 Good morning. My name is Hulman Haddad and I'm here to tell you how the United Nations World Food Program is using Ethereum to distribute aid to refugees. DWFB's mission is to end hunger and traditionally this has been simple where people are hungry, we bring food in kind like corn, rice and so on and so forth and we give it to them so they can eat. For a few years now we've also been doing something what's called cash based interventions which means that instead of giving people food in kind we give them the means to purchase their own food and this can take the form of bank cards, mobile money, e-routures and so on and so forth and the idea behind it is that the beneficiary is in the best position to make their own purchasing decisions so that essentially instead of saying okay here some food eat this they can make their own choice and there's an element of dignity in there and also that it can have a multiplier effect in the local economy. In order to enable this sort of modality we normally have to rely on financial intermediaries such as banks, mobile money companies and cash agents. So the services that these financial intermediaries provide falls under two broad categories. One is practically accounting which is creating accounts for beneficiaries, providing a distribution mechanism such as SIM cards or bank cards and so on and so forth, authorizing transactions when beneficiaries spend their entitlements and then eventually reporting to us so this let's call it accounting overall and separately it's actually paying the supermarkets or retailers where beneficiaries have made the expenditures. So a typical setup could be that at the beginning of the month the WFP gives a list of beneficiaries to the bank saying here's our beneficiaries, here's how much each one is getting and they will create the accounts and load them and we also have to advance money for usually a month to the bank to enable this essentially transfers to happen. The bank then contacts the beneficiaries saying here's you know you got some entitlements the beneficiary goes to a store initiates a transaction, the store pings the bank for authorization, the bank checks the account if they have enough entitlements they confirm back and then eventually post-factum on a periodic basis the bank will pay the store for the actual transactions that have taken place and let's say at the month end we get a report from the bank saying okay this is what actually happened. So a number of issues that exist with this model one is cost of course the bank will charge a fee for this and at our volumes that cost can be quite high there's the element of risk because we're advancing money to the bank and especially in the kind of places that we work that risk can be quite elevated if the bank goes bankrupt or comments fraud or whatever. We have to share beneficiary information with the bank they need to know essentially who they're opening account for and you know privacy could be compromised. There's the issue of reconciliation so when we get the report back from the bank we make all sorts of efforts to verify the report to make sure that the transactions that they say took place actually did take place and this is very laborious and time-consuming and we can never really be a hundred percent sure. There's element of control so normally for everything we want to do to add a new beneficiary increase entitlements whatever we have to formally advise the bank and they would do it on our behalf and in addition to the administrative burden there could be a time delay which could mean somebody goes without food for a while. We make a lot of effort to reduce our financial risk and also to protect beneficiary privacy but this also means the contracting can take quite a long time which again would delay operations and then there's the issue of flexibility so for example if we wanted to do e-vouchers and mobile money and ATM and these companies don't talk to each other and we don't know in advance where beneficiaries are going to redeem their entitlements we can't really enable this. I mean the best we can do is you know divided a third third third or half and half but it's not ideal so it's not quite flexible. So what we've done essentially is replace the financial intermediary with blockchain and so what we do is now we load the you know create blockchain accounts for beneficiaries load them with the entitlements we inform the beneficiary they have entitlements that they can use they go to a store they initiate a transaction the store now pings our blockchain instead of the bank for authorization we check the entitlements provide confirmation and then at the same frequency that the stores got paid before we know essentially which store to pay how much by when and we make the transfer to this door directly ourselves. Just a clarification here we have bank accounts and this transfer currently is happening through regular banking channels in the future it could be crypto for example or a digital national currency but the difference is that we have 350 global bank accounts that we use for you know making staff payments buying corn and so on and so forth and the cost of payments directly for us is really cheap the main expense to us with the previous model was this accounting issuing cards POS terminals that was essentially the main cost so yeah just to clarify that you know we have our corporate bank account where our bank accounts are and then usually we have to have a separate bank to provide those distributions and that's what we've replaced. So we did a quick proof of concept in the Omarkot village in the Sin province of Pakistan in January and based on that learnings essentially we've launched the pilot in the Azraq refugee camp in Jordan since May this year in the past five months we've been assisting 10,500 beneficiaries we've distributed $1.6 million $200,000 transactions without a glitch and we've reduced the fees essentially by 98% so that is what that accounting portion costs us and the 2% is what's costing us to make these direct payments in the future with crypto or something else that could even go maybe near zero we're not sharing any beneficial information with anyone so again in a refugee context that's very important we're not advancing any funds to anyone so our financial risk is greatly reduced we have full control so if we want to change entitlements at someone new we can practically do it in real time so the assistance is not delayed and very important for us we've authorized every single transaction so we have 100% certainty that you know what we think happened actually happened and that significantly improves our reconciliation process and the associated payments so it actually reduces the effort on our side then because we don't have to worry as much about financial risk and privacy protection then we can actually contract people faster which means we can operate faster and we're much more flexible now because all the accounts are on blockchain we can connect to as many outlets as we want and they're all coming out of the same wallet so we can provide that flexibility to beneficiaries which again it's great so you might ask why are you doing this on a blockchain you could do this essentially on a traditional system and that's completely true we could but one of the main reasons is that there's currently a lot of fragmentation and duplication in the system and this is only for illustration but essentially imagine if you have Bob and Bob is receiving assistance from four different agencies each agency would have their own identity system for Bob they would have the contract with a separate bank and they would channel assistance through each one of these and of course you know there's a cost and the effort required to maintain all the system the contracts with the bank and they don't talk to each other so that really the things are not optimized or harmonized in a way so what we're hoping to achieve and actually one of the reasons this exists is that the technology in the past has prevented collaboration among the agencies but it's various ownership issues so if we have a system and we have a contract with the bank and someone else has something else we don't want to use theirs because it's not cool and vice versa and what we hope is that blockchain can cut through that as a neutral mutually interoperable equally owned all of that you know open source system then it's not really owned by anyone but it's just a mutually operated system that everyone can use and this way we can have beneficiary let's say identities or wallets on the blockchain for Alice, Bob, Dave and Erin and various UN agencies can link to that and provide whatever service they want to provide and giving our various networks like for us as WFB with food retailers someone else might have health network like hospitals schools whatever else it might be but we can really dramatically increase the convenience to beneficiaries and if we're all using the same system we can really optimize and harmonize the global aid effort into an unprecedented degree with big data artificial intelligence whatever it might be and I think extremely importantly is the element of financial inclusion I mean even now without using crypto we were adding quite a bit of that to the people who may not have a bank account before but with the advent of that investments, loans, microloans insurance, international remittances would be available to many hopefully soon at low or no cost another reason why blockchain is that essentially so we have now let's say an identity for a person on a blockchain and we as WFB are adding financial transaction histories to it and just showing that you can spend money is not very meaningful but if you show that you can save a little bit of it that might actually count towards a credit history and this credit history now is not tied to a specific financial institution let's say in Jordan so if a Syrian refugee were to go back at some point they could transport this history with them and maybe get a small business loan and get back on their own feet if we manage to then bring other actors on board we can maybe add education data, health data, asset registries, work permits and make this profile increasingly more and more useful to the beneficiary wherever they are without the fear of persecution for example because of their origin or whatever and I don't think it's unfeasible to think that one day maybe even government IDs could end up on the blockchain and we've heard here a lot of things about privacy protection, zero knowledge proofs and I know it's at an early stage but essentially this is the path we've just started and we hope as technology advances connectivity becomes widely available the price of smartphone goes down in a few years time this could be a full-fledged essentially tool for beneficiaries to have more convenience and flexibility so our invitation is to other humanitarian actors who don't have a profit model built in is to not to reinvent the wheel so we have note set up we have essentially smart contracts for cash transfers and we're asking if they have resources they join us open source free and dedicate their resources to making the system better, more secure add additional modules like identity, supply chain, health, whatever and also make it available for free so that it can essentially grow to be a system everyone can use. Currently we're using a private chain and mainly that's because the transaction throughput that we require can't be handled on the public chain but one day I mean philosophically we're leaned if we can ensure security and privacy and all of that to move there so the system is self-sustaining and doesn't even require the backing of WFB or the UN or whoever else for it to function. So yeah please join us and we'll see how the world is vulnerable. We really look forward to collaboration. This is a photo from our proof of concept that we did January of this year in Pakistan essentially and for this we used the test net to distribute food and cash to these beneficiaries essentially to illustrate the point that it can be done and using that learning then we've done our Jordan pilot which is now going well it's five months in production and we're soon going to scale up to 100,000 beneficiaries and a few months after that to 500,000 which is all of Jordan and from there other countries could follow that have you know that could potentially benefit from this solution. I have many many people to thank of course WFB itself for providing this safe space for this idea to take root. The Ethereum community at large who has been very supportive for allowing me to be here today and get this message out and also Rich Boto, Rich put your hand up who we met at the Singularity University around this time last year where this was just an idea and we had very little resources and support and we told him about it. He helped us think through it and he pretty much hacked together what we used in this proof of concept in the photo that you see and he did it for free. No matter how much we insisted he didn't accept any funds so in a way he gave us that initial push that we really needed to get this pilot flying essentially. My email address is there if anyone is interested in more information and since I have a few minutes left I would actually like to show you a video of how the system is currently working. So in Jordan they happen to use IRIS verification even before we went so we wanted to seamlessly integrate into that so we don't change the beneficiary experience especially if we're going to fail and the system would have to be rolled back essentially. So what you would see here on one side is the cash register at the supermarket in one of the refugee camps that they ring up the normal bill, the amount comes up the amount is put into the IRIS scanning terminal, the beneficiary scans their IRIS, that IRIS then goes to be checked against an identity the identity comes back and at that point the terminal has both the amount and the identity of person trying to do it, they ping our blockchain can this person do this transaction, we reply and that's essentially the lot of oranges. On the bottom right you'll briefly see the amount, 1280, beneficiary scans their IRIS, transaction goes, the IRIS imprint goes to the database, comes back with the identity now it's come into our blockchain with the amount and identity and that's it. I have four and a half minutes so maybe one or two questions I think there was one already there. In this case you have this IRIS scanning infrastructure but in general in order to have established the blockchain identity especially a long term one you would need a private public key pair and do you have any solutions in place providing people with public private key pairs in such a way that they can control them and they have adequate security with them? Yes so good question one of the things we cannot assume right now is hardware and connectivity for everyone especially where we are right now in fact the internet is cut off again we're building towards the future so as you might know Google, SpaceX one web a whole bunch of people are working to bring affordable satellite internet everywhere in the world and the price of smartphones is just dramatically dropping so I think by the time we're ready the standards for identity have been sorted out zero knowledge proofs whatever we can then transfer beneficiary private keys to them potentially under devices and the device could even have a biometric authentication to trigger the private key but for it to be the data to be stored with each beneficiary and not centrally not sure if I answered the question. I'm working with a group in Bangladesh handling the refugee crisis with Burma or Myanmar it's about 1.5 million people first question how many beneficiaries can this scale to sorry how many beneficiaries? Transactions can it scale to can it do a billion a day because we're looking at these kind of numbers with 1.5 million refugees the other question is can it also not just do payments but learning outcomes and then settle with some incentivized collection of meals or some kind of incentive and what do you think can we connect later because it's a long question yeah definitely and how do we become partners with you? Sure get in touch my email is there and so on the first question as I mentioned we did our proof of concept on the test nets we're doing a private implementation right now and we can handle quite a bit of transactions per second so a million a day won't be a problem in fact we're shooting more for more because WFP itself has 80 million beneficiaries that we serve in 80 countries and we've heard over the past few days the efforts are in place to essentially speed up everything potentially even the main net and so we don't foresee the throughput to be a limitation either now on what we're doing privately or even in the future and then yeah I mean definitely that's I think the beauty of smart contracts that you can program whatever logic you want into it so for example we have school meals which means that generally speaking let's say if a child attends school then the child in the family gets food so it's in a way to ensure children attend school and also that some nutritional outcomes are met and with smart devices and everything yeah we can essentially sort of automate that so that you know the entitlements go into a smart contract in escrow and if certain conditions are met then they're triggered to the end recipient or recipients 56 seconds for a really quick one in case there's one cool thank you