# FHE eats ZK for breakfast - Lauri Peltonen | ETHDam III - 2025 - Channel: [CryptoCanal](https://streameth.org/cryptocanal) - Date: 2025-10-07 - Duration: 11:02 - Watch: https://streameth.org/watch/yt-eXEcaPqb34w - YouTube: https://www.youtube.com/watch?v=eXEcaPqb34w ## Description Welcome to the 3rd Edition of ETHDam, hosted May 9–11, 2025 in Amsterdam. This year, we brought together the brightest minds in privacy, security, and AI for a unique 48-hour hackathon + conference combo. 🌷 https://www.ethdam.com// 🌷 ------------------ FHE eats ZK for breakfast - Lauri Peltonen | ETHDam III - 2025 🎀 About the Speaker: Lauri has done EVM development for around 7 years and entered the ZK world about 2 years ago. Since then, he has been developing various privacy-focused ZK projects and engaging with different ZK ecosystems. He has given multiple privacy-related talks at Ethereum and Zero-Knowledge conferences across Europe. 𝕏 Follow: https://x.com/LauriPelto ------------------ About ETHDam & CryptoCanal ETHDam is powered by CryptoCanal, an education and events platform rooted in Amsterdam, expanding into Rotterdam and ZΓΌrich. Keep up with us to see updates on future events: https://www.cryptocanal.org/ Follow CryptoCanal on X: https://twitter.com/CryptoCanal Join CryptoCanal TG Community: https://t.me/CryptoCanalCommunity Join CryptoCanal Discord: https://discord.com/invite/XJVjpCqQBz CryptoCanal unites crypto enthusiasts committed to making a positive impact. Unapologetically political, we prioritize education, events, and services while championing cypherpunk values like privacy, sovereignty, and censorship resistance. ------------------ πŸŽ₯ Credits: Intro / outro by babyPRO - https://babypro.art/ ETHDam Photography by Paulus – https://concretestate.eu/ MC of ETHDam - Laura Brown - Cofounder at JobStash & Veri, and your resident crypto ginger. https://linktr.ee/laurabxyz ------------------ Special thanks to our partners who made ETHDam possible: 🌹 Hackathon – Bouquet: Oasis Network https://oasisprotocol.org/ 🌷 Hackathon – Petal: Circles https://aboutcircles.com πŸ’› Conference – Gold: Zano https://zano.org/ Dash https://www.dash.org/ Bitvavo https://bitvavo.com/en 🩢 Conference – Silver: Igra Labs https://igralabs.com/hero πŸ’› Conference – Copper: Lido https://lido.fi/ DeTrip https://detrip.travel/ Cake Wallet https://cakewallet.com/ The Grid https://thegrid.id/ Calimero Network https://calimero.network/ 0xbow https://0xbow.io/ Mina https://minaprotocol.com/ JobStash https://jobstash.xyz/ Cyber Capital https://www.cyber.capital/ POAP https://poap.xyz/ Acronym Foundation (Supported our Top 10 Hackers) https://acronymfoundation.org/ 🌱 Sponsor: EF Ecosystem Support Program https://esp.ethereum.foundation ------------------ 0:00 | Introduction & Background 0:09 | FHE and ZK Overview 0:32 | What Is Zero-Knowledge? 1:54 | Prover & Verifier Explained 3:40 | Use Cases of ZK 4:48 | ZK Flow & Trust Models 6:14 | Learning Curve of ZK 7:01 | Introducing FHE with Example 8:14 | Comparing ZK and FHE 9:14 | Illustrated Conclusion 10:06 | Q&A: Tools & Adoption ## Transcript Welcome to Ethan to Ethan to EA to EA to EA. So today we will talk about FHE bit a little bit and uh ZK and how they play together. So I'm I'm uh I'm a freelancer doing ZK EVM develop development. been doing it for a for some seven years now and um maybe like two years ago I focused more on the ZK side. So what will we actually talk about? So there's these two difficult concepts that I actually mentioned. There's a Ziggler knowledge called ZK and the second one is FHE which is full fully homorphic encryption. It's a mouthful but hopefully we will understand a little bit how it works and what it enables. So we will quickly go through what these two are uh what kind of use cases they en enable how they work a little bit. Uh then there are some challenges because they are rather complicated topics. So there are some challenges in how to use them, how to learn how to how they work and how they can be utilized and some comparisons. So well let's start from the ZK part and let's actually start uh what does it mean to be zero knowledge. So the actual term is used quite um in different context and it means a bit different things which causes some confusion but in in a strict sense it means that we have we have some privacy. So we have some data and if we transmit it the the receiving party gets zero knowledge of the original data. So there's zero data is revealed of just some properties that we want to reveal but not the actual data. So in ZK there's this two two entities prover and a verifier. So we have a prover and they want to prove they have some claim for example I have a number or I have some certain algorithm that you know I know I want to prove that I have these things and then there's the other party the verifier okay I have a claim I prove it and then someone wants to verify my claim or is it is your proof valid do you actually have what you claim you have and then these two interact in certain certain patterns So there's two way ways they can interact. They can either do it interactively. So they kind of talk with each other and the verifier asks like more details or more challenges or then then there's just like non non-interactive. So the prover kind of publishes the proof somewhere and whoever wants they can verify it. So the flow flow starts with claim like you have your claim maybe I have a secret number I don't want to reveal for some reason and you input this into into a special approver program you can write this yourself or maybe someone else has written it for you and this outputs a proof. So this proof is just kind of like random string which you can then input into a verifier program. So once once you have this proof you can verify with this special verifier program whether the proof is valid or not and the verifier and the prover they are tied to each other quite tightly usually so they they have the context like the verifier knows that what I am verifying so you cannot verify other proofs usually it kind it depends a little bit also so this enables two main use cases uh verifiable computation. So it means you have some this goes back to the algorithm that you have some claim about algorithm or you have some computation that you want to prove that you did it correctly. For example, you can ask your cloud provider to do some computation for you and then they just they can prove that they did did what you asked them to do. But this can also be used for scaling. So if you have some computation well blockchains are about computation. So you can scale your blockchain especially like L2 rollups by using ZK just to prove that you have done the block production and all these rules. But the second use case is the privacy. So you have some some data that fits some criteria and you define what are the criteria but you don't want to reveal the data for whatever reason. So these for like whatever financial transactions you have maybe like uh just blockchain transactions or just any stuff you do online in the internet. So the bigger picture is that you have these two entities like the proverify and they are somehow connected. They maybe they talk to each other or they are just like non-interactive but still you have this and then you have someone who wants to who has a claim and they want to use approver. Okay, now I have a claim. I will go and use this prover and out comes a proof. Okay, and someone wants to verify this proof otherwise like you generate proof but nobody cares about it. So what's the point? So there has to be someone who cares about this proof. So they get the proof in some way from some online database or whatever. So they get this proof but what do they actually do with it? Well, they ask the verifier program like okay now I give you the proof. is the proof valid or not? And this is kind of like a bit complicated already. But then then there's all sort of like nuances on top of it like uh some techn technological stuff like trusted setups and like different trust assumptions for example like you have this verifier but how can you trust what it does? So you get this program from somewhere but how do you know it verifies what you think it verifies? So this lead leads to like like under underneath there's all all sort of different mathematics like we're we're not going to go here but all sort of different weird So it's it's super complicated and just don't go there like smarter people do that part luckily and we don't have to touch that. So the typical learning path is like first you try to understand this what is what does the term mean and what what is zero knowledge and then then you're like okay now I may maybe understand something what are the use cases for it where can it be used so was it like privacy or was it was it something else and or like how how do they discuss well like did they talk each other within with each other or not like you get you're starting to get a bit confused like how does the system actually work and can I trust the verifier? How can we establish that this kind of trust connection? And then somebody says, "Ah, screw that shit." Like this is getting too complicated and it's difficult to architect this kind of system also. Like is it is it worth it? But so let's let's take take a few steps back. Uh let's imagine this like awesome service. Let's a service that doubles any number you send it. So we we sent number four. It gives us eight. Awesome. We have a great service. But then the next question is can we add some privacy into this kind of service? Uh yes and no. We can encrypt our value. So we encrypt number four and send it to this awesome service. And what do we get back? Well, we don't get anything back. Like the service cannot do anything with the data because it's encrypted. We have full privacy but we get nothing back. So it's totally useless. And this is something that FHE will fix. And where it actually shines very well, it's the same system. We encrypt number four. We send it to this service that supports FHE. And they sent encrypted eight back. And the service itself, they had no idea what they they doubled something, but they have no idea what they doubled. So the only only one who knows the number four and number eight is the original user who encrypted it. So this is the beauty of FHE and if you kind of compare like these schemes and designs how they work it's a bit obvious like which one is easier to understand and to architect. So for the learning path is a bit different I would say point one there's sub you have some encrypted some data you encrypt it and other other people can compute on based on the data and that's about it there's nothing else basically you have to learn about it so they both offer privacy but uh how ZK does it is just totally different way they do it like there's the prover and verif fire interaction and it goes it's a lot more complicated and with FH it's kind of just works and of course like it's not a silver bullet it has its own own issues there's a lot more processing and there's lots of lots more research needed for FG but in theory it works a lot easier so ZK works and it has been established and a lot of project use it it's great fine but it's complicated to get into use. So somewhat works, but it's a lot more easier to understand and to explain and to take into use. So if if FH can get their research done and they can get their together at some beautiful day, then the it will probably eat many of the use cases especially the privacy use cases that ZK provides today. So illustrated conclusion, we will just eat all the ZK privacy. Some beautiful day. Let's see. Yeah, that's my slides. Thanks. [Applause] Thank you. I love the advice of getting your together. So I really like that. And we've got time for a few questions. Um permit over here. Cool. What tools are there out there right now that that people can use to implement this onchain or or or off? You mean Yeah, that's uh there's not too much onchain stuff. Zama is doing some and then there are few other projects but I think those are in theory they work but I think it the computational effort is way too much still. So you cannot in real life you cannot use it yet. Cool. Do we have any more questions? Yes. No. If anyone if you think of any questions later on in the day, um Lai will be walking around. Um so, thank you so much. Thanks.