What Is The Big FHE Unlock? With Guy Zyskind

20 Dec 2024 · 27:47

The problem is adding confidentiality to smart contracts without making developers rewrite existing applications. Guy Zyskind argues that FHE can compute over encrypted user data while ZK and MPC remain complementary, citing sealed-bid auctions, hidden token balances and private payroll as target uses. His proposed path moves from an Arbitrum-based L2 toward a co-processor that apps can tap for selected features, but he says FHE remains expensive and will probably require specialized hardware.

Speakers

Topics

Transcript

asking people to rewrite everything that they do and come to us and changes like it doesn't make sense what was it that turn turned you on to FH you said you were maybe a disbeliever 3 to four years ago what flipped and I'll let you in on a little secret I think welcome back to the rollup your home for technical talks and open conversations that shape the modular future nothing on today's show is to be construed as Financial advice you can see our full disclosures list at the roll up.com co/ Ventures sit back relax and enjoy today's episode after a quick word from some of our awesome Partners Redstone is def's modular Oracle supporting over 100 clients on over 50 chains offering 1,000 assets

Redstone is the leader for Bitcoin staking and yield bearing assets they offer the most robust customizable price feeds available in any model join the Redstone exhibition [Music] today guy welcome to Bangkok thank you thanks for being here thank you for having me we're at uh we're at ARB verse and uh having a lot of fun um I think time and time again we've heard this this concept of uh the casino funding the research lab and uh I think that is particularly true after we've seen a couple of green candles The Vibes are up uh the casino is working as intended but

there's a lot of energy being put into the tech and the research that is enabling this entire ecosystem um I I so you know I want to talk about FHA and get into some some of that um generally how are you how are you uh thinking about cryptography you know cuz meme coins and and Defi and yield farming gets a lot of the flashy lights yeah but that's really not why this industry is going to survive the next 5 10 20 50 years I mean I mean this might be a somewhat unpopular take from someone more on the research side but I've been

in crypto for really long like since 2013 and I I I totally agree with you that like you know the casino funds the research lab and and I kind of started from that like ignoring the casino and just being on the research side but then over time I kind of fell in love with the casino and with everything like I feel like you know this is almost like the biggest social human experiment that we have going on like like potentially in my lifetime In Our Lifetime and so I I see it as like something that's like by direction right it's like it's like I'm perfectly fine seeing like meme coins take the spotlight H I know that the technology we build will eventually underpin that use case but also many others and it's fun it's like part of

the fun in games as long as no one is getting hurt I mean I'm good with that but I think it's it's it's it's all like you know there's no way to really cut and and I think that's a realization from the last like 3 years or maybe slightly more there's no way to cut crypto in between where it's like this is research this is the casino right they feed each other yeah absolutely it was kind of a match made in heaven I mean the the technology that enables a lot of these a lot of these Primitives uh has only been possible in the last two to three years because we've had so much brain power and energy that has gone into this Bas layer and so we talk a lot about ZK and programmable cryptographic methods but I think fhe is one of those things that actually

vitalic called it like this Egyptian God card that is is potentially going to like really cause this paradigm shift in the way that we're building and thinking about these systems and so why do you think FHA doesn't really get the the love or like the the same credit that you know ZK does is it just because it hasn't been around as long or you know what why why is that uh two reasons one is like time yeah yeah I mean it has been around for quite long like 15 years give or take but it did seem like such like Moon math like even more than ZK that even people like on the cryptography and research side they were very very skeptical like you have to understand people were skeptical of ZK

like 10 15 years ago people until like 3 to four years ago were very very skeptical of of FH myself included by the way and you're only in the last like few years you're seeing the transition so that takes a bit of time uh that's a I guess that's that's the main that's the main aspect of it and the other aspect is that you know not everyone still knows it like people are still getting familiar with it what what was it that turn turned you on to FHA you said you were maybe a disbeliever three to four years ago what what flipped so I'm an MPC researcher by training so I come from NPC private compute ZK all that um I made a mistake I was one of those like when uh before even zcash

came out I was one of those who said it's too slow it's maybe going to be for like one use case no's going to use it and I was obviously wrong like ZK went like this you know and like and and people didn't expect it myself included so that was like reason number one I'm kind of seeing with FH a similar trajectory as ZK but just with like a five or seven year delay um the other aspect is that you know I thought I found it also secret Network before so I thought tees are going to be enough I think tees are very powerful but more is a def face in depth I don't think they're enough on their own and again I'm an NPC researcher so I thought NPC would get us to where FH will and we can talk more about it later maybe but like

I've come to believe that FH even though today there may be on Park give or take FH is going to is going to like grow and scale much much faster and and I think what we're realizing is that like when when one technology comes out it feels so Innovative we feel like it's going to it's going to solve all the problems yeah but then as it goes we realize that NPC is good for one one class of problems ZK is is good for another class of problems and and there is no Silver Bullet and so my question is what specific class of problems is fhe good to use putting aside ZK and NPC what is it specifically that you know the other

programmable cryptographic methods aren't able to do that FHA is I mean FHA allows it to compute over and crypted data yeah like you have a transaction you know maybe it's a transaction for a smart contract that smart contract has like user data you encrypt that data and you just run it over the encrypto data that's it no like ZK can do that MPC does it in a very quirky way so I don't think that's the right technology to do it um but you're right you need you still need to mix all of these right like ZK is the proving technology like even with FH you you'd want to do like ZK over FH when that's feasible even MPC like tomorrow I'm giving a talk at our

encryption Day event and I'm going to talk about NPC work that we did to basically split the decryption key so that you know yes there's FH encryption thata is encrypted but sometimes you need to decrypt data who holds the key so you don't want one person you want the network to kind of split so all of these come together but when it comes to Computing over encrypted data and really really kind of like the core privacy aspect of not seeing data that you operate on F is like the the way to go yeah and I think another misconception here is that uh these are mutually exclusive and so you can either pick ZK or you can pick fhe but that's not the case you can combine these two technologies for really really powerful use cases um and and I'm sure you have

ideas and you're imagination runs wild here um but you already kind of mentioned them you can you can uh compute over encrypted data uh and then do ZK on top of that or I have a feeling you can do ZK underneath and then you can run fhe on top and so there's all kinds of fun like uh use cases that you can use multiple different uh privacy prever preserving uh Technologies and and get a get blockchain to a place where uh web 2 traditional Finance is right now because everything is centralized and safeguarded we can actually do that on chain in a programmatic way right and I'll let you in a little secret I think

every researcher operating in the space knew this knew that there will be this kind of like combination and we will get to this end game of combination I feel that I don't know if it's the the market the ecosystem the vises whatever you want to call it like they created these like fake camps but but like people people in in the inside always knew that these are like fake camps M yeah I I I think so too and I think you know we we cause we we call this the uh Us Versus Them Dynamic uh and and that tends to get a lot of clicks because you end up with lovers and haters and things that are very contentious but in reality when you get down to it uh we want to solve all the problems in the best way that that's possible and that is going to

take multiple uh different types of Technologies to to solve these complex problems um and so I think kind of moving moving into specifically at Phoenix like what what you guys are specializing in I'd love to hear a little bit more about how you guys use FHA what is kind of your what is your timeline or roadmap or Vision uh for generating more adoption of FH technology so I mean what we focus on on FH is uh specifically confidential smart contracts um and specifically right now like H evm base so like we allow you to write solidity as is H we give you like library in solidity that b and will give you like encrypted types and then you

know you can basically magically compute over like encrypted data so I know let's say you have like an auction contract right H and I'm guessing you know like sealed bid like second price auctions they are way way more efficient than public auctions because if like in a public auction I can see what you be and I can kind of front run you and kind of try to kind of like snug a better deal at the end of the day but if everyone's bid bids are private that and you do like something called a second price auction then it's like mathematically proven that you get like the best results for everyone um so that's like an example of a smart contract that will only work let's say on ethereum on L2 on arbitrum on us on optimism but it will only work if people can encrypt their

bids and that's what we focus on basically allowing developers and users to decide what data inside of their smart contract executions and business logic they want to keep private and they can selectively share it or or not and there are many many many use cases uh uh for that um I think you had another question that I um essentially like H how are you how are you using that to generate more adoption right and confidential smart contracts like how are you going and and and preaching this to the world and and and you know those who want confidential smart contracts coming to the technology and you giving it to them right so we do it two ways we

do the usual approach of like you know getting developers hackaton we do a lot of like brainstorming sessions and like internal hackaton and kind of like build like eat our own dog food and like Building Products that we think are are interesting and we don't necessarily do that to a you know build like we're not a Dev shop but we do that to kind of like show people like what you can do with FH and with confidential smart contracts that you cannot do otherwise so that's strategy number one strategy number number two um and I think that's that's a that's a bit more complicated but if I but maybe I'll pick one example so like we think there's and we have good data to show this that like there's a lot of interest around like

confidential ERC 20s I'm not talking about like tornado C mixing I'm literally talking about like you know I send if I send usdc I send you know ARB or wbtc or CB BTC and all that and I don't want people to see my balance and you know and and how much I have maybe they can see my identity you know whatever but like at least at least that there's like a lot of Market especially with stable coins because people use stable coins a lot for payments like we've had like companies that are in crypto saying look we pay our employees in stable coin we don't want our employees to see each other's like salary doesn't make sense so there's a lot of that so we think that's like a core building block that if we build and then we bring other teams that

kind of build like other money Legos that def building blocks but that tie into this like confidential defi confidential 20 aspect then we're going to see a lot of growth and enable a lot of like use cases that we've never seen possible before right and we need that confidentiality primitive uh and and FHA is the technology that allows us to do the computation uh on top of private transactions and so right like I you know we have a company that pays people confidentially on chain um then we can do things like maybe you know there's some sort of taxes or payroll expenses or or things right and so maybe as simple an arithmetic as you know 10% Needs to Go to paying taxes right I the 10% you don't need to have any

information about who or what of of that underlying transaction we can take out taxes we can put some in a 401k and an IRA we can split up the different different token amounts and and break these things up into whatever and and uh alloc these these things across the different places where they need to go without having the the confidentiality broken right and I I think this is this is a particular use case that you know for blockchain applications to look a lot more like they do in real life in web 2 and and traditional companies we need to add these privacy preserving elements on chain imagine if the world's GDP moved on chain guys say GM to safet the newest protocol by safe isn't a

blockchain and it's not a new layer to it's a transaction processing Network built to send money anywhere and swap anything on thousands of layer twos from a single Universal balance in sub 500 milliseconds it's fast it's smooth and it's simple and it's built with safe multi-sig level security in mind safet unlocks smart accounts across multiple chains allowing users to supercharge security scalability and speed so get started today at net. safe. Global and remember to follow at safe on X welcome aboard on our journey into the singularity this is the endgame road map for fra and fraxel the new L2 out of the

frax ecosystem that is totally based on fractal scaling fractal a modular L2 blockchain built by fra with modular and fractal scaling in mind fractal scaling is based on the mandal BR set of fractal images and it explains how l2s become l3s and the entire app chain life cycle goes through its Evolution the frax ecosystem is made of some of the top builders in the space and has stemmed from its OG days of the frax stable coin it is now in the LST space the LRT space and this frax L2 joins the existing

Suite of fra products including fra swap and frax lend you can see frax fairy the bridge to go from any chain arbitrum optimism or main net into the fraxel main net Bridge your tokens and bring your FRA assets into the fraxel ecosystem join today to get the best out of fraxel you are still early join today take part in our Expedition into the singularity 100% agrees but you would also be Sur surpris like big financial institutions they see that so we had I mean this was H revealed a few days ago so I can now talk about it we had JP Morgan come to us and say like we're

looking into real world assets right like we want to tokenize like private Equity Funds we basically want to tokenize everything we believe in ethereum we want to do it like on ethereum and and and you see that M but they said like look our clients they're not going to buy shares in a in like a real world asset fund and that if people can see like how much ownership they have and how much what like they literally told us like that they need this now this is not like you know for people to avoid taxes this is literally like one like a Fortune 100 company saying we need this in order to bring this product to Market one of the funny things I've seen around this conference is there are more events side events with bank co-hosting and sponsoring I've

than I've ever been I've ever seen at any crypto conference before and I was actually uh in the elevator with with uh K plus and uh it's like a the actual Thai Regional Bank here funded by the royal family of Thailand and I'm I'm in the elevator and I asked like you know why are you co-hosting this blockchain event are you using blockchain in your bank and she said no we're not really using blockchain in our bank but we have this uh to asset tokenization arm of the bank and I mean quick sidebar it's like H H how much of a 180° flip from JP Morgan Jamie Diamond from Black Rock Larry think Bitcoin is for money laundering D to we want to tokenize the world it's

insane where did that come from yeah you know uh I guess I guess they needed to pack their bags at the bottom before before they started chilling I don't I mean well maybe maybe maybe I'm not a conspiracy theorist so I'm not sure I I I buy into that but I think people change their opinions and I think you know when you're a public figure you're a CEO of a of of like these public companies you know you you go up on shows and you you have strong opinions weekly held like you say very very strong things and then you you you get more information and you learn that it was not Incorrect and uh you elegantly move to the other side yeah but I think that's good for us I I think so too man it's been fascinating to watch I think

early on it was uh very easy to just hate to just hate on blockchain hate on bitcoin because it was the status quo and now the incentives switch now the incentive is banks are starting to buy buy crypto and tokenize assets El Salvador governments are starting to buy Bitcoin and tokenize and and now the incentives are such that it's it's uh in their best interest to talk about Bitcoin in a positive light because the train is leaving the station and they they're they want to be on board so I think that's actually the critical point so I remember like 2015 I talked to like Banks and regular all that and there was this kind of like permission private blockchain let's do that and and the

talk then was like let's reduce cost and it wasn't clear how much cost you reduce if you reduce cost but the reducing cost is never as interesting as like new Revenue like streams right and now they get new revenue streams right like like you have custody Banks or like might like they see fire blocks and coin they like I want to be in the custody game because that's additional revenue and JP Morgan and does it say like okay I can get revenue from investors and take fees by by basically tokenizing everything so I think that's the other that's the other yeah I think you're right I think they see that they're leaving money on the table because crypto has now grown to an economy that is at scale to a certain extent it's going to continue to grow but it's big enough where they can

they can't help but take notice and and they see that their services are prime to enter this market and now this Market is big enough for them to to pay it attention right one thing we haven't talked about is is arbitrum we haven't talked and we're here at arbitrum ARB verse right so I let's talk take kind of the the last few minutes to to talk about arbitrum uh I'm mostly curious why did you choose to build on the arbitrum stack so just so that you understand how an unnatural decision it was initially and and and that is to arbit room's credit of of like the amazing stack that they built like when we started we built like this uh basically a cosmos based Network and we have a lot of experience in Cosmos like validators all that it

was way easier it was faster it was working and we dumped all of it to basically build on the arbitrum stack why do we build on the arbit stock because building something like an FH chain roll up whatever you want to call it as a lir one it does not make sense because FH is still expensive you need probably specialized Hardware like you're going to need sequencers you're going to maybe have a lot of validators over time but sequencers those that are actually ordering the blocks and actually executing defy computations like you really want to work in in like in in a rollup model where you can kind of move as fast as like you can without waiting for like other validators to sign off on consensus and all that and you know then once you understand that

you understand that FH rollup is the only thing that makes sense and then the question is okay arbitrum optimism both are great and I say arbit one in in two ways I mean we love all both teams and stacks and all that but like what arbit gave us that no one else gave us is a way to do fraud proofs over F so FHA is not solidity under the hood it's like rust code very optimized you know it's mathematical operations how do you do fraud proof on that on chain and arbitrum you know with the with the offering like they they they allow to basically compile anything to web assembly and do that and they actually helped us like they not only gave us the solution like they worked with us closely to kind of figuring out the nuts

and bolts and how to get it to work and from security perspective that's literally was the only way to move forward um and then B is they're amazing people like they're extremely smart they care about FH they like FH they see the vision as we see it like working with them together side by side and I think you see us working a lot more with them together with especially with the offin ls team because like there's such a great alignment like many of them are researchers as well so they they really align with the vision yeah absolutely uh to arbit trums credit they've also been able to amass and and they've been able to amass this um this community and hub of liquidity

and users yeah that really is unparalleled in the rest of the space and and so we're actually we're going to have plume in here in a little bit and they're doing some great work with rwas also GMX and and a and all these defi powerhouses are on arbitrum and it it is the home of defi I don't think it's it it's wild to say that and so when we're thinking about FH um there this needs to come from somewhere right and so I'm curious about the forward compatibility of FHA and Phoenix on top of these existing defi uh applications and so is it possible to run the confidentiality on top of things

like a and GMX and like like rather than rewrite smart contracts from scratch using confidentiality can we put confidential on top to an extent so I first of all in ourl to we're going to support like both modes and we're going to we're building like really crisp us to make it easier for users because we do want like also like we want to connect like to the entire arbit ecosystem and other ecosystem and to tap into like other liquidity sources and and so forth and also it makes sense for like projects to come and just like Fork what they've done so far and maybe later on we can work with them and they can add like other features with confidentiality but it can start from

the from the edges it doesn't have to be like rewrite everything or copy everything so that's like angle number one angle number two is like and that is something that we are working very closely with the with arbit and this is completely based on the arbit stock like quite frankly I don't think that Phoenix end game is our L2 I do think that the L2 is like extremely important both for us and the arbit from ecosystem today but the end game is like everyone's going to want to build their own rollup or chain everyone's going to build their their apps like whereever ever they are asking people to rewrite everything that they do and come to us and changes like it doesn't make sense so we're building this as a co- processor you can think about it like a lightweight rollup that

basically Taps into the same security of the chain whether it's the L1 whether it's arbit directly whether it's like orbit chains it will tap into the same security level and model of those chain and will settle to them right but inside it would run a sequencer and that and basically run like a mini FHA arbit room stack and I think that's like the end game that's and that's how like developers like you know they will have their apps and then they can just tap into the co-processor and say hey maybe I do this next feature by tapping into the Cod processor already have it on arbitrum or eum L1 I'll just tap into it for one thing and then maybe over time I'll grow and and and expand my product and that's really how we think this gets adopted yeah and ultimately this is

going to be helpful for the web 3 rollups and applications that are already live they're going to be able to tap into this co- process but I think it's even more important for the web 2 projects that have not come on chain yet they're looking for the most lightweight modular solution to en to enable confidentiality privacy into their solution and and and the co-processor I think is going to be a phenomenal way for them to get involved and so if you were to say you know one one word uh to or you know one last piece here to those who are watching who you know we have a lot of application Builders a lot of uh uh people who are working at rollups and working at applications and and you know if you were to kind of let them know why they should uh come use the the Phoenix

co- processor what you know this is your chance to do so thanks um I mean I think FHA and and it's this is getting to be accepted not at this point I think it's like the next FHA and I think 5 years from now like a like blockchains in general they're going to change you're going to have three steps encrypt on the user side en crypted transactions ordering and computation is done over encrypted data I mean flashboards are doing it with te right now on the ordering side like Phoenix and other we do it with FH or other Technologies on the smart contract Level side and then eventually when you need to reveal something that new decrypt and like if you're a developer in the space then I

think it just makes sense that if you buy into this Vision that this is like the ne DEC Computing Paradigm of blockchains five years from now it makes sense for you to kind of like be of the curve like it's kind of like being early to Z and then being able to kind of like uh set the stage and be the first one to build a cool building blocks and basically take like most of the pie awesome guy thank you so much man this has been an absolute pleasure learned a lot as well thank you for having me I mean this was awesome