ZKV presents Privacy in Cosmos: Oasis Labs Showcase
5 Jul 2021 · 10:46
Anne Fauvre of Oasis Labs presents a showcase of the company's privacy projects in the Cosmos ecosystem as part of ZKValidator's Privacy in Cosmos event. The talk covers how Oasis Labs approaches confidential computing and privacy-preserving infrastructure for applications built across the Cosmos interchain.
Speakers
Topics
Transcript
thanks to everyone um yeah so i'll start by just briefly explaining the problem that we're trying to solve here which i think everyone on here is very aware of already but really it's just that we know that web 2.0 build created an incredible new internet where you really can do so many things it's created entirely new economies it's created entirely new ways to use data but of course there's a number of problems with the internet as it sits today i think we're all very aware of this for us we think a lot about this in terms of data and data ownership and who has control of that data as well as many of the leaks that have gone along with it so many problems i think we're all aware of here so when we think about what can be enabled by blockchain technology in web 3.0
we really think about it as a way to return data ownership back to the individual and we think by combining privacy and blockchain technology we can create entirely new type of tokenized data and we call this the ownership of the ownership data economy uh specifically so this is where we come in as i've said and i'll give a little bit of background on how we're designed so the oasis network is a privacy first network um it's essentially designed with an architecture of two layers there's a consensus layer and as i mentioned that's actually based on um the tendermint code that's used for the cons for cosmos as well we love working with
that team it's been really awesome to to work with them um and then we also have a layer we call uh paratimes and paratimes are essentially multiple runtimes uh they allow you to do any sort of compute that you'd like so we have some that are decentralized and non-uh non-private we've some that are private we have some that are permissioned and private and and use privacy and confidentiality technology it's really designed to be quite flexible based on the use case that's needed um and then throughout this whole system the system's really designed to support uh tees um and secure enclaves at every level of the network so ttes for those who don't know are trusted execution environments and this is one aspect of confidential compute technology
so today i'll be speaking a little bit about two confidential paratimes that are being built and developed on the oasis network now and how what kind of applications we're seeing but at a high level what you'll find in these is that compute is essentially run in a smart in a secure enclave um in many secure enclaves uh in those run times and that actually serves as a bit of a black box for computation data is encrypted you run analytics in that encrypted environment and then results uh are essentially come out of that environment as a team we've also been investigating a number of other technologies and we're working on projects with enterprises
around the mpc space and differential privacy and are working on how those technologies can interact and connect to the oasis network as well so many privacy technologies but i'd say teas are really at the core of what we're doing today um ultimately what this gives us this design and these paratimes is a network that's really performant the design is quite scalable versatile and resilient so right now we use the cosmos consensus layer and you know our excuse me the tendermint consensus layer we're able to connect to that and be quite versatile in this way um and then confidentiality as well which i mentioned um so yeah as i mentioned we'll talk about two paratimes today one is what we
call parcel and this is a paratime being built by the team at oasis labs and this paratime essentially is built mostly for businesses who are interested in accessing and unlocking the power of data for their users and for their own businesses to enable trustless data sharing um it's actually designed to be a less blockchain like environment so the api interface is actually much more similar to a non-blockchain environment we've tried to abstract some of that uh some of the complexity away uh to really ensure that we're providing people with the uh i think in the experience that's easier to onboard onto especially for enterprises and then we also have a more um uh confidential smart contract paratime
that the oasis foundation is working on now that we we believe will really be used for privacy-enabled d5 apps as well dexes lending platforms etc and that's in the works that should come out in august um so i'll go into a little detail here on how we think about this so for parcel the way we're thinking about this is we wanted to build a platform that really works for enterprises to allow and enable something we call data back tokens and so if you think about this this is in some ways sort of the next step of nfts we think about nfts now as um you know a static way to represent and own individual pieces of a good those are mostly static like
pieces of art or images or videos but we really think that there's another layer of this where not only can that an nft be a representative of something but it can also be tied to a certain amount of value or data backed value not just some sort of speculation or excitement and so we call that a data back to nft or token and really what happens is that token represents the data that you upload onto the network and it's a singular representation of that data and what's really important here is that that data needs to one uh it needs to exist on a blockchain so you have some auditable record of how it's being used that you have control when to use it how to use it when to pull back
access of that data and then second this is where for us we've seen secure computing and confidentiality come in we've seen this as a really compelling use case for many which is applying privacy technologies to this to ensure that as data is used or as you're providing access for different teams your sensitive raw data as an individual remains protected and secure and so that's how we use confidentiality so to do this we've designed essentially this tokenization engine where essentially we are able to issue data back tokens and that can represent images uh you know videos etc but it also can represent future ip or future royalties um so that for example as uh data is being
accumulated or used the value of that token can go up essentially the second is we can actually then attach encrypted data to that um so you can analyze that data or have it to be used and then that's compatible across our ecosystem so it can be shared with users as they wish there's a little uh architectural design here see what that looks like and i'll just speak to a couple of the examples that we're already seeing being built so the first is we're seeing we're working with a company called nebula and what they really wanted to do they're a private uh version of 23andme so they do genomics kits for many of their users and they wanted to make sure that their
individual users owned their genome so the first step was actually setting this up on the oasis network directly and users actually have the choice to hold their network their data on the network and that gives them the ability to pull it back at any time it gives them the ability to describe who has access to it and when the next step is actually tokenizing that genome data as well what that means is that every individual at nebula will ultimately have a token that represents that data and as that data gets used by additional people or as it gets um uh you know uh used by different research entities the value of that token will essentially go up and so you'll see users actually being incentivized and rewarded
for uploading their data to the network as well now this might sound a little weird um and i think a concern we've heard is well do i want my genome on the blockchain but that's also where privacy comes into play which is by having all of this stored in a confidential way you can ensure that the data data analysis can be run on on that data without uh actually giving someone all of the sensitive data of your genome and in turn you own control of that we're also seeing this we're working with a music company we'll just call them banned crowd for confidentiality reasons at this time but essentially we're working with a company that wants to give more ownership back to musicians so the idea that they have is they
essentially raise money by tokenizing royalties and selling tokens to their audience so i as a musician i need i want cash now so i can essentially sell a future aspect of my royalties and then what will happen is the data from spotify actually feeds into the system and that links essentially the spotify data is how many plays i've had of my music and as i get more plays and as my royalties go up so too will this token and i can share that token with my users i can share it with followers of mine or investors and so they'll have some representative uh value that ascribes to the future value of my network as well so very cool to see we've been excited
to see this um play out and more to come on both of those examples i'd say the tokenization engine is in its final stages and you can play with a round with parcel and the paratime now and the tokenization engine will be out uh later this month so you go to oasislabs.com to try that out we have documentation on all of it and would love to have more folks playing around with it and trying it out the other one i'll talk about is cypher and this is a more traditional um blockchain network it essentially enables smart contracts confidentiality at the smart contract layer and this is something that we're working on now this will be a wasm-based smart contract environment using rust although we're working on an evm
compiler for this as well um and essentially we'll allow we we suspect and have heard that the the biggest benefit here will be private defy applications and so there's a few private d5 private dexes in the works as well as a lending protocol um and an amm so excited to see those go live again all of these are using uh tees um uh at the uh at the paradigm level uh to enable that that confidentiality almost done i just said this will be connected to ibc which we're excited about yeah and uh so many if you guys are interested in getting involved those are two i think great examples of privacy in action on the network we're thrilled to be a part of the cosmos ecosystem
the network as i mentioned we'll have a bridge to ibc very uh very soon later this year and uh yeah that's a good wrap up of some of the privacy applications in the works