 All right, everyone, thank you for coming on time. I, hopefully, people will trickle in, but I think in interest of time, we should sort of kick off. So what I'm going to talk about, as the title says, blockchain, biometrics, and geolocation. This is a case study of the work we have been doing as UN, and I'll introduce myself in a bit more detail at the UN, what the project is, and some of our experiences in actually doing this project. So just a quick introduction about myself. I'm Shashank Rai. I'm the Chief Technology Officer for an organization called the UNICC. And another gentleman who generally presents along with me when we talk about this project. Mr. Dino Delaccio is not here with us. Due to personal reasons, he couldn't come over here, but he is the CIO of the Joint Staff Pension Fund, UNJSPF. So in a way, he's my client as we serve the UN ecosystem, and we built out the project for him. Since we're going to talk about digital identities a bit, I always like to start off with this cartoon from Peter Steiner, which was published in the New Yorker magazine in 1993, which I think is still very valid. And sort of as it says, on internet, nobody knows you're a dog. And unfortunately, in 2022, it's still valid. On internet, nobody knows you're a dog. Right. So I'll do a quick introduction as to who we are, what is the Joint Staff Pension Fund, who is what is United Nations International Computing Center, then actually go down into the business challenge. What is it for you looking to solve? How did we solve it? Some of the lessons learned, which is the essence of this talk. And then my favorite question, which I always like to ask as to why use blockchain. And in fact, over the last one and a half days that we've been around here, I have been going and asking the people the same question, why blockchain? And I'll be honest with you, in 50% of the cases, I never really got a solid answer as to why blockchain. And as my technical team likes to put it, we are then just putting up very expensive databases. Right. So quickly about the staff pension fund. Staff pension fund was established in 1949. I'm sure most of you know the UN General Assembly. That's the big tall building in New York. They do need to discuss very important stuff, but they also discuss a lot of administrative things. And other administrative things is forming bodies like the joint staff pension fund or International Computing Center. The pension fund, as the name dictates, they are responsible for a pension pot. So obviously, that the pension fund operates from an IT point of view, at least efficiently, is in my personal interest as well. Because obviously, I like to withdraw my pension from them when I happily retire. Today, they serve about 25 odd UN entities. And for those of you who might not be deeply familiar with the UN ecosystem, as I like to sort of joke about, UN is united in name only. It's quite a multilateral organization itself. For example, you see on the right the names of the 25 UN entities that the staff pension fund supports. But UN is far larger. Some of the names you will know, like UNICEF or UNESCO. These are household names. Others you might have heard of, like UNHCR, High Commission for Refugees, or the World Food Program. But they are ones that you might not know if you're not in the industry like UNCTAR. They establish standards on trade. Or UNIDO, which helps industries in developing nations, et cetera. When we talk about a pension fund, obviously the size of the kitty or the pot matters. So pension fund handles close to $91.5 billion of assets. Today there are about 137,000 people registered with the pension fund. That's all the staff of those 25 organizations. They're contributing into the pension fund. And there are about 82,000 beneficiaries. So people who have retired out of the UN ecosystem and the pension fund actually serves those people. So the people who are withdrawing the pension from the UN ecosystem, that's about 82,000 people. From their geographical location point of view, they're pretty much spread across the globe. So you can think of a country, and there'll be somebody from UN who's retired over there. And for this, 82,000 people, the fund is disembarking roughly close to $3 billion worth of disembarkment per year. Quick word about us. Again, we were established by the UN General Assembly but in 1970, so we as International Computing Center are about 50 or years old. Administratively, for whatever difference it makes in today's presentation, we are part of the World Health Organization. So I'm a WHO staff member. And our mandate as International Computing Center is to really provide digital services to UN organizations, to other international organizations, World Bank, Asian Development Bank, sort of entities which are UN-like, as well as member states, which is just a fancy term for countries. So if a national government comes to us looking for some mighty help, we do that. So that's the context of who we are and what the pension fund is, et cetera. So I'd like to move into the business problem. What is it that we were trying to solve? What is it that the pension fund faced and came to us to talk about? So every year, those 82,000 people that I mentioned, they need to prove to the pension fund that they are alive and they are living at the address they promised to live at, specifically the country. And the country is important because the pension, if they may choose to do so, to receive the pension in local currency, the pension is slightly adjusted based on exchange rate and other factors. So obviously, if you promise to live in a particular country, it is important and the fund knows you continue to live in that country so that they can continue giving you adjusted pension. Now, the way the pension fund has been doing it for the last 70 years of their existence is every year they send out a letter to these 82,000 people. They own their registered address and they wait for those letters to come back with the signature of the individual. And then based on that, it is basically a verification that the person is alive because they try and match the signatures. They assume they are living on the same address because they posted a letter to an address and they got a response back and they continue issuing the pension. Needless to say, none of us have to sort of think about how long this process is, how inefficient this process is and the challenges it brings. And not to mention the fact that the fund has always been challenged by auditors to try and quantify potential fraud that may take place and is always very difficult. Obviously, if somebody continues signing the letter when they are 150 year old, I think that's a dead giveaway that they may not not be alive but somewhere around 85, 1995, it starts becoming challenging and they try and scrutinize the signatures more, what not, but then at that age, people might be incapacitated so they might not be able to sign and it gets far more complicated. So that was the business challenge. Is there some way, can we bring in modern technology to solve this problem? Can we make this more efficient? So the fund, as I said, the role of ICC is to provide them technical solutions. The fund, Dino and the team came over to us. We had had a lot of discussion as to how can we explore new technologies to really help them out in this area. However, again, one of my standard jokes is if you've worked with one government, you know how much red tape is them and how slow the bureaucracy is. Multiply that by 200 governments that we work with. So it's an exponential factor. My point is there is a lot of audit, there's a lot of accountability that has to take place and rightly so because we're spending public money, we have to be accountable for that. And again, when we're dealing with some sensitive information such as people's personal information about the eligibility for payments, there were certain key guiding principles the fund laid down. And one of the key ones was no matter what technology we use, we have to make sure we are taking good care of the personally identifiable information. Now just a point to note over here, from a legal point of view, UN is not bound by any national law. That's how UN was created back in the days when League of Nations was converted into United Nations. All the signatory countries said, UN is not bound by the national law. What that practically means, things like GDPR don't apply to us. Now having said that, it doesn't mean we will play willy-nilly about people's privacy and information. So the course of last few years, UN has drafted UN privacy principles and these are sort of the guiding commandments for any solution we work on that will comply with these UN privacy principles. So, but obviously any technology solution we would look at for the pension fund had to comply with the privacy principles. As I mentioned, strong audit trails. We're talking about transforming a 70-odd-year-old process. It's deeply embedded into everybody's mind. This is how it works. This is how we track it. 70 years of audit by different national bodies because we're generally audited by national governments. They send their national audit offices to come and audit us. So it's all well laid out, what they will ask, what they won't ask and we're not trying to transform that to business process. So audit trails was very important. And finally, obviously we can't sort of go around playing with technology where it is not sustainable and it is in some way, shape or form, contradictory to the sustainable development goals of UN. So with those guiding principles and the discussions that started with the pension fund, I think we all just said blockchain is the answer as long as you have got the right question, right? That is how we all, and as I sort of started this conversation with some of you when we started my presentation, which is why blockchain? And even in the last one and a half days when I've had this question with a number of people just over here, I am not convincing 50% of the cases that they need blockchain. So our point was no, blockchain is not that hammer and we're not going to look for a nail. There has to be a reason why we should have blockchain in the solution. And there's quite one I'll try and answer myself at the tail end of my presentation. So what we did from a solution point of view is we built a mobile app. The idea of the mobile app was that the beneficiary, the people who are receiving the pension would undergo an enrollment process. They would actually use the phone and they'll try and take a selfie and that selfie will basically be used to build their biometric profile. The important bit was that this biometric measurement of the face and the resulting biometric data is stored locally on the phone because we did not want anything going out of the local phone. Then they undergo an enrollment verification process. I'll talk a bit about this in my next slide. And once this enrollment verification process is complete, they actually use the same mobile app. So the pension fund prompts them saying, hey, it's time you prove that you're alive and sort of living in the country you promised us to live in. Can you use the app and prove that? So they open up the app. They again take a selfie. But while they're taking the selfie, the app prompts them to change their facial features. So it'll ask them to smile. It'll read if their eyes are blinking, et cetera. Why is it doing it? Because first it's taking the biometric profile of the face. It's matching with the biometric profile that's been verified on the mobile phone. It's doing the liveness check to make sure nobody, your loved ones are not holding your photograph in front of the camera and claiming your pension. And obviously it's trapping the location based on the GPS of the phone. And again, on technology side, we had a lot of discussions on GPS spoofing and how it is avoided. And it's a cat and a mouse game and we are in that cat and a mouse race as well. But broadly, this is the three things we do. And through this process, what you see, for example, in the image over here is the person goes through that whole process and they get a valid DCE and then they are eligible for their pension for next one year. That's a business process. Now if some of you are wondering, why am I here at a hyperledger forum and talking about this whole thing when there is no blockchain involved, the answer is, well, there is some blockchain involved. The underlying platform for all of this that we built was using Aries and Indy. So we do have a UN-Indy chain running. And again, we've had different discussions both to a coffee and in different forums about the need for a sort of chain. In that sense, should it be a public chain or whatnot? I think from a governance point of view, there is no way I can see. Just as I mentioned, the privilege of the immunities, there is no way, for example, UN can start embarking and doing business on public chains where national laws apply. So we built an Indy chain. The wallets are in the cloud. These are all user wallets. So obviously we, in that sense, control them. So we open up those wallets and we write the transactions into the wallet. But what's getting written into this is basically three sets of transactions. The first credential that we write into it, into the wallet, sorry, I should say credentials. The first credential that we write into the wallet is basically once the verification is complete, the enrollment verification, that this is the person who they claim to be. Then once they undergo the liveness check, we write the second credential that the liveness check is complete. And finally, we tell the third credential, which is the DC has been issued, which obviously has an expires because then they have to get the new DC done. So that in a very short essence is the solution. A bit on the enrollment side and the fun and joy that pandemic bought along with it in the last couple of years. So while all of this was developed, and I'll talk about the timelines in my next slide, the intent with this project started was that once a beneficiary has downloaded the app, they would actually be in touch with somebody at UN because each UN agency has a number of staff which looks after their beneficiaries, the retirees. And the idea was they will be in personal contact with these officers of the UN or WordFood program or whatever organization they belong to so that the officer can actually sort of certify that, yes, this is the individual who has enrolled into the system. And based on that certification, they can then go and produce the DC, et cetera. But obviously when the pandemic helped, the best laid plans failed and we had to quickly scramble and try and figure out how can we sort of avoid the pandemic bottleneck or the hurdle the pandemic had placed. So what we did, or well, when I say we, it's more the smart people who work for me, what they did was they adapted the app to build and they built in a scheduling system in the app. So sort of you see that little screenshot over there which is in French because it's a multilingual app. You choose a time and date of your preference and you book a call and then we built the whole system whereby the pension fund has a service desk, a help desk who get on to a call with the beneficiary on the pointed hour and they do sort of a verification where they're looking at this video feed of the person. The person also scans a government issued document photo ID. So they're looking at that document on their screen and they're sort of saying, yes, we looked at the document ID, we looked at the details, we spoke with this person, it's the individual who they claim to be and they approve that enrollment. And this is important because the distinction we like to make in this solution is that we are not at this stage of the solution taking what the person is claiming to be on the face value. There is human in the loop who is doing a degree of verification. Now obviously it's not the most optimal solution we would like to, for it to scale, we would like to bring in a degree of automation in some other projects. As UNICC, we have done that just to sort of sidebar mention when COP26 took place in Glasgow last year, the climate change conference because it was a hybrid conference and they had to resist a lot of people coming in. We took this solution and we used machine learning algorithms to actually scan the government issued ID and do a face comparison of the selfie so that the onboarding of the users could take this quickly for the COP26 conference. So we now want to bring some of those automations over here, but at this stage, this is basically a human in the loop verification of the person. Right, from a timeline point of view, the sort of the idea, initial idea started in about mid-2018. Dino had just taken over as the CI of the pension fund. Obviously this was one of the biggest challenge the organization had been facing. So that is when the discussions all started and then when that whole discussion started, obviously the technical experimenting from the teams at ICC started because none of us knew what Indy was, none of us knew what blockchain, I mean, we all had heard blockchain and I think that was one of the peak things where I think every techie was investing in crypto at that stage. I think it was 2018, 2019 when Bitcoin had first hit 20K or something. So we were all quite excited, looking to retire early maybe without the pension funds pensions, but our dreams are slightly dashed. Anyways, back to the topic. So that's when we started experimenting with the technology, trying to understand how it all works, what are the intricacies of it. Then in Q1, Q2, we started building some POCs, proof of concepts, both on the mobile side and the technical challenge obviously was to do all that liveness and biometric in mobile itself as well as trying to understand Indy and then obviously Indy pulled a fast one on us. Suddenly you had Indy and Aries, so we had to scratch our head around that and work around with Ekapai. That's fine, I'm not complaining here. Then we moved into a business pilot and obviously as the techies in the room will know, a business pilot means IT's in production, so it was really a production network for us and there's an implication of that I'll talk about in lessons learned. So around 2019, tail end of 2019, we moved into a business pilot. The idea was in 2020, we would go into a full, what do you call, production deployment. But as I said, we are UN, we would like to go through our own channels. So most of 2020 was really spent in getting the right set of approvals and I think that was important. While I joke about it, obviously, it's a very transformative process and it needs to be well understood. There's a lot of conversation about why blockchain and what blockchain, I'll touch upon that also in lessons learned. Probably it was not best on our part to sort of use the word blockchain. Maybe we should have just said, we just distributed databases, nothing to see here. But it went to the pension funds board, they were quite actually excited about it. And I must again call out the CEO of the pension fund, Rosemary, she was quite into it. When she had just taken over the pension fund and she was like, this is where we are transforming. And obviously, I think we've all sort of seen that meme that what brought about digital transformation in your company? So it was not the CEO, it was not the CIO, the CTO, it was COVID. So we were at the peak of pandemic lockdowns et cetera at this stage and the fund was scrambling to go paperless because historically there was a lot of paper that moved through. So we got the approvals in 2020 and then in January 2021, we went live with the whole solution. And since then, the solution has also gone a lot of external audits. We got ISO 27001 certified earlier this year. So obviously we went through that whole process of certification. And again, that was an important one because we wanted to demonstrate. And when I say we, I'm speaking of both the pension fund and ourselves that while this is new and innovative and it might see that we're the techies trying to bring in glitzy toys to play with, it is mature, stable and proven. We are right now, it's work in progress working on ISO 27001 certification. For those of you who don't know, I mean, I don't know if you know 27001, that's the internationally recognized certification for information security management. Which sort of just goes on to say that for this solution, we are trying to follow good practices. We wash our hands before we eat our food kind of stuff. And 27001 is the certification for data privacy. So again, we want to prove, again as I said, none of these national laws apply to us, GDPR doesn't apply, but we still want to prove that we are very cognizant of the sensitivity of the data and we are treating it carefully. So that's the progress. As it stands, this is sort of the user distribution across the globe. I think you only find a couple of gray spots there. That's North Korea. And again, sort of, I'm not sure which country is that, but otherwise, yeah, we have basically users all across the globe, all sorts of connectivities, all sorts of mobile devices. And it has been a very interesting experience sort of trying to roll this out across the globe. Lovely. So lessons learned. What did we all learn around all of this? I, well, I wasn't sure this is business tracks. I didn't want to bring a lot of tech stuff, but if you want to please do to my, whatever extent possible, I'll try and answer some of the tech questions, but I divided the lessons learned into two parts, the business side of it and the tech side of it. From a business side of it, really nobody cares, except the fact that the moment we set blockchain, there was a huge perception that built up that this is something we are using which is destroying the planet because everybody associates blockchain with Bitcoin and everybody knows how much energy Bitcoin networks consume. So there was a massive backlash and it was a tremendous effort on part of the communication department of the pension fund to really try and reach out and explain this. In fact, at the bottom, you see the link for their, the FAQs for the public face of the pension fund. There's like still a lot of material around. No, this is not energy consuming. And we tried and also explained this because practically speaking, we have this running in three data centers across the Atlantic. Those are all UN data centers because of again the UN privileges and immunities, UN laws. So from a practical point of view, they're like I would say six hardware machines, but then you have layers of abstraction because of the VMs and then the pods in the VMs, et cetera. So really from an energy point of view, basically from a hardware point of view looking at six physical machines doing this. So that was one big challenge. And the second big challenge, everybody has heard that blockchain is immutable. So once you write on to it, you cannot change it. I generally also try and ask, what about the 51% attack on Bitcoin network? And then most of the people who talk about immutability do not understand, which might be a lot of people around in this conference as well. But we had to sort of go again on the war footing to explain that no personal information is being returned in the blockchain. The way the solution is designed, actually all the information is being held in your wallet. Well, in this particular case, obviously Pension Fund already owns the information. So there is nothing new Pension Fund is asking of you. The only new element in this is your biometric model, but we are not taking it away from your phone. So it's in your phone, it's securely put in your phone and it's with you all the time. But again, those were sort of the perception issues that we had to deal with. They were obviously the adaptation problems. So we realized everybody tries to take selfie this way. And this is not a best angle to get a biometric profile. So really to try and get the selfie this way and you're looking at an average age of 50 plus for our consumer over here was a difficult piece. Again, we tried to build in a lot of cues in the app to sort of prompt them to move the camera up and take good selfies. Now we have adapted the app to also try and detect if the lighting conditions are good and try and prompt them to have lighting conditions. People had their family photographs behind them and then you had like all sorts of biometric profiles being built off the grandkids and everybody, you know, everything being done. So those were interesting challenges. Device diversity. So I'm a big Android fan because as a power user I feel like I can do a lot of stuff on my Android. But as an implementer I just love iPhone because then you know it's a homogeneous system. The kind of and some very high end, very expensive devices from some big manufacturers and we run into all sorts of problems with native code on that. So that was a big challenge for the team. As I said, for the biometric piece we were very careful on security. So most modern phones, so I would say after 2017, have a hardware chip on that that generates a pair of keys for you. So each app can use that hardware chip to generate a set of public-private keys and then use that to encrypt and store information locally. But we use that feature because we use it to encrypt the biometric information on the local phone. But what that implies is that any phone that is older than 2017, we cannot be used. And that did bring its own set of challenge on itself. And then finally, as I said, the safety concern for biometric and we again had to do a lot of user information sessions et cetera to sort of explain this. But if you sort of look at the user view of it as I said, apart from that blockchain concern, nobody actually knew there's a blockchain, they don't care there's a blockchain. In my humble opinion, that is how a decent blockchain solution should be. You don't want to know which knife was used to cut the onions when you walk into a restaurant order of food. And to my mind, this is just back kitchen stuff. From a technical point of view, remember I mentioned the business pilot and then we went to production. So obviously in the business pilot, we didn't want to invest in a lot of infrastructure, logically so. So we had just the two data centers in Europe running itself. Now we wanted to build up the third node resiliency across the continent in US. And we found, obviously, the seed nodes had the IP addresses hard-coded and whatnot you have it. And really to expand that across the Atlantic was a big challenge. But again, I'm blessed with a very smart set of technical people who work for the team so they were able to expand. A lot of help from the community, we were able to extend that out. So that was one piece of lesson learned. We had some challenges with the Kappa scaling because we were using the SQLite. Then we moved to Postgre as a back-end for ACAPI. That has obviously helped. But that brings a different set of complexities because now the Postgre has to be replicated and whatnot across multiple data centers for this to really work on our side. On the development side, obviously when all of this started, we were just a team of five, six people playing around with different things. The solution is scaling up. I'll talk about how this is scaling up within UN. So we're onboarding more people and it's a bit of a challenge. Even for the most seasoned veterans on technology, we get it's a bit of a learning curve for them to understand what's this whole business about ACAPI, how it is working, and in the blockchain bit. So that's a bit of a challenge from what you call manpower or person power point of view. As I mentioned, the wallets are still running in our data centers for the moment. Our intention is to push the wallets out. In a separate set of projects, we have now made mobile apps with the wallet in the mobile app. And there, again, the last one year where the JavaScript library came in to really help build the mobile apps, it's a breeze. I know some of the people from the team are online and they would really love to thank everybody and anybody who's involved with that JavaScript framework for the wallet in the mobile. That's been a real blessing. But as I mentioned, the JSP of wallets are still in the cloud, but our intention is to push it out. Right. So the famous question, why just not a regular database? And actually, half my team was thinking I'm crazy when we started talking about using this because they said, we can just do this in Postgres, and life will be so easy. And we all know how this all works, why all the complication of blockchain, et cetera. And I'd say two reasons. And with a bit of hindsight, obviously, I'll claim this saintly divine idea that was used at that stage to bring in blockchain, which is now evolving within UN. But there was a degree of thinking into this. But one of the primary drivers for the pension fund was immutable records. Because it was so transformative, pension fund really wanted to make sure that as these transactions take place, they're recorded immutably. Now, traditionally, when we have looked at, let's say, the old school databases, if I may say so, you put solutions on top to make them immutable. In terms of really having immutable database engines, technology in 2018 was not so advanced. I think QLDB, for example, on Amazon came in 2019. And I think that's an amazing solution. There are now various very mature open source implementations for immutable databases. And I genuinely think 50% of the cases where I hear blockchain, probably they need an immutable database. And that's not my words. Those are the words of now CEO of Amazon and DGC. He sort of said that when they launched QLDB, do you really need a blockchain or not think about it? But we did not have immutable databases, so we were looking at solutions and we looked at sovereign and that made sense. But more important, another aspect which has been a bit of a Achilles' heel for the pension fund is the user information. Because of those 25 different union organizations they're serving, each of them have their own ERP, each of them have their own corporate system. And within UN, we jump around different organizations. It's very difficult for the fund to keep track of the actual user data. And that's a big administrative overhead for them. So with that in mind, we already started looking at, hey, if this product is successful, if we are able to bring out a digital identity solution, then it might be possible for us to really expand it and then have a UN ecosystem solution which will allow transfer and exchange of data. And that was sort of the eye on the future bit. And I'm quite happy to say actually, and I'll play a little bit of video over here, that now at the real highest levels within the UN, there has been a consensus that's reached to really roll this, expand this further out into what is called the UN Digital ID Program. And we have got the World Food Program, the UNICEF, UNHCR, High Commission for Refugees, the UN Secretariat, as it keeps saying, the big building in New York, the Pension Fund itself, and UNDP United Nations Development Program, the six big organizations. And that's roughly good 80% of, probably 60% of UN staff, UN workforce. So you're looking at about 130,000, 150,000 odd people to roll this out as a digital ID for the UN staff, a UN workforce, I should say. So not just staff, but consultants, volunteers, everybody. So the program has been conceived. It's been funded by these multiple organizations. And this is really where the blockchain, the decentralized nature of this solution is helping because now we can sort of go out to these organizations and say, hey, you know what? You go define your schema because there have been attempts in the past using traditional technologies for this to work and it doesn't scale because one of them is using SAP, the other one is using Oracle, the third one is using something else. And then you say, okay, I need to exchange data with your ERP and here is my API and points, how do we do that and all that hodgepodge? And then you bring PKI in and it's a technical mess. So what we're saying is we will build, using indeed the data exchange platform. You manage your schema lifecycle, you manage your credential lifecycle, and I know trust over IP is not mature yet. So if we're doing a hack with the central registry of our own for the moment, but if as a consumer of a credential, if I'm verifying somebody, I can then basically say, okay, you say you are part of WFP, please provide me these credentials and the user being in between using the UN Digital ID wallet can give those credentials to the verifying agency. So it took again a bit of a process within the UN to get over here over the course of 2021. This was the idea was socialized. We had a number of internal UN hackathons, et cetera, where we showed and made this how this is working. And as a result, this project is now going ahead. And from a business point of view, I think this sort of video really gives us an idea as to what this is. Digital transformation is changing the way we manage our data, our information, our interactions and our identities online. The United Nations is ready to digitally transform how it deals with identity with a system to streamline information sharing, daily workflows, access to platforms and buildings, operating across agencies by providing its personnel with a universal system-wide identity solution. Introducing the UN Digital ID, a unique and digital identity for UN personnel from the day you join to the day you part. All of your personal, HR, medical, travel, security, payroll and pension data in the palm of your hand, giving you full control on what you share and with whom. With blockchain and biometrics, the UN Digital ID makes verification efficient, secure, transparent, immutable, portable and universal. It's been piloted by different agencies and the UN Pension Fund, where they've replaced current manual processes with certainty for who and where pension recipients say they are at any given time. Imagine a regional field officers just joined the UN. She uses the mobile app to obtain a digital wallet, stored securely in her smartphone and only accessible to her with biometrics. Even better than a physical wallet, she can store all her credentials issued by any UN organization in her digital wallet. She has immediate access to course certificates, travel clearances from UN DSS, medical records from allergies to vaccinations, also making any transfer to another organization a breeze. As innovation transforms the world, we can improve the way we manage our identities online. UN Digital IDs, a building block for digital cooperation, unlocking the promise of the SDGs. And you see some of the names over there and as I said, other agencies have joined in. Digital transfer. We've got that, thank you very much. And now the tricky bit, to move to the next slide, bear with me. There we are, right. And my just last bit, a word of thank you and then we can certainly take questions from the audience, but clearly all the app users, they've been through a bit of a journey with us. From the very initial app, where we had this mask of Zoro kind of a band to tell them where their eyes should be to a bit more user friendly now. Pension Fund staff has been amazing. They've kept up with this and obviously there's the whole learning aspect for them as well. I'd like to shout out to the BCGov people. A lot of work that you had done on your initial implementation of Indian Aries. We were heavily influenced by that. So thank you very much for all that. The community over here again, as I like to say I'm sort of presenting here by standing on the shoulders of giants. So a lot of people have done a lot of work and some of my amazing talented team who I think are online and looking at this. So thank you for to them as well. With that, I'm open to questions if you have any. Please. So a couple of comments on the video. Our services card has the same sort of, started the same sort of way with live in person, scheduled online and then they transitioned because of COVID they had to go one step beyond where you were and they went to a filmed interaction doing the same sort of things. Look up, look down, smile and then having a human look at it and that was enough human interaction to allow for a reliable verification. So none of this holding a picture and you know, moving around and stuff like that. So anyway, just wanted to pass that on and that was just, they were doing, I can't remember, they got it down to like 35 seconds of a transaction and they can do 20,000 in a day I think it is or something like that. Yeah, so anyway, just so you know. You said there was a lively as check when they use the wallet now? The app. Yeah. Yes. So the intention is, what the way it works is you get a notification saying you need to get your DCE. You open up the app, punch in the pin what not. You hold it in front of you. It basically, we boiled it down to again something like 15, 20 seconds, even less. I'm talking about worst case scenarios. It basically does the facial biometric, compares it with the biometric profile that was previously created and parallely looks at liveness. So again, it'll prompt you some states to say please smile. So to check whether you're alive or not. And we do that randomly. Not at the moment. Yes, I, yeah, so yeah, I think we just need to sort of work our way through the UN legal system slightly and slowly, but the intention is there that. Any help? Yeah. No. I'm afraid you have to buy it for us. Hopefully it should, so from an algorithm point of view, basically just using ML net and face net. And we've sort of trained the algorithms, et cetera. So yes, there is a degree of knowledge which I think will benefit the community. The thinking is there, you just have to bear with us as we just navigate our sort of little challenges through multiple UN agencies to make this open source. Please. Yeah. Yeah. Yes, indeed. So the solution we came, we have built, it is still not cross platform. What we do is we generate a long random passphrase in the app. We use that to encrypt the biometric file, the matrix that's been generated. We pull the biometric file on our servers, but that random passphrase that we have generated, we use your phone's native capability of Apple and Google to store it in the local key store. That still does not work for cross platform because if you move from Apple to Google, it will not work. And as I said, I like Apple in that sense because it's a bit easier to do in Apple. In Google, it will prompt you with the question. And if you click no, then if you want to revert your decision and that decision of not saving the passphrase in the Google key chain, it goes across devices. So if you switch to a new Google device and then you use the same password, your decision of no on the old device goes forward and the settings and what you have to guide the users through is like, you better go off and have root canal then you're trying to make those changes. But that's the solution we have come up with and we will, unless there's another better solution that comes out from the committee, we're thinking to continue with that when we push the wallet into the mobile app. Please, by all means. So totally of the record, well, I think it's recorded so I don't know if I can totally go off the record. The way I put it is our UN remunerations are quite comfortable. But we have seen roughly 52% split towards iPhone versus 48% towards Android. So it's these two platforms really. So we support Android and Apple. I think on Android we ran into trouble where there were some old devices but from a business process point of view, the fund is saying, look, this is one way, we are not stopping the old paper-based way. So if this doesn't work for you, that will continue. And there is a large population which is not comfortable with the devices because you're dealing with 80-year-old people. You're dealing with people who are incapacitated. You are dealing with people who are in some sort of vegetative state and they have legal holders who are signing on their behalf and they need to get pension. So we've got all those use cases. The main line process continues. What this does really for the pension fund is really, it gives them then time to focus on those because you've got a bunch of these already sorted out automatically. But when we extend this to UN Digital ID, I think we can still continue with phones. That's gonna be a big challenge. When we discuss in the substantive work of UN for refugee population, person of concerns population, then yes, we need to figure out how to do this offline. We've come up with some hacks but really any ideas on that side will be really welcome. Please, sorry, I'll just go ahead. So since we've launched, we've got roughly 15,000 people who, so that chart that I showed is the spread of 15,000 people. That's from the 82,000% population so it's not really huge. So we're making some process tweaks now. And different ways. So there are a lot of pension fund associations, so retired beneficiaries from the UN system, there's an Indian association of retired union beneficiaries, there's an American association, so sort of talking to them, trying up with them, making them people aware of it that this is available. Unfortunately, well, I wouldn't say unfortunately, wrong choice of words, but there is no incentive we can give to the user in this case. It's only when they run into trouble with the paper process and they come to know of the alternative, they jump onto this. So there is no incentive. We can't say what we'll get $2 more per month if you go for it. It's not allowed by the UN regulations. So that's the challenge. And yes, I think from on the usability also, we've made a lot of changes so hopefully that adoption slowly moves up. Nancy, and I just realized I was asked to repeat the questions which I have not been doing because people online and maybe it's a bit too late now for you to tell me that. So the question was, is there any regulation to undergo that certification? One, and two, what are the benefits if any of that? So no, there is no regulation. There is no mandate to undergo that certification. And it's a fairly common yardstick for a solution to be measured on security. So a lot of mostly commercial entities do it. We again, the pension fund and ourselves decided amongst ourselves that we will undergo this certification because we wanted to demonstrate the strength of the solution and sort of, it's like having any certification to say, hey, we are meeting certain bare minimum standards. So I know for example, I do believe a lot of national governments, state governments, they, when they build solutions they go for these certifications. It's just that because it is, there's an outward element to it obviously, which is you are basically saying we are ready to be audited by independent third parties and open up our systems to them and get their buy in and then their certification. Plus it does bring a degree of hygiene within the solution itself. So now the technical team hates it because it's not that they do a PR and you know, death psych ops and the build isn't, you know, goes out. There has to be breaks and checks and balances and whatnot you have it. So it's both there. Please. So in this case, it was a bit easy actually the first roll out. So because pension fund is, the board itself is a complex set of entities. It's on the website. I mean, with you and everything is public. So you will be able to see this on the website. So it was, I must admit, it was a bit easier to sort of go in and again, pandemic really helped because the pensioners were struggling, especially if they had to go to offices for some reason. So, and again, again, we're talking about a vulnerable population. So it's not really like you're talking about a 15 year old allowed to go out. You're talking about a 60, 70 year old, most vulnerable set of people who you cannot ask them to go out during the pandemic. So that was a big catalyst, but it was a single board. The advantage of that approval with the digital ID then was that when it went to this multi agency forum, it's called high level committee for management in UN. So when it went to this forum, it went with the idea or with the mark that this is already working. So this is not bringing in some new fancy tech. Some, I do recall, I think when the conversations were taking at that level, somebody started saying it must be one of the big fours or some high paid consultancy who's bringing in this ideas around over here. And the response was no, actually, this is working for UN and so it's proven. And that then helped quite a lot. So sorry, it doesn't give you the recipe if you're trying to sort of see how to push the program further in your own organization. But yeah, I mean, and kudos to pension fund honestly for them to bite into this early on. Any other questions? No, and if you do sort of think of anything, I think we'll all be parting later on the evening. So the beers, if anything comes out, feel free to just pass by and we can certainly have a chat. Thank you very much everyone for your time. Thank you.