some application requires waiting I don't know several minutes to complete the operation definitely that's not for people um and the biggest challenge is to match current technologies uh like to real applications I know that a lot of protocols and web three infrastructures they're working now on privacy preserving applications uh and yeah I believe in this future that in in five or in 10 years the majority of the five applications applications will be regulated but they should be regulated uh in the proper way.
Uh they should be regulated in the way when user can keep privacy but at the same time prove that they they're allowed to perform something relating to passports. Uh problem number one uh we have like a common standard for passports but um the implementations are different from country to country and if you are working with zero knowledge groups you need like to have ideally you need to have some constant algorithm for verification of everything and if we are working with different like implementations of passwords in different countries we need to reimplement logic uh they are using different signature algorithms with different hash functions and uh yeah from and when you're operating with constant circuits that's a headache. uh additionally some of them support very difficult cryptography and if you're working with the K you uh yeah you need like to cover this difficult algorithm with additional like constraining etc etc etc and uh all users have limitations depending on their client devices uh and it means that you're not able right know how to generate all proofs locally. So, and there is a trade-off what exactly you can delegate to some trusted server. Uh what exactly you need to hide and what exactly you need to control uh on your local device.
Uh as I said, we also working on the intersection of the K and AI and um good models. they're also computationally difficult and when you're trying like to cover them with a key it means uh you're adding requirements for the device where exactly this algorithm will will be conducting uh so yeah this like the biggest one yeah I respect to teams who we which are trying like to reduce the complexity of writing zero knowledge applications for developers for example I don't know a team and no uh because sircom is a like this like technology for gigs. Uh and yeah, if you want to like to to bring like more people and more developers to the K applications definitely you need to create appropriate frameworks and appropriate SDKs and tool base for them. Uh and like yeah we have the same problem for identity cases. uh because um I don't believe that developers will investigate how our technology works the majority of them of course uh so we need to provide a clear SDK a clear tool base uh and probably documentation uh to allow them to build their identifications like in in an easy way uh and probably yeah that's what we are doing right now we should like uh open source our frame may work for Z the K friendly AI uh until the end of May.
Uh so and then we can yeah we can gather the feedback of how how difficult it was for point of view cy punks are the people who are fighting right now who are fighting for digital uh digital privacy because uh Yes, I said that like
we want like to have a privacy in the physical world and u yeah this sometimes is difficult to achieve but at least if you're if you would ask people do you want to like to uh to be private um the majority of them will answer yes but in reality um
nobody cares nobody cares in the digital world internet remembers everything. Uh and uh early or later uh we we will need like to find with results uh with results of public applications. Uh if we can design like a private internet and private like web three global infrastructure, it means we care about the future about the like humanity future. Um and yeah, cyber punks are trying like to implement this future for like for all of us. I think like yeah there is no single instruction for how to be a cipher product.
Uh and probably there are different ways. First of all, yeah of course you can learn m and cryptography and like implement some applications. Uh then like Propaganda also matters. So you if you're fighting for privacy, you can be like a public uh a public person and spread like yeah you should use like privacy oriented applications for example. Sooner or later u more and more people should start uh using like such application as signal instead of telegram for example is like a clear choice.
uh and uh but yeah we we need to have like a strong uh ecosystem for that. We need to have privacy preser preserving applications with a good UIX first of all because uh initially people uh see on user experience and only then only if some problems happen they uh they can yeah pay attention on the privacy. Uh so probably yeah the solution is if we have like a strong and uh user friendly applications uh that presumes that presume privacy we can reach this like uh we can reach the dream of cyber funks. Again there is no like general advice for developers. It's more like related to owner of applications.
Uh
yeah this like advice for them not to forget about privacy. So when you're designing your final product when you're implementing some like new additional features uh yeah of course first of all you need to think about like how the final user will like consume your application but at the same time like a strong recommendation not to forget about the privacy. Yeah, I I would highly recommend uh everybody to pay attention on applications that uh achieve a privacy through cryptography and mass because mass is beautiful. Uh this like probably uh for me it was a way how I um how I started working with crypto that's like Monero and Mimblewimble. Um you can yeah you can see how mass creates some properties uh you can achieve like in private applications uh and uh it's not possible like to achieve that with trusted entities with like trusted mediators with some responsible parties because sooner or later the responsible party uh will reveal your data uh this responsible party can deny of of the service and you cannot like use your application.
So um internet and like application stuff on that should be built in the manner where they yeah in ideal world they should be decentralized uh they should be permissionless uh and they should support privacy by default. uh only like with this set of requirements uh the user can can be safe in in the current world. Yeah, there are two like basic books from Bruce Schneider and Vanbal. They're like quite general, but anyway, if you want to like to have a quick introduction into cryptography, I would highly recommend to to read them. If you want to be a cryptography guru, you should read you you should read uh a graduate course to applied cryptography by uh Den Bon and Victor Shup.
So probably yeah if you if you're able to read this book um you yeah you could understand a lot of about like cryptography and how privacy can be reached as a cryptographer uh I would recommend you to read to try like to read every paper you can find uh because first of all like yeah that's like the building blocks for the privacy technologies in the future and it can provide you some ideas you didn't expect that you didn't expect to have. I wouldn't say there is like one single website that aggregates everything. So probably year print and uh Eric are the best sources. Uh and uh yeah there are like a lot of different journals focused on like one specific on like another specific areas. Yeah, like a lot of different sources where you can find cryptography pieces.
Uh, web three privacy. Now, this is a good source for that. Um, and yeah, I would start investigating protocols that brings privacy into like current world. Uh, from internet perspective, yeah, that's like yeah, tourp. Uh yeah, you shouldn't like know all details of the protocol but like some basic principles should be enough to understand like how how how it's constructed uh from like uh the centralized applications.
Yeah, cash has a very beautiful construction. Monimble vimble the cash as well. Yeah, for messengers signal first of all uh some messengers should support broadcast encryption uh matrix for example. Yeah, matrix has a different structure but you can instigate that as well.
Automatic transcript — names and jargon may be misspelled.