Baseline Protocol The Baseline Show - Office Hours, 17 May 2023

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: The Baseline Show: w/ zkPass part 2

Description

The Baseline Show will host the team from zkPass for part 2, to learn more about the use cases and work they're doing!

View part 1 @ https://youtube.com/live/fVhRldvGVIk?feature=share

Learn more about zkPass @ https://zkpass.org/home

Baseline Protocol is an emerging standard for state synchronization across different systems of record. This standard can be applied to traditional corporate systems of record and utilize any kind of database or state machine, including different blockchains or distributed ledger technologies (DLTs).

Join us every Wednesday at noon EST on our live show, 'The Baseline Show'. Each week, we host industry leaders who share their insights and expert opinions on data governance, security, digital supply chain, multi-party workflows and collaboration.

Tune in via YouTube or LinkedIn Live.

#BaselineProtocol #DataManagement #Blockchain #DataSharing #MultiPartyWorkflows #DataGovernance #InformationSecurity #BusinessAutomation #BusinessSynchronization #DataIntegrity #IndustryTrends #LiveShow #LiveStream #LiveDiscussion #ExpertOpinion #YouTube #LinkedInLive #B2BBlockchain #DigitalSupplyChain #Innovation #SupplyChain

A

Hey Sona, could you please let our co-founder uh bing? Yes,.

B

We're live. Everybody welcome, hello, everybody um and Josh I, don't see anyone in the waiting room, but once he's in, we will accept him.

C

B

Awesome all right welcome to the Baseline show this week. We are streaming a little earlier than usual, um so we'll see how this time works out for some of our viewers, as well as our live studio. We are here with the ZK pass today for part two of a live stream.

B

We had them on a couple weeks ago feel free to check out that stream afterwards, but we're here for part two to learn more about what they're doing the use cases the actual projects, and we will find synergies with what we're doing in the Baseline world, so Annie I think I can start by handing off to you yeah.

D

Sure, thanks Sona and thanks for inviting us to the second part of the Baseline show um you know. I was very happy to introduce you guys to what briefly you know like CK password trying to build here and I might just quickly cast a slide on our use cases and elaborate a bit more from there.

D

Okay, let me know if you guys can see it.

B

D

Still, loading.

E

It's a bit slower.

D

No still not yet um you know what I'll, probably just just talking it's it's not that complicated. It's basically a slide that shows. Obviously we have pretty broad um kind of like scenarios where we have. You know many use cases starting from the Pretty uh straightforward, zkkyc compliance onboarding.

D

You know, for example, um CK parts we hit here can enable users to prove that you know people need um some sort of like regulatory related requirement, such as your age being over at 18 or 21 years old, or your nationality, whatever it is without disclosing much of the exact um you know detailed information such as your date of birth, your your you know whether, if I want to prove that I'm, not American without letting them know exactly what nationality I am I will be able to do that so use using ZK pass.

D

Users can essentially generate zero knowledge proof that can confirm that they meet the certain, whatever age criteria, providing a more privacy, preserving solution.

D

um For you know some of the age restrictive services or products such as Gamepad project, with a bit of violent content in it um and also zika pass cam, also being integrated into websites and online services that require age verification to ensure that users do actually meet the required um thresholds, and um you know there's other things like online gaming, identity, verification, social media platform authentication.

D

uh We essentially can be integrated into social media platform to verify user identities, reducing the risk of fake accounts and improving platform security.

D

um A lot of other face uh use cases uh include in the um D5 kind of scenario we can achieve uh approval fonts, you know, ZK pass can be also uh integrated into D5 platforms, to provide users with secure privacy, preserving identity, verification as well as you know, enabling seamless onboarding reduce the risk of Fraud and for landing platforms. We can also be deployed by money, market or Landing platform to verify borrowers, not just identities but also credit worthiness and also even if they want to verify their real world asset own chain asset.

D

Whatever it is right, we can provide a pretty secure, zero knowledge way of doing that. If you like, I, mentioned last time, I believe in my slides, if I want to prove, for example, I have over um 300 000 or whatever it is in my package account to be able to achieve like a good Landing rate. I can be able to do that without disclosing, exactly how much I have in my bank account uh and when it comes to governance and voting. You know we can do Anonymous voting uh participation qualification.

D

Essentially, that can be achieved through uh kind of zero knowledge uh prove technology as well right. So we can allow users to prove their eligibility without revealing their identity and using ZK pass. Folders can generate zero knowledge proofs that can confirm that they are registered and eligible to vote while maintaining an anonymous- um and you know many other things for a more secure privacy.

D

um Sensitive information such as even you know, one's health records. You know whether we we can also be adopted to enable secure and privacy preserving access to patients, possibly electronic health records, ensuring data integrity and patients, privacy uh and stuff. Like clinical trial. We can also possibly be deployed to do like screening and applications and and registry and Etc um educational background research. You know like if you want to verify whether someone indeed graduated from a certain faculty University.

D

You will be able to do that, instead of them just putting it on LinkedIn, because they there could be fake profiles and you know just just paid claims, so all of that can be verified because essentially you would have a login to your to your to a graduated. You know faculty or your alumni, so most likely you will have that you will have a transcript. You will have a graduation certificate um and you know an extension to that is kind of your experience, checks which you will be able to do, that.

D

Your employment history check um and Etc and I think I'm going on a lot and that's you know just a few examples of uh the use cases that we can be applied to, and uh you know let me know if you guys have any questions on top of that.

B

Mark Hadley, you looked like you're ready to ask a question: do you.

F

Have one you always say that do I always really I mean I'd, be a lousy poker uh player. I mean having too many tells and stuff like that. No I think this is really interesting under the uh ZK knowledge um type of applications. Is um it really when you start to where you have a need to where you need to gather data so you're in a real University setting, and you need to query uh I, don't know financial data on a certain.

F

uh You know: population, demographic within a certain region, and you want to find out. Okay I need to find out of these people. How many are there? uh You know, of course, how many are male and female and um so forth, but also how many are homeowners. How many have uh no credit card debt? How many have higher credit card debt I mean?

F

All of these are really sensitive things that you could not ask directly, but you could, in a Range proof type setting come in and say, I need to at least Identify, say I'm doing a study. I need to identify people for the population that I'm going to be drawing my data from and how do I find these people, and so you could do a query like this. uh You know with the data saying: hey, I have a permissive purpose, but I am not violating any privacy.

F

I am not getting any identification of any individual on there, but I am getting actionable data that these people meet the criteria of what I need to conduct my research on, and so have you ever considered something similar like that to where that.

D

That is definitely a very true and valid. um That's what we you know hopefully can achieve one day which is very, very high level of use of profiling um without obviously revealing their identity, and but this does require Leslie a big accumulation of use uh user base, because you know for any research like that. It's you know it's minimum thousands right, if not tens of thousands kind of like populations that so yeah, hopefully in the future.

D

You know if everyone's kind of used to being able to prove you know like attestations, um I'm using the zero knowledge kind of technology and be comfortable with it, and you know, essentially just everyone have a bunch of svts and attestations to to to state that they they are this. They graduated from here. They they have this Etc so yeah, hopefully that that that that will be very valuable for sure.

F

E

F

Talk to a lot of data scientists, because that would be a really strong tool for their toolbox is to be able to say, I think the theory. These large data sets- and you know a lot of that's going to be population data, demographic data- that is public domain and that is in you know the the government space um yeah and uh but you need the raw data and they're not going to give you the raw data, because again, privacy right saying, hey I, just need to know these people.

F

uh You know it's almost like you know: I'm a fisherman tell me which side of the lake that I need to go fish yeah.

D

Well, I I can't even install it. You know like how valuable that will be, but yeah like um it. Definitely it will be a long way to go because that that that sounds like you know, like everyone, a lot of people needs to get on it and for us to have like that kind of you know, base population pool first yeah.

F

You know I'm thinking, if you're doing you know like uh trying to match user data and stuff, um you know generally users you're going to already have them vetted and onboarded onto your platform. So as users, these would be people that you don't know that you want to try to entice to maybe become users, so marketing would absolutely yeah no.

D

What we're thinking in terms of uh go to macular Marketing in general, is obviously you would have you know like the projects or business side, or have you call them? You know, or we can refer them to business user? For now who have verification needs? Who will come to us? They can be um I, ideal, IO, Pro platforms, or they can be a game fire project we actually do realize.

D

There are a lot of um projects who have verification needs to prove that their users, their sign ups, are not from a certain jurisdiction which is kind of like the most common need, and obviously see with that. You can. Obviously um you know like you- can accumulate a lot of initial user bases um through you know their sign ups, their their users, their potential users and, on the other hand, we will encourage. You know individual users who would like to essentially just mean to like sbts about their qualifications about their.

D

You know their experience about their Dow membership about their uh if I own, a Porsche I want to tell people I want to flash it. I want to tell people I have a Ferrari just like how I want to flash my twenty thousand dollar nft profile picture, similar thing right, so kind of that's what we're thinking uh like both sides. We can, um you know, start facing and try to entice. You know get people to sign up and you know try it out.

F

Yeah I mean it gives you the capability to do a lot of things that you cannot do now, just because of governance and privacy.

D

Yeah not definitely- um and you know it's essentially, you know in a more broader sense anything web page based as long as you have a log into it. We can essentially turn that into a online test station, and you know it's it's it's very valid. It's it's! It's not tempered! It's it's! It's true! Yeah.

G

F

G

I think just really with.

F

The advancement proliferation of a lot of the chat GPT and the generative AI type use cases because now they're starting to say hey, this is really cool, but wait a minute. The uh the the fuel for ingest in this do we have it properly sourced and was it legally obtained? And, yes, we're scrubbing web data, but you know what is it and so you're starting to get a lot of regulators looking at this and a lot of people are saying: well, I, don't want Regulators with their flashlights.

F

You know badges walking, you know poking around, so this is the way that you can at least have that arms length thing to say: hey I'm I'm within the guardrails here.

D

Yeah yeah, not definitely I, think uh you know, even when it comes to like web 2 Worlds. You know there's like a lot of potential application there I like because it remain you know like remain anonymous or like that layer of like privacy shielding is quite important to to a lot of service providers. So you know we'll we'll see cool.

D

Thank you. Thank you.

E

So I wanted to ask I'm not sure if you showed this last time, but did you guys show what using ZK pass actually looks like for a user, whether it's like a business who's trying to provide kyc for customers or a customer who's like engaged like getting a bank account and they need to provide kyc? Did you show what that looks like, or could you.

D

um I think we could, we could show a demo of um you know when you are trying to link a SVT to prove your let's say account. uh Internet bank account balance, um which is you know. We we have a few scenarios. One of them is like proving your legal Identity or your Australian, which um you know um you can do that by logging into your mygov Australia, which is like the government official website to approve your everything about yourself. You know your essentially Legal Information or your digital IDs.

D

There um and I think it's worth showing that demo, uh which uh Ryan, if you can pull it up, I think my casting is a bit off um and uh that one is essentially proving um that you know it can be in our case is only proving you have over 30, but you get the idea. It can be three three hundred thousand, it's only thirty dollars, because that was uh almost absolutely an internet bank account, but uh um yeah we'll play that, for you.

B

And on that note, can anyone plug ZK pass into their own applications and use cases, or is it more so on the user side.

A

Okay, you guys see my screen here. Yes,.

B

A

Okay, let's get started.

A

Okay, uh once you log into the homepage of the secret pass, uh there will be a gksb team, you click mint ckspt and we have already classified as the shower type into seven seven types of Chen legal identity, a Financial social education experience and skills, uh because right now, I want to approve that I am an Australian. So right now we have to choose the verified data data source. Is the official government website? It's called mygov.

A

My God is a digital identity of the old Australians, so I want to verify. I am I, am coming from Australia and I'm an Australian. You click meet now.

A

And now it is, it is under processing the the Maori party computation and TRS sections. So probably it takes uh 10 to 15 seconds.

A

Once it's done it's now the gkp process.

A

So the gkp will be uh will be generated on I mean locally and you have to submit it on chain and you can verify very function.

A

Once it's done, you can see all the information about GK spt on your profile. It shows the data source you are generated from and the website your verified fields, all of the information, and you can find your gksbt on chain in the Explorer. Just click the hash here and you can find on Channel.

A

D

B

Thanks Ryan here to.

D

Play one: that's proving your nationality, but you get the idea. It's it's um similar concept when you're trying to improve your internet bank balance is over a certain limit or uh between us or limit or whatever it is. You can. You know pretty much Define what you want to verify um like in terms of the criteria and essentially yeah like we, because of the fact that we know these credentials were generated from a trustworthy data source, for example. In this case, it's my gov dot, dot, dot, a.gov dot, a? U which we know.

D

Obviously that's the um Government website, which holds every Australian's digital ID, so yeah through that you know, essentially you can prove I am Australian or I am not American uh with this way without disclosing way too. Much of you know, instead of me sending you a copy of my passport bio page.

G

I have a question: if you don't mind.

D

Yeah, of course, Keith yeah.

G

So um and a lot of the you know the research I've done and the kind of the the um ZK presentations I've been to one of the the problems of the underlying problems is the you know: what's what's being used to produce the ZK circuit. All these domain, specific languages and the possibility of um you know introducing errors within that circuit is generally much higher than um you know. Logic kind of this pragmatic logic that we've been using in other languages for a long time, so someone who's asking like.

G

What's the guarantee that, while you're constructing that zksbt that the the circuit that you're using under the hood, whether it be for um you know, nationality or or um you know, private medical data or or Bank data? What's the guarantee that there is no law in that circuit?

G

And then, in fact, you are proving something slightly different than what you're say: you're proving right because in the end, I just received kind of like a a a zero knowledge proof that anybody can verify and but they're trusting that the that the logic inside that circuit is true, because when I run the verification, it's just kind of verify and say: oh yeah. This this proof when run through the circuit is good right.

G

So what kind of guarantees do I have there? And you know what would you say to somebody kind of asking that that type of question or that that's especially like non-technical right, and they want to use your system like what guarantees do we have that you are very buying exactly where you say, you're, verifying.

D

Right are you saying like, for example, the one that we showed you? How do you believe that it was actually credentials generated from that data source? Essentially.

G

um It could be that or it could also be. Let's say what you're doing the the the money thing with the bank accounts that um let's say, I I missed a portion of the last presentation of under the hood, the debate with Andreas under the hood. uh But let's say just to simplify the argument of like. If we're doing, we.

E

G

You have more than three thousand dollars or thirty dollars, and the circuit that's being used, though, is written incorrectly and has a flaw in it right and there's like a greater than rather than a less than just for, like simplicity's sake of argument like but I can't see inside that circuit right, that's a black box I'm, just verifying against the circuit right. So what's what guarantee do I have uh you know that that is correct right.

D

I want to last time, uh I think our our you know, Tech, Team, being and and then and Josh. We we kind of um it needs it, probably better. If we can illustrate with that um technical process flowchart, um but you know maybe quickly if either Josh and Bing can bring that up again um and essentially, you know just just to obviously one all of them. We will be open source so that will be audited so which you know for for those of you who are more technical background for to have a peace of mind.

D

It's you know more than welcome to and uh yeah and um I think Josh and.

H

D

H

Okay, so I can walk you guys us through on the tax side hold on. Let me let me open my slides Yeah. So, basically, uh what we are showing here is: okay, how user is able to kind of get the data from whatever they want to over three data like a data source? Let's say that you know the um the legal side you go for those. You know a d Bank. um Okay, my cover ID and the bank account those credits you're going for those Bank website. That's basically what we do so earning data.

H

You are ready to related to Identity from the web tool or through oversight. Okay, we are able to use the ticket as a protocol convert that one into a 3K SVT, and this is conditionally.

G

You were going to share the slide right. I, don't really see anything right, yeah hold on okay, so I'm still looking for it. Sorry.

G

um Go ahead: yeah, oh okay! There.

H

We go okay, so okay.

G

uh Yeah I can see it.

H

H

Okay, so before before the slides, I'd like to you know, uh work you guys through this one. This one is kind of a more concise um like a procedures to like how, basically it works. So, let's just assume okay in this user for now. Okay, maybe let's say compound- needs an anti-activity right. Okay, so they want to over this activity and they want to verify okay number one, the user's real name and number two they're kind of like a world two bank account or something deposits they may just want to verify.

H

This is a real user and the data source they want to use is Bank of America boa okay, so the first step would appear so the compounder put the you know the verified field and, let's say the name and the bank balance there in the template and the pointer data source and then user site Alice.

H

She wants to join this event. So, okay, she going to pick this pick this template from the compound site and then the protocol is able to read the yoga Alice log into Bank of America website, so use her password and the username that that'd be the first step. The Second Step would be you know, based on the TRS level. Okay, uh the Bank of America has do, has a responsive back to you know the protocol up on the local Alice site, okay, with the name that kind of balance.

H

Right, that's going to be the second step and upon this point, okay Alice would have two choices. Number one Alice has a potential to Fortune the data I said: okay, my backer bank balance will be only ten dollars, but I want to change that one to a hundred or 10K, or something like that. So she has the potential to do that, so, which means we do need a magnet or a system to prevent Alice from forging the data. So that's how we designed this.

H

You know the next step, which is the MPC multi-party computation, and then we engage multi-party, computation node, which is the question over here. We okay the function of this node, it's a two number one permanent Alice from from 14. data. So how does it work is okay, we have engaged the two keys in this process.

H

Number one would be the you know the increa encrypted key, so this key is two okay to encrypt whatever you know the responses from the Bank of America, which is you know the name and that kind of balance just to prevent the notes from you know accessing you know Alice to private information, so Alice only Alice in this case has the incredible key. The notes do not have it. That's the first key. Second key: is the market key Market key, so this key is the data Integrity key?

H

Okay, so the arrow side and the node side they have a ship also has a share of the keys So. Eventually, okay, once we finish the MPC, we need to check this case if the key is verified, which means Alice do not afford your data. If we cannot, you know, you know, verify the key, which means LS already forged data. So that's that's the case. So that's the function of the MPC okay, so we run the MPC once the MPC is finished. You see the thermal in the demo.

H

Okay, we do have this NPC right and after MP say: okay, so which means okay. We already get responsible for the Bank of America and we already prove Alice do not afford your data, then okay, we base of these responses. We generate this zkpd analogy proof, and now we transfer this. We generate this DK SBT on chain and the Passover this sticker speaking to compound so campaign, is able to verify okay Alice as a user. We approved username and that kind of balance and the camera do not.

H

You know, have the exact information of the ls due to the energy proof, so that's basically the whole flow. If we go one step further in terms of like the uh the detail, more detailed technical side would be like this, so the first one would be okay, um the first one would be okay, so Alice as a user. P right is the user side also, so they just do the you know, client, hello, so just use the traditional uh TRS level.

H

So just um simple accessor token have the you know: client, Hollow, handshake or two that water sites and then so, once I do have like a um certificate certificate. You know pass it over back to the the Alice side. So based off this. You know: certificate, okay, the user side and also the node is able to generate this there kind of like a um so. So this is the public key and then we come by because we have this two part details into three parties.

H

So we combine this node and the users um the key into a combined one, which is here you see and then so, based on based off this combine the keys. Okay, we generate discipline secret and we have this. So that's that's the pretty much the statement we have um based of this. You know the keys, the public keys from um MPC nodes and the user side.

H

So what I'm describing here is just you know the typical, like a TRS level, how to how the protocol get access from the Bank of America and also how we are able to you know the protocol helps you know the the data from you know, for the user and without user to budget data. So and then, once you get this secret and then so, each ship party generate their Master secret and then eventually over this MPC, we checked this. You know you see the step. Seven is that we check mark So in this market.

H

We're checking the market key and, and also that did a quarry from from the user site and the step eight would be the request you see. So this is the Q that would be the encrypt um so encrypt. You know the function besides the encrypted key and also query and then so step nine would be the responses, and then we also do the encrypt encrypted side and also we check this Market key and so with the with the those combined, the pretty massive Secret.

H

So that's the step, nine and then up so the step. 10 11 and 12. That would be once we kind of finish that one okay, we need to check the you know the market market key and encrypt key together. That's that's the whole NPC, so this once the NPC is finished, and then we move we jump over to you know to to the zenology proof. So for this zero knowledge, proof, okay, we we need to check the few constraints, so the number one the first one would be.

H

You know this is the first line is encrypt K and the responses. So this is the the public input. This is kind of more like a public input, so to verify that one and then we verify there's a market key and to check the market. So so that's it. Then we need to you know, check this.

H

This kind of the asset, which is the query which is the kind of like the responses um or the queries we get from from you, know the user side and also the query of the token, which is the questions and the asset is more like responses from the you know, the server side and the base of this orders verifications, and we so we generate the proof based off.

H

You know the above um for you about four state and then that's basically how we generated the KP and you see for each, um but for each kind of like a template or for easy.

H

Like jkp, we have a different like constraints, you don't need to change um the constraints here um so that just be changing like a gamma, take the circuits and then so that's how you generate proof and then just you know um then transfer this proof on on my on chain and then you're able to verify, use the you know say another contract to verify it and and check so basically, that's the whole um process.

H

um Let me know if you have any questions on this.

D

Yeah um I think uh Josh uh went on pretty detailed in terms of the process, but just to add on a bit on the Black Box uh front, uh so it won't be a black box because uh you know right now we're just uh going through like testing phase, but eventually, obviously we will become open source. So it's you know open to anyone's literally going through the codes and, for example, if it's a shot 256 circuit, you can literally see pretty much every step of it.

D

So you know like you, you will be able to Auto go review it yourself and essentially, for example, if let's say um today, we have a community um programmer uh who's from I, don't know, let's say Brazil and be like hey. Is it okay past weekend actually help? You add this new um Brazilian bank to your data source, which you don't currently have so we we will be open to kind of like delegate that to community which they can.

D

Obviously, you know help us adopt new data source, and you know it's it's literally a matter of like changing the template and which then will be uh audited um by the community by our internal Community uh committee and and the community and external audit, and which then will be combined into the code and obviously everything is open source and it's subject to pretty much everyone's audit. So if that makes sense.

G

Yeah, that makes sense- and that's awesome, to hear I- do have some questions related to this flow. But, since you touched on, uh you know open sourcing um Annie is there any idea you can give us a road map of kind of uh you know when that would happen when people can start getting involved. Is that already happening? You know when can I when can I use the the ZK pass kit? Anything like that.

D

Yeah, so in terms of roadmap, we are right now um kind of like doing our pre-alpha test match, so we're a bit far from magnet yet, but won't be that far um so um uh likely gonna be uh Q3 for mingnet. So um you know, in terms of like open source will be uh after Tesla, We Believe yeah.

H

G

H

We are going to kind of like do the auditing and you know be open source. You know uh when the mainland is is out, so that's basically yeah.

G

Yeah: okay, and are you looking for um individuals or entities to eventually be running a ZK past nodes, as well as part of the net or.

H

Yeah, that's that's true. um We do. As you see, we do. Have this MPC decentralized network, uh helping. You know the whole system to uh do the modified computation um yeah. We we are going to have a community of like um the Runnings.

H

um Could it be the individuals also, you know a lot of every season. They are kind of like very interesting. You like internals as well, yeah. You are chill so that's that's kind of like the the reaction we are going. Yeah.

G

Okay, so for people that may potentially uh be interested in in running notes, you know what what is their? Is it just if they uh like the idea and want to participate in this type of system? You know what can they, what can they gain right.

H

Yeah, so um in terms of the the benefit of running the notes, I know, but there are two type of benefits number one would be okay, um so running notes is a belongs to the community like a rewards.

H

um In the token timings we do have a larger portion of tokens allocated to those you know on the network. um I think we have.

H

um We have like a Five-Year Plan of those token allocation to those decent products that our company needs. That's pretty good number one uh number two would be okay for um for each verification, we are going to charge your second fee for um just kind of like this would be the income or income source for the protocol and for each verification fee. uh We are going to allocate 50 to those notes and we, and so that would be the benefit and the the the investment that the costs may need to do.

H

Okay, they need to stake certain amount for tokens at the KPS. In order to be to run the load. Stuff will be the number one I think the initially design we have like a 10k of gdps number, two okay. They need to have to come and get a AWS, AWS or Google Cloud type of for um like a computation power in order to run in the ecosystem yeah. So that's kind of from like the how we designed the whole. um They took anonymous.

G

Okay, all right well since you've brought up tokens when, when will this token be available, and how can I get some.

H

Okay and we um I think we are launching the tokens um pretty much very close to the mainland ledge um near the end of the code of three or maybe a quarter. Four I think it should be a random code of four or something like that. Yes,.

G

Okay, all right thanks, yeah sorry I had to ask once you mentioned tokens. I think you need to ask you know when, when they're going to be out right because uh yeah.

H

E

Like what I.

H

G

With ZK pass, they could be hot right.

H

Yes um right so basically in tokens is very important, um like a part of the ecosystem, um because we need those tokens. So, in order to run in the notes, we need those tokens um the each Innovation they need to stake certain amount of tokens. That's you know we have to have those tokens um pretty much before the mainland yeah.

G

Okay, cool thanks um so now to get back to my question to this chart. I know um last time: Andreas uh put you guys through the the fire um I'm I'm gonna, just assume. For that part, you know I trust. uh You know the the process of um um you know the certifications and the keys. What I'm interested in, though, is kind of uh what portion?

G

Probably the next slide, where uh the ZK circuit, uh the actual ZK proof is being created and what portion and the flow or the user flow am I able to kind of create a flavor. Let's say if it is bank account and everything's done correctly, with certs and keys, but where can I as the user set well, I want to prove greater than three thousand right. Is this going to be like in the interface where I can alter kind of like uh from a set of pre-made circuits, or you know.

D

uh You're absolutely correct, so essentially you can um whatever it is that you want to verify, let it be over three hundred thousand thirty thousand whatever fifty thousand you essentially key in that which you know they are like the um you know like that. Various fields that you know can be then obviously decoded into the circuit for the verification so yeah, let it be greater than smaller than in between uh or whatever right. Essentially there are a few scenarios for for stuff. Like you know, numbers.

G

Okay, um can anybody speak to kind of how you're accomplishing that under the hood, you know what um ZK language or DSL you're using is that being like generated on the Fly? Can anybody give some insight as as to that or the route that you plan to take? If it's not fully there.

H

We use and how how it's been implemented.

C

Okay, so uh for the first version we will use the clunky GK algorithm, and uh so so so. The most uh work here is to write the customer gate here and give a few uh optimize for for the circuit, and so actually there are about uh three main parts. So the first one is the authentication of the response. So here there is actually just a chart 256 circuit, which is a common uh in in the blank, and we don't do some.

C

We don't do much optimize for this part, so the second one is to pass the response yeah to a specific grammar for the response, because actually all the pattern of the response is fixed when, when uh when we do and all we Define find the template, so so we should pass if, if the response is actually what we have defined in the template, so this is the part and uh for for this part, actually we don't need to do it in in the circuit, because there's a lot of, if else case in the circuit.

C

So if we put all these together, the circuit will be very, very large so, but we we can just pass it and uh out of the circuit because actually each input, we are output. If we open the pattern or we open the grammar, the response.

C

It's a response format yeah to to the verifier, so the web verifier can can can just uh no no which, where which we are in the circuit, is actually the uh data or which wire is actually the pattern or the P. So so, if you could directly January circuit or check, if, if the.

C

Is right or not so so we could do this out of the circuit, so the second part is the assert. We need to make sure that the third response, uh some some value in the response, for example, the age, is actually larger than 18 or something else. So this part we have already uh release a few I mean like uh hold.

C

For example, we will release some great sand or equal or Yin, or some some of or something like this logic, so uh any I think I.

C

Almost uh the developer can just check our code and we can just rewrite a few parts. I mean change, some part that we will can generate their own circuit. So this is how how it will do for the proof.

G

Okay, all right yeah, very cool. um Thank you guys. That's that's. Most of my questions.

D

Thanks Keith, um you know just um but obviously the technical part that they they they're they're the expert of it and it's you know if it's just worth, highlighting that our speed of uh CK proof generation is currently set at around 16 seconds, which I believe is.

D

You know pretty much way ahead of um other competitors, um and uh you know I think our team is still working on optimization to hopefully bring that further back down to around 10 seconds, which obviously will make us more commercially viable than a lot of the other competitors who might be taking over one minute. So yeah just like to highlight that and yeah.

B

Awesome any final questions from the group.

F

H

Thank you. Thank you. Yeah. We have been working on this for like a very long week, we're starting to get past like from uh last year, quarter one. So you think, even before that a week kind of like yeah like um pretty much like a half a year of like a CK related like a deeper research, that's yeah, so.

B

Actually, it would be cool to hear the story of how ZK pass started and how the team came together. If you could end with sharing that with us,.

H

A

H

So yeah yeah, you can see something go.

D

Ahead, you go ahead. Josh.

H

um So pretty much you know DK um so overall been it's the co-founder that I can eat, and so he he's a background is all he has been working uh in kind of cryptographic, space for like a 15 plus years and he was like attack need in entity.com, it's more like a T-Mobile type of a Japanese company right or ATT type of uh T-mobile I will see yeah.

H

So so he has been in that space for very long and he has a you know um the first project he didn't related to like the case, but somehow he found some there's some problems on that. You know um polycastle.

H

um Basically, he had some ideas um about. You know how how we should do something and then so I I my background is kind of I'm a co-founder to more like a strategy and a product site on I graduated from University from Missouri, with a PhD in structure, engineering and I I worked in New York City at GMS as a senior engineer from 2015 and 218.

H

um after the interview like uh I, think I quit my you know, I think it was a 2018 January I quit my job, so I started. My first kind of like a blockchain projects and then based on how it didn't work out, so that was kind of like a token as a real estate. It's more like a real world. Assets Now easy is like a hard topic for now, but it's kind of a little early I think so you know we. We cannot be on the market together.

H

At the you know, one of our colleagues in a girl party, so we kind of want to hit it off. Okay, so we uh he had some ideas about how the world is pushing forward in terms of electricity key and also there's some something we need to do on the you know: identity, side and then I heard okay. This is great project.

H

The idea is so maybe we should do something, so we kind of like a talk about this 3p I mean obviously treat the key kind of like a period of fall about a half a year or something and we both- and we have a few other like we have a Dean- has one classmate classmate and we kind of talking about GP and read papers. So we didn't.

H

We decided to we okay, we are able to do something in this space, so we officially want you to give us I think it's called a one um 2000 last year and then so we have okay, we we know anywhere well and so she's she's in Australia and she's from like a traditional Bank banking site. Okay, see it's very easy, she's, very good, like Outreach, into like a university and also the business development side. So we have so so any and also Ryan. They are more on. You know the business development side.

H

So that's how basically the initial record the core team of secret pass, and then we gather into violence, incubation I, think last year and the up season five. So out of season, so the fourth season, five they list they picked like I, think the 12 teams, out of maybe a thousand applicants, so the capacity is okay, one out for 12 and we are the one of two teams uh they categorize as infrastructure. So that's how you know we are not at the idea project.

H

We are not kind of like you know, then it's very simple application. um um Those are projects, so we we kind of like the infrastructures. So that's that's kind of how balance you know put that label to the project and then, after the three months of like incubation, okay, we had a demo all day, so okay 12 teams competing with each other. Okay, so we take it passes.

H

We're number one I, think a balanced, Labs kind of like are broadcasting this um this this kind of like a demo day, because that was the first one they ever did in the matters. um You probably see I think they already have like episode. One coming out right, right, I think it's it's episode, one is already out so and then so we did it very well. So we are number one in this demo day and then so that's how we um in terms of tracking tracks inside and also the city investment.

H

We I think we have a very good jacket, and so we have a bunch of vision, Manchester. You know after testing that, but we already have like 100k of like, like a register to to try the test net. That's number one number two I think our Twitter accounts is something we get like a adk of charge. This is all kind of organic and the balance designer of course may help with the traction side as well, and so what else and see see the run?

H

The investment I think we have very good um uh investment like on investment list. So that's that's kind of like yeah, so.

D

um That's good I, don't have too much to add, and thank you guys so much for uh you know just doing this with us for a second time and uh yeah, and uh you know any other questions can always let me know um whether in a group chat, offline, Etc, yeah,.

B

G

Great I'm, sorry just one last thing so yeah how do I stay up to date on ZK pass? You know it seems, like you, guys, have a lot of great things.

G

um You know to me as a developer and something looking forward to this open source, stuff I'm, just kind of like you know, waiting for these things to come out. So how do I stay up to date? How do I know exactly when, when the tokens come out when you're open source like how do I, you know.

D

Appreciate it Keith uh definitely follow us on um Twitter and obviously you know put down your address for the uh pre-alpha test net and you know we'll keep you posted.

B

Awesome sounds great, so look for ZK pass on Twitter, LinkedIn and the social channels, um and we will stay in touch on when things are open source and follow along with your guys's work closely thanks for joining us again and anytime, you want to hop back on, let us know and we'll slot you in. Thank you. Thank you. Thank.

D

You guys have a good week. Everyone thank you.