Arcium: Parallelized Confidential Computing Network - Yannik Schrade. Ep. 564

Transcript

what we were designing was a system where users had this privacy utilizing Zer knowledge proofs and every private transaction that one performed um generated an encrypted transaction that was stored on chain and then we had a dedicated network of nodes that using MPC was able to screen those transactions essentially but importantly there was no trusted entity that could screen or block a transaction it was the network as a whole that ran Ure multiparty computations over those encrypted transactions arbitrary program execution in a confidential encrypted way but to also have the program itself be fully encrypted and to have encrypted

Random Access Memory this sort of replacement for you purchasing some trusted execution environment and now you can use this Global superc computer to execute anything if you're looking to stake your crypto with confidence look no further than course one more than 150,000 delegators including institutions like Bido pent capital and Ledger trust corus one with their assets they support over 50 blockchains and are leaders in governance or networks like Cosmos ensuring your stake is responsibly managed thanks to their Advanced meev research you can also enjoy the highest staking rewards you can stake directly from your preferred wallet set up a white label note restake your assets on IA or symbiotic or use the SDK for multichain saking in your

app learn more at car one and start taking today this episode is proudly brought to you by nosis a collective dedicated to advancing a decentralized future nosis leads Innovation with circles nosis pay and metri reshaping open Banking and money with hashy and nois VPN they're building a more resilient privacy focused internet if you're looking for an L1 to launch your project nosis chain offers the same development environment as ethereum with lower trans trans action fees it's supported by over 200,000 validators making nosis chain a reliable and Incredibly neutral foundation for your applications nosis

Dow drives nosis governance where every voice matters join the nosis community in the nosis Dow form today Deploy on the evm compatible nosis chain or secure the network with just one gno and affordable Hardware start your decentralization journey today at nosis IO welcome to epicenter the show which talks about the Technologies projects and people driving decentralization and the global blockchain Revolution I'm Sebastian K and I'm here with my co-host Felix luch today we're speaking with Yanik Shada he's the CEO and co-founder of archium it is a private compute platform that allows for all sorts of interesting use cases privacy uh in

crypto and Beyond so we'll be chatting with him today about archim the architect Ure their use of NPC and much more Yanik thanks for joining us today thanks for having me Sebastian on fixs I'm very excited yeah so tell us a little bit about your background and how you got involved in in the encryption and privacy space and how how you ended up working on arim yeah sure so um I think um the reason why why um I'm in the space of of building arum building confidential Computing decentralized confidential Computing and privacy technology at the end of the day

it might boil down to me as a small child reading 1984 I think that might be the the honest answer um I think that that really shaped um my my perception of the world and and my views of how the world should work and how um how privacy and freedom matters um and um reading that as a child I think really really influenced me um so um at a similar age I started programming teaching myself programming then studied law um for a bit and then um through founding my first startup pivoted to um mathematics and computer science in the process away from studying trust law um and then

through the study of of mathematics and computer science I met my co-founders um basically all of us caring about um elliptic curves at that point because that's what we what we learned about in University and then getting into zero knowledge proofs um and through the process of that end ended up um founding archium which back then was called elusive um and yeah with elusive we we focused on building transactional onchain privacy um and with the added twist I guess of of adding um trustless decentralized compliance um so that there's both privacy and still um safety

on the end of of the users so unwanted elicit Behavior could be excluded from from from this um yeah onchain onchain privacy architecture um and for that we leveraged confidential Computing with secure multiparty computations um and through the process of Designing the system and building this technology what we realized is the the huge potential um and the Revolutionary potential of being able to run arbitrary computations over encrypted data without having to decrypt the data first so um data in use can remain fully encrypted and anything can be processed um on top of that data and so um that was really a pivotal moment

this realization for us um which then allowed us to to evolve further into arum and fully Focus just on providing this what we like to call distributed Global super computer that allows for confidential Computing awesome thanks for the quick overview we're going to dive a lot into arum over the course of this of course I think for me also interesting because that's like how I got to know you I think working on Elusive and also like working in in the Solana ecosystem can you maybe break down how you are you started in salana and in know kind of why you how you came to the conclusion to at this like compliance angle I think

that was like kind of like one of the main differentiators there in U in in what you were building or in salana yeah sure um so um I think how we got into salana is the is the story of of many teams that are building on Solan was really the the developer ecosystem um and so um the the final co-founder of our team actually I met at a Solana Hecker house so um uh in the year 22 there were a lot of hacker houses and and um this really um allowed developers wanting to build new interesting projects um to meet up to to talk with experienced Founders and then um yeah

test test the systems that they were building so um that's what we also did how we came um came about building on Solana um and um yeah back in the day with elusive um what we what we were building was essentially a cach like onchain privacy system which utilized zero knowledge proofs um as as core primitive and the issue that um basically all of those systems face at some point really is um compliance concerns right because um we've we've seen that with tornado cach essentially right so um with tornado cache um there were enforcement actions by different

governmental agencies because yeah they claimed that tornado cach um didn't prevent malicious um behavior on the on the platform and if one utilizes mathematics um to provide perfect privacy guarantees um um yeah that's that's the issue that they run into and so what we were designing um was a system where users had this privacy utilizing zero knowledge proofs and every private transaction that one performed um generated an encrypted transaction that was stored on chain and then we had a dedicated network of nodes that using MPC was able to screen those transactions essentially but importantly

there was no trusted entity that could screen or blocker transaction it was the network as a whole that ran secure multiparty computations over those encrypted transactions and essentially in a within a blackbox this virtual Black Box we're able to look into the transactions and um assess what happened and then they could find external consensus over um yeah malicious parties that that tried to use that system and then after the fact um reveal reveal the the corresponding information um and so that was extremely powerful because um yeah prior systems really attempted um to to add compliance by screening

transactions um up front um and with this systems uh a screening after the fact uh expost was possible um which I think um would be the ideal solution and that's something that um is possible to to build with arum um but um we really realized that um yeah I think through the technology that we were developing and we'll we'll dive more into this later I guess um but um the complexity of this MPC technology I think there was just just um one area we had to focus on and we really saw our strengths in in building this Computing platform and um optimizing the compute both for

trustless but also performance um and so that's what we ended up doing and you in the case of uh you know you mentioned tornado cash like I when I when I heard about that I I couldn't help but be reminded of you know the the cyppher punk stories of the 90s you we had Mark Miller on the podcast tell us about how you know he used to print the RSA algorithm on T-shirts because the US government considered that encryption Technologies were um were considered under uh US law to be um to be weapons like to be Munitions and so therefore you you couldn't export them and there was this whole um you know legal threat and sort of uh threat of persecution on on those people in the 90s you know that

that were simply just freely expressing uh ideas and and and to the point like just making math public uh and so like I I heard that you know the tornado cast story and like it it felt to me like very similar where you know these guys basically like wrote math that this stuff was this stuff exists uh it cannot be stopped and and I think that we'll look back on turn cash in the same way that we look back on you know the the suffer punks in the 90s and what they were doing um there's really you know little one can do to prevent malicious activity when you're using en cryptic Technologies and I think it's like a huge societal debate but I hope one that you know people that sort of subscribe

to the idea that encryption you know is just freedom of speech and you know is sort of like a right um I hope that you know those people will fall on the right side of History you know having you know you you said earlier you came into the space and you were kind of inspired to work in crypto having read 1984 you know how did you how did it make you feel that you know you were sort of constrained to to build a technology that enabled compliance um when you you had such this you know this very strong background and and sort of ideological pedigree of like non-compliance in in the in in that sense yeah so um I I think that

um so first of all yeah I'm am 100% on this on this um Cipher Punk track um I think one of my greatest um um accomplishments in the in the Privacy space would be um getting my entire family to um to your signal and having the entire family group chat um be be present on Signal instead of WhatsApp um so I think this this compliance question um is an interesting one and the way I think about it really is um that we gave control in the hands of the people using it at the end of the day there was no external action of enforcing compliance it was essentially

being able to have distributed consensus um and giving users the option to use whatever system um so it's um the design that that that we designed back in the day with elusive really was um decentralized compliance um so no centralized actor should be able to to have any control it should be the users that have control and I think um at the end of the day and um yeah we never ended up getting to this point to to see that in action and and experience those um those those Game Theory and network effects but I think at the end of the day um yeah it boils down to do users

want specific other users um using the servers for um specific kinds of of of use cases um and I think it's a bit more more easy um to to have these mechanisms um with this um yeah with those financial mechanisms attached um whereas it's way more different if it's just pure um yeah expression of ideas right so um I think that's a system that made sense that um at the end of the day found a natural equilibrium between both ends yeah it's it's hard not to not to also bring up you know the the whole whole um story around the the telegram

founder um who was arrested in France uh I think like last week um yeah any any thoughts on on on that and you know what it means for the for the state of privacy Technologies uh in Europe yeah I mean to be honest it's very it's very difficult to say um I think telegram is is is difficult is difficult for me um because um on the on the encryption side of things um telegram always um especially in yeah more more mainstream media I think um is being portrayed as this

super encrypted private chatting platform that that cannot be hacked um whereas in reality um there's no default um yeah encrypted chatting between two peers right um so um I think the story is is difficult yeah it's it's it's interesting to see um at least like you know here in France I've been watching mainstream media talk about uh talk about telegram as this uh encrypted messaging in fact they they they they don't even use the word encrypted they use this this other word that in French that that doesn't even mean encrypted I mean you know the and it's like it's as if they were calling it a you know a cyber messenger instead of an encrypted

messenger uh and um and it's it's really painted as this kind of like blackbox thing right that you can tell that the story is spun in such a way when in fact you know like every anybody who uses WhatsApp you is in fact using you encrypted technology and anybody who uses Apple messenger is using encrypted technology and telegram is really painted with this with this very uh negative sort of lens so it's it's interesting to see how you even the mainstream media here and I'm sure probably in Germany and other places as well is like very biased against these Technologies for some reason and but anyway not to make this all about about that um you we do want to talk about um archim and and what you guys are building so let's let's talk about you know confidential Computing so you know what does that mean and what are the

different ways that you know it has been implemented in the past and what's kind of new uh about the the way you guys are going about building it so yeah I think confidential Computing has been around for for quite some time already um and at the end of the day it means that computations can be performed over encrypted data um without having to decrypted data so it's it's about so-called data in use right we have data in rest um which just lies encrypted on some hard drive but data in use is the data that actively is is being used in a computation and so confidential Computing allows for that data in use to be secure that's what

what confidential Computing tries to achieve and there's different different um use cases I guess for confidential Computing um for one creating a secure execution environment right having sensitive data critical um critical business data for example right that that has to be executed over in in some Cloud environment and if that data were to be leaked um um bad things could happen to companies Enterprises um governments or individuals so um building secure execution environments um at the same time confidential Computing um allows for data collaboration essentially so um

individuals can have their data remain private while being able to interact with others so those are two core aspects of what confidential Computing tries to achieve and um there are different kinds of technologies that um have been utilized in the past um to to achieve this and I think the most prominent one um so far has always been trusted execution environments um and um a few months back I was in in San Francisco at the so-called confidential Computing Summit which essentially is just a conference of um trusted

execution environment um manufacturers um and Intel Microsoft um all those folks there um praising their trusted execution um environment technology and um yeah I gave a talk there and I called it trusted execution is dead and we have killed it and so my take really is that um there's new kinds of Technologies now um that can replace those Legacy systems that require trust so um what we're trying to achieve at archium and what the entire decentral calized confidential Computing DCC movement I think to some degree is trying to achieve is to um add more trustless to to confidential Computing and in our

case that's by utilizing multiparty computation so maybe you can kind of I guess bring up the tees already and we have seen over the years a lot of different types of attacks on these I think uh yeah like side Channel attack um kind of supply chain tax um and still like like you're saying right in the summit everyone's talking about that I think also in crypto space right now te seems to have another like sort of resurgence and the popularity um is that or like you

know why is that maybe and how does MPC approve on that or like how do you yeah tackle that at arum like the sort of and maybe explain also a little bit what these types of attacks actually are yeah yeah sure so so trusted execution environments are dedicated Hardware um that um comes with security and privacy promises essentially from from the manufacturers um so um usually that's dedicated Hardware chips um that have those secure enclaves is what it's called so virtual machines uh Within window those chips um where data should remain secure encrypted and cannot be

accessed outside of that of that enclave and an enclave um has a code base associated with it so only this specific code base can be executed over that over that state which um sounds like a great concept sounds like a difficult to realize concept at the same time um thinking about um exploits and hackers so those systems suffer from from from many problems I think the most easy to grasp problem is the um actually the complexity of building building architectures on top of trusted execution environments anyone who has worked with with tees will will tell you that it's very difficult to build um

secure code on top of those enclaves because developers really have to um have to ensure that data is being handled correctly and they need to give rigorous attention to detail um that the boundaries between um the secure and non-secure environments um are being respected um and that also at the same time brings um yeah increase time to Market I think because there needs to be a very um very thought through development processes and those systems at the same time also um on their own come with quite high costs associated ated it's dedicated hardware and um

dedicated Hardware that requires specialized folks that that are able to maintain and develop um on on those platforms those are more of the soft factors that I don't like about working with trusted execution environments and the bigger problems really are associated with the trust model as the name already communicates the execution here is trusted and um the trust model in my opinion is fundamentally flawed um because as you said MTS are really susceptible to a whole range of of of um more or less sophisticated attacks and um I don't know if you guys saw this but

one and a half weeks ago or something like that I think there was um some um some researcher on Twitter that was able to extract um the uh root provisioning key um from some um Intel sgx family processors um which is a key that can be used to to fake so-called attestation report so trusted execution environments require this process of attestation where they prove to you that they are a real trusted execution environment and you can trust this environment so you can send your data in an encrypted way to that environment and this attestation service um yeah requires some third-party Trust

um um yeah regardless on the processor side of things um there's side Channel attacks and Hardware vulnerabilities and side Channel attacks really um boil down to you as uh as the person having access to to the processor or you um being able to just track the execution of some program execution and through this process of being able to look at for example power consumption um or um or tracking other metadata I guess um that get gets exposed through the execution of a program be able to understand the execution path in a program right so if you think of an if

else statement and if some some condition holds something complex is being executed it takes one hour and in the else case the program just ends if the computation takes one hour you will know okay um the condition was true because the first Branch was executed and so that's that's um the most simple form of uh performing a side Channel attack right and so those processors are prone to these kinds of um side Channel attacks and there's um many many um exploits that have occurred in the past and many fixes for those exploits but it's always new um exploits that that get identified um so in the blockchain

space we've seen a number of those kinds of exploits actually um I think the most notable one would have been um secret Network I think that was end of 202 um where there I think it was called a seat key or something like that some some master key that was shared between all all the nodes in the network all the tees there um was exfiltrated um and then anyone who ever ran a note in the network would have been able to um yeah REM remove the privacy of all transactions right so um there's this inherent danger of side Channel attacks um firmware micr code and SDK bux it's just this large development stack with

um incredible complexity and both on the hardware but also on the software firmware micro code level um where at the end of the day humans are building those systems and so humans in most cases are also able to to exfiltrate information those are one kind of of exploit but I think more important is as you as you mentioned Felix Supply chains because when someone trusts a trusted execution environment they're trusting a supply chain uh usually proprietary supply chain which at the end of the day is just a chain of single points of failure um single points of failure that can occur in the attestation process can

occur in the uh manufacturing process and I think uh very striking um example of how trusted these systems are is um when when Apple unveiled their private cloud computing platform which is their own tees that they they supply for uh for for um model inference um for their apple intelligence I think something that they stated in their docs in their um release article was something like okay they physically ensure that the um that the machines from the factory get placed in their um in their data center right so you can imagine um

people standing there with their assault rifles protecting crates of of of uh computer chips uh which is insane right if that's the trust model that you're working with that there has been someone that was able to physically guard this chip from A to B um I think that's not something that um that we should put on our trust and so I I wanted to ask you about like the because you know we've been focusing on sgx here and I think like in crypto at least you know sgx has gotten a lot of negative uh attention because of the secret hack and um and and and some other issues and like I'm I'm here on the Wikipedia page for for antas JA and there's about you know a dozen different

attack uh vulnerabilities that are mentioned um but so why is it then that you know we never hear of like the Apple te and our phones being compromised or you know even perhaps more interestingly like Ledger devices uh you know they they claim that you know they're super secure and and they have like this whole uh um what do they call it the the the do the dungeon or something like that you know their their their whole like security lab that just prize all day at trying to crack a ledger so why why are tees different or or are they you know is this just misconception or lack of understanding of the

technology yeah I I think um as ja has gotten a lot of attention the architectures are are significantly different in the details of course but um the core architecture and the core Reliance on on these kinds of Supply chains um Remains the Same and um yeah I think in general um there's um yeah I think think so so we can we can go more into that how I think trusted execution environments can play a role in the future I think they can play a role but um yeah just this this approach of using these systems and saying well so far we

didn't see any exploits so we can be sure that we can trust those systems and we can use them for for all all future I think that's a that's a flawed flawed approach um because at the end of the day it's it's just um the sort of nonpr blind trust um and in the case of Intel sgx we've seen how how fatal this can be really and um I think what's important um and that's something that I've I think never seen anyone talk about when when talking about these kinds of Supply chains um um in the case of inas GX that's something that that uh we've seen with different kinds of of team that

that use that technology was that um okay there's the supply chain and Trust into Intel essentially and the person that has access to to the to the computer trip but there's also someone who deploys the code for the um for the enclave and they upgrade the code they they they give updates and they have a private key they use to to sign those updates and so there's a single signing Authority that can update whatever code they can have their own local trusted execution environment update the code and exfiltrate information and at the end of the day in most crypto projects that use trusted execution environments yeah easiest single point of failure

really is some development agency um sitting on some Island I guess um that has a private key that has full control over everything um I think that's the that's the most shocking shocking aspect and that's overlooked a lot when when talking about the trust in the manufacturer most most of the times um it's it's even simpler than that it's a $5 wrench Attack Just you have to get the boat to the island yeah exactly okay yeah that that's super interesting I think I guess we're we still haven't even talked about NPC so so let's uh get there I guess you know we're basically focusing

on these Hardware based models but now we also have like purely cryptographic Ally based privacy techniques and um how do they work and uh like what are their trade-offs and and how do you like navigate that tradeoff space I guess that's going to be the discussion for the next few minutes hopefully yeah yeah sure so um essentially um the other Technologies are zero knowledge proofs fully homomorphic encryption and MPC um zero knowledge proofs we've spent a lot of time working with zero knowledge proofs are um yeah great technology and at the end of the day Zero knowledge proofs are even part of those fhe and MPC proving systems so um but but CK

snarks CK snarks those those more more prominent proving systems um yeah they don't allow for what we are trying to achieve what we're trying to achieve really is to be able to have some encrypted data send it to the Some Cloud environment some some blockchain Network whatever send it to some some some computer and have it processed without that party that processes it um having access to the data and and zero knowledge proofs are this um two-party protocol where one party is the approver that has access to the data and the second party the verify verifier not having access to the data so zero knowledge proofs on their own um can't allow for those autonomous confidential

execution environments fhe fully homomorphic encryption um allows for that um essentially the def definition of of fhe would be um to be able to to um execute arbitrary um operations over encrypted data so um there's an encrypted representation of of some State and with fhe um two operations essentially addition and multiplication are homomorphic um over that um representation which means that they yield an output that again lies in this representation so the um is a very very good concept um and um

the problem with fhe really is the Practical um usage of the technology um because um the multiplications um for fully homomorphic um encryption um result in noise accumulation and noise accumulation requires um bootstrapping so bootstrapping operations essentially reduce the noise of this encrypted output and if there's too much noise that output can't be can't be processed anymore so um bootstrapping is required and that um results in very high very high um performance latencies and and computational overhead and that means

that when using these kinds of um fhe systems um there's many orders of magnitude um performance latency Associated and so um um it's It's Not Unusual and I think um comparing the the NPC protocols that we have with some fhe operations um It's Not Unusual for us to see um something like us being faster 880,000 times something like that so um incredible um performance differences and for very very simple operations um fhe can be used um but um yeah more complex computation

um make this technology fail at the end of the day in in Practical implementations and so the system that we are using um is multiparty computation and this multiparty computation uses so-called um somewhat homomorphic encryption sh and somewhat homomorphic encryption basically um uses the efficient aspects of homomorphic fully homomorphic encryption that we we also utilize there but for multiplications there's a smart trick Associated I guess that allows us to to efficiently perform those multiplications um as well and so um that's why this technology makes most

sense and on top of that um what's also very important to realize with fhe is that um moving from encrypted representation to encrypted representation is very nice but if what if um one wants to move out of the encrypted representation right if we have confidential onchan applications and we want some transparent settlement to occur um there has to be a way to selectively disclose information about the um the encrypted State and that's not possible with fhe by default and with NPC that becomes possible um and so trust model wise what's the case is that the trust models for for systems and

practice is the same for MPC and and fhe applications with mpz being many orders of magnitude faster and so um that's why we arrived at um building those systems with MPC super interesting and do you think like I guess in general this is also something that FHA can ever address in a way like is it something that you know over time the algorithm will get better and like probably right I mean but yeah we'll be curious about that that mostly boils down to to Hardware acceleration um so Hardware acceleration is is what matters most for that there will be significant improvements um for sure um and what's

important is that those Hardware acceleration improvements also um get utilized by this somewhat homomorphic encryption system in in in in MPC um MPC still requiring network communication um that's that's that's one difference requiring some more Network Comm communication at the same time um fhe um in order to achieve verifiable compute with FH you also require um consensus mechanisms if you if you do it in the in the blockchain setting um if verifiable compute which is highly important for for any computation especially confidential computations um um if if verifiable compute is achieved

without using consensus there's even more um significant performance slow down so Hardware acceleration is what it boils down to and um I think these kind of Hardware accelerations will also um be utilizable for for the NPC protocols um especially with with reent um yeah with recent advancements in those those protocols um greatly reducing the the need for for um network communication um so there's different um yeah options to to be able to um pack a lot more data within communication round so um yeah I think at the end of the day um f will um

become more performant but um yeah at this point in time NPC just makes more sense so it's always about um yeah being able to supply the required privacy technology um at this point in time and for the next years and so um it's the same with um with um Fusion uh energy I guess right um we also need energy now and for the next 10 years um without having Fusion Energy so does this work with any type of computation can you do can you perform really complex computations using uh arum and um what what languages

can one write computations in NPC does this implementation support like specific languages can you maybe explain that how what what we trying to achieve essentially as as um as the vision with arum um and that's why I like to call it a super computer really is for arbitrary program execution in a confidential encrypted way but to also have the program itself be fully encrypted and to have um encrypted Rand Random Access Memory encrypted Ram right so to have a fully encrypted computer you really um have this sort of replacement for you

purchasing some trusted execution environment um now you can use this Global supercomputer to execute anything arbitrary um programs that um are essentially arbitrary encrypted Ram programs um the current stage we're at um is that we support arbitrary computation um support and uh for that we have our um a dedicated compiler um that compiles arbitrary rust code um into the um into the op code micro code format that our Network understands yeah so um one could essentially be uh be writing assembly code so on the op code level what the

what the network understands but um what we offer mainly is this dedicated um yeah rust compiler now depending on the use case um different languages make more sense so um what we have at argum right now is two distinct back ends to distinct MPC protocols one MPC protocol that is um highly focused on being um as trustless as possible um for especially onchain applications um and we have another backend that is um fine-tuned for um floating Point operations so in order to perform um operations over um large matrices with floating points for machine

learning um and so for that back end we we support a um a python um SDK essentially um so one can use tensor flow and pendas like they would normally um and now have it be confidential and have one party hold this data another party hold another data and then collaboratively train um yeah some logistical regression or or tree boosting um model with that so um essentially rust and python 4 for the machine learning applications that's what I would say yeah awesome let's maybe also take it back to like arum as this network

overall right like so we talked about you know why you're using MPC and then confid naal computation but to achieve this at scale right there is a a network and there is the crypto component of it so can you provide us like a overview over the arum structure architecturally as a as a crypto protocol and and how how that functions essentially arum itself is a you could say stateless Computing Network so we utilize um existing lures um existing blockchains um for State

Management and computation scheduling so um you can have your own smart contract on Solana for example that um has confidential functionality and this smart contract then um calls an arcum smart contract that inserts a new computation within the RQ mempool the network picks that up executes the computation and settles it back um to the corresponding L what's important is that the argum network itself you don't write a smart contract for arum um you you can use our um smart contract um SDK and build a confidential smart contract but the network itself just runs

confidential computations and it picks those up from those dedicated onchain M Pools that can exist on different lures and so I can build a confidential offchain application and send such a computation request to the mol um and it doesn't have to settle back to the to the dedicated smart contract um now how the network itself functions is essentially it's a permissionless network of nodes and so um the three of us could deploy a Noe in arum and we could open a so-called computation cluster so um our computation cluster could now be used by anyone to run confidential computations or or we could

restrict that computation cluster to say okay we don't care about providing computational services to third parties we just want to run our own confidential computations so it's fully configurable um you can have clusters um containing nodes from the network you can run your own note um and what's so highly important um about the trust model Associated there is that it's the this honest majority trust model so any computational cluster only requires one participant to to be honest one participant that doesn't um collaborate with all the others participants in a malicious way so um with that trust assumption you can um have arbitrarily sized computation clusters um and um

associate them with a so-called mxe MPC execution environment so that's essentially this virtual machine instance that you create as a developer you create this virtual machine instance where you have encrypted State you have your functions that can be called um essentially analogous to a Smart contract if you will um and that environment has one or more clusters Associated that then process these kinds of computations um now the problem one on a theoretical level would run into is that um such a computational cluster has this ideal trust assumption only requiring the trust that one note is

honest right but um for our blockchains we have with with practical Byzantine fall tolerance we have the honest majority trust assumption and here dishonest majority trust assumption one out of n really is perfect but the problem is that this also um becomes highly dangerous for censorship because um one malicious party could also censor computations and um that really has in the past been a practical limitation for those very effici ient and trustless MPC protocols um because yeah deploying them in this permissionless setting could really mean that some malicious party could just DS the operation essentially and so what we

have is a new kind of cryptographic protocol that allows to enforce um execution and also correct execution um um for for these computational clusters so if you have this computational cluster um and our cluster in this example and Sebastian just randomly decides to um share some invalid data with me um at some point I will notice that invalid data has been shared and I will stop the computation um as I'm an honest participant um the problem now is that by default Felix and I wouldn't be able to cryptographically identify that Sebastian was the one that

caused the computation to fail by sharing invalid data now with our new protocol we are able and actually everyone in the world viewing um our our computation from the outside is able to identify that Sebastian is the one that caused the computation to fail and what that means now is that we can punish Sebastian for causing the computation to fail and so that's where blockchains really come into play that we are able to enforce execution of those computations um and enforce this kind of correct Behavior so we are able to add accountability um to those um computations which makes it incredibly

powerful now because then even smaller sized clusters um yeah can be performant run those computations um and the the deployer of of these clusters can can be certain that execution will occur um and so that's where a staking and slashing mechanism comes into play so compared to other you know privacy projects out there um you know how do you think arum is fundamentally different like what what is really the the the key differentiator that you guys are offering as a product yeah I I think um that really boils down to be trustless and

performant at the same time um because what we've seen really in the past is that um that systems either require trust via trusted execution environments or have incredible slowdowns with fully homomorphic encryption and um this sounds stupid uh saying that as as the founder of archium but um what I've seen in practice and we are right now moving into um our private test net and have teams um building with arum really is seeing a lot of teams that have attempted building complex systems with fhe um and yeah pivoted to using arcum because um on the one hand it's it's

easier to use and on the other hand um it um it just yields um the performance requirements that um applications require and so um I think that's really what it what it boils down to and um at the end of the day um all of that sort of is user experience although trustless might be more hidden from the end user and uh you know in terms of applications you know obviously there's applications here in crypto and you know I'd love for you to maybe talk a little bit about some of the applications that you you're seeing and that users initial users um are are

building but you know applications here go beyond crypto I think you know if we're if we're talking about the The Trusted execution environment Market you know potentially you know clients of tees of you know companies that use tees could um shift using um Technologies like arum uh maybe also describe some of those use cases and you know efforts that you're making to address that part of the market as well yeah sure so um I think that's also one of the um more interesting aspects to to to building arum um that it's not just um yeah solving problems in a crypto space but solving cryp uh solving problems um for

the entire world for for both Enterprises governments any kind of of organization I think um and um that's also essential to to the architecture design that we've we've designed and I mentioned that earlier that you as a developer don't have to deploy a smart contract or build a smart contract in order to run confidential computations you're essentially just accessing This Confidential Computing Network um and what that means is that um on the crypto side of things um there's a lot of use cases um especially in Defi and deep in um there's a lot of teams building um building amazing stuff with with our

technology um and also um I think and that's I I think something that um is less important to the crypto space actually and more important to the to the traditional Computing space is being able to run confidential um AI more specifically um confidential machine learning with arum because in the past um confidential machine learning really um hasn't been possible and um with this NPC based confidential Computing what's so striking is that completely new types of digal interactions I guess become possible um because if all of us have isolated data silos isolated data sets of highly

sensitive data that it on its own is valuable right all of us care about our privacy we would never share that data but um this data if we were able to in some shape or form pull it together and uh train AI models um using using that data that could become very valueable and I think a very striking example for that would be Healthcare right so very sensitive patient data um that now can be utilized in a trustless way new models can emerge that um can help all of us without any one of us ever having to expose their sensitive data um and this really is this collaborative element and um what we've seen is even

um logistic firms right so um we we talking with listic firms that um want to improve Supply chains um regardless of blockchain they don't care about blockchain all they care about is not giving their sensitive supply chain data to to competitors but at the same time um they can improve the supply chains um they can all have um the sort of win-win situation um by using this trustless um yeah confidential Computing technology so I think that's really the power of um decentralized confidential Computing to um to address traditional markets as well and um that's also what what drives us that it's not just building

applications um for crypto sake but building applications for Humanity's sake at the end of the day I think so when um when considering the you know the rkm network are like I think one of the um one of the important things to consider here and I'd love to get your take on this is that the censorship risk so you know we in the case of like uh an application that would uh handle the healthcare data of like an entire nation um if we had like a small number of nodes there would be a risk of censorship and so therefore you'd probably want to have as many nodes as possible in there um does are there uh basically so exit mechanisms to prevent

uh censorship and also does does the Network performance is the performance maintained as the number of nodes scale or do we end up you know does that uh start decreasing as you add more notice to the network yeah sure so um we have we have multiple mechanisms to to to combat censorship I think the the first mechanism that I outlined earlier is um socalled on on a technical MPC um level it's called cheater identification so to be able to cryptographically identify anyone who who who tries to stop a computation from from occurring and we can we can force everyone game

theoretically to run a computation otherwise they get slashed they get punished um now if such a misbehavior happens and the misbehaving party doesn't care about um losing all of their Stak assets um there's fallback mechanisms there's so-called cluster migration so to be able to move from one cluster of nodes to a new cluster there's also cluster forking mechanisms so that nodes that um want to zensor because maybe there is some um some some some kind of computations happening where one node says okay I'm not com comfortable um processing these kinds of computations so there's really also this

tension of of sensorship um um because um we want to force nodes to run any any computation at the same time there might be um based on on where the the node is being operated also legal implications to not be able to to process certain computations um and so we have both mechanisms we can allow for nodes to say okay in the future I will not want to process more computations for this corresponding mxe and they can do that they need to pay for the so-called cluster foring then um and there can also be um those forceful migrations if a node um just doesn't jointly compute the function that they have to

compute right maybe taking it back to the use case for like a potential final question we talked a lot about arq being this platform and actually you already also mentioned initially you were building Like This Confidential confidentiality thing on on elusive right like kind of more application focused almost or I guess in general The elusive product was more application now you're this platform and obviously if you're a platform you have to do like ecosystem building you're mentioning you've been speaking to a lot of like Logistics firm like in healthare like traditional businesses um and I think you also like mentioned like you yourselves met in the Hecker house ride in in Solana which I think quite

interesting and probably one of the best ecosystem building examples in all of crypto in the recent years right seeing like it brought also you together and just generally how successful uccessful salana has been uh with that so yeah the question basically is like how are you approaching this how are you getting people to build on on the arum platform yeah so um we currently have way too many people trying to build an arum um that's um that's the current status so um a few months back when we when we announced archium we we started accepting um developer and node operator signups um for for the private test net um which we started rolling out and

currently we are in the cohort one phase so the first group of of developers that um get Hands-On support from us um get access to to all of the tooling and and can start building applications and step by step we are we're expanding um those those cohorts and and adding more teams so folks can join our Discord um Can can register to be able to to um be accepted to those private test net development rounds and then they get full access to arum um and can build their applications and get um yeah development support from from our end that's say congrats on the success there for sure yeah I hope we'll

we'll see like a bunch of cool applications uh SE any anything else from your point or no I'm uh just like always uh amazed that like these Technologies you know as great as they are you know the the technologies that we really need you know like private compute more privacy Technologies are not the ones in crypto that you know get the most attention or uh get the most capital and so I'm I'm just really glad to see arum uh like you know raised 5.5 million recently and you know from some pretty big VCS and um and hopefully like take

this technology to Market because I think it's it's really important you know I also at some point you know brought all my family to signal and thought that was a really big achievement and even though you know signal might be back doored and we have no real way of finding out um you know but uh um yeah I think like this stuff is important and um and it's also political I think we shouldn't we shouldn't forget that like encryption is a political issue more than just a technological issue and so I'm glad we can talk about it here on this podcast thanks for coming on Yanik thanks for having me guys