Paul Brody - On-Chain Privacy Is Here - TOKEN2049 Singapore 2025

Hi everybody. So I'm here to talk about privacy. And it is it is really good for somebody like me who's been talking about privacy for like 10 years. I'm a bit of a broken record, but it's finally getting the attention it deserves. We have seen so much discussion about privacy this year.

Uh it's been great to see Vitalik talking about privacy. It's been great to see the Ethereum Foundation making privacy a priority, but a lot of what people think about and expect for onchain privacy just isn't correct or isn't based on kind of the state-of-the-art. And in particular, we're seeing a bunch of almost privacy solutions that are not quite there and are not quite good enough. Right now, there's kind of three major ones that we're seeing in the market. Number one, mixers and privacy pools.

These look like tornado cache or some flavor of tornado cash. Number two, trusted systems. I think trusted systems are in many ways the antithesis of what we came into blockchain for. We didn't come here to depend upon a centralized uh centrally controlled centrally managed system. Uh and so they present risk.

And then we have a whole bunch of flavors of what I would call limited privacy systems. These are things that conceal some of the data but not all of it. And the idea is that this will somehow be particularly acceptable to regulators but will still be good enough to protect your privacy. None of these are going to be good enough. Where we need to get to is something very clear and very straightforward but very difficult to execute.

Fully mathematically encrypted composable systems. Right? And these are things with full endto-end encryption. They run entirely on chain and they will allow you to build complex programmable ecosystems that look just like the mainet, you know, the Ethereum mainet over time, but everything running under privacy between the parties involved in a transaction. Why do we need to do this?

Why has things why have things moved along? I think there's really kind of two fundamental answers. Number one, the use cases have expanded. So a lot of the early use cases around privacy are like hey I want to pay you some money right and so fungeible tokens work great for that but fungeible tokens are fine for mixers to a limited extent they don't work that well but you can't use nonf fungeible tokens in any form of a mixer and you can't interact with defy ecosystems or smart contracts using these kinds of mixer systems so the use cases have moved on and then secondly the best-in-class systems that can reverse engineer your transactions have gotten really good. Um, I saw this slide uh a couple of weeks ago at a conference in Munich.

This is from Chain Analysis. I'm using it with their permission, but you need to understand what's going on here. Chain Analysis is following a pile of money from a stolen NFT through a crosschain bridge and a mixer, right? And now it's not perfect, right? I think that if you look at it very carefully, and we won't have time to do that today, basically about $600,000 goes in and about $50 or $60,000 comes out at the other end.

But the bottom line is this. They can track your money. They can follow your assets even when you're running them through cross trainers, right? You don't have privacy that you think you do. I think there are four fundamental requirements that are needed for us to really mature them.

Privacy game number one scalability right we need to use privacy transactions for everything that means we need as much scalability in the private section of the blockchain ecosystem as we have in the mainet number two programmability I need to be able to build defy services lending services amms I need to be able to build all of these technologies and tools under privacy and I need to be able to decide exactly how much privacy I'm going to give you in each one of these Third, I need composability. Every time you move an asset into a privacy system and then out of it, you are leaking tons of metadata. You are exposing important information about what you're doing and where you're doing it. So, you need to be able to have composability under privacy, which really then brings me to the last point, which is you need permanence. Stuff has to go into the privacy layer and quite frankly, with rare exceptions really needs to stay there, right?

stuff goes into that black box, it has to stay there. So that is what we are trying to build at EY and in particular over the last decade and believe it or not it's a decade. This year EY is celebrating our 10th year in the blockchain business, right? I'm very very proud of that. I helped start the blockchain business at EY in in 2015 or technically what we call FY fiscal year 2016.

Um but we've got two key things that we're building. Number one, it's called Starlite. Starlite allows you to take a Solidity smart contract and turn it into a zero knowledge circuit. You do not have to be a mathematician. You do not have to be a zero knowledge expert, right?

It generates the API so you can do the testing. We have been building this since 2021 and more than 80 companies and banks have been using it. Now, Starlight allows basically almost unlimited programmability and flexibility in a privacy ecosystem. Nightfall is our scalable mass volume transaction service. It allows you to do high volume standardized things like payments, swaps, uh, and other standardized things at very low transaction costs.

Today, even with compute costs, it's between 5 and 10 cents. We know we can do up to 400 million transactions a day in our current capability. And our road map takes us to in excess of about 4 billion transactions a day. Right? So we have standardized scalability and we've got um uh almost limitless flexibility.

And by the way, both of these technologies are public domain opensource, right? We are making those as a contribution into the ecosystem. That's why it's called Nightfall and Starlight, not EY Nightfall or EY Starite. Now, where do we want to get to? Our goal is basically full composibility and interoperability between these systems.

So, you can have a pile of assets in a privacy layer like Nightfall, right, which is a shared privacy space. And then if you want to do something specific to them from a from a uh engineering standpoint, from an AMM, from some kind of like DeFi service, you can take them and put them into a customuilt Starlite contract and execute those at scale. do other types of zero knowledge activity and then put them back into Nightfall. All the entire process fully composable under privacy. So basically the goal honestly is relatively straightforward.

I want to be able to have any kind of businessto business transaction fully executable under privacy and you'll be able to put assets into this environment. You'll be able to move them around. you'll be able to do business with your other parties. And then if you really want to, you can also take them out. But just keep in mind, every time you put an asset in and every time you take it out, you're kind of making it highly visible to your competition.

Now, none of this is intended to come outside of a regulatory framework. One of the things that's really important to understand when you're talking about enterprisegrade privacy and the road map is there's no anonymity here, right? This is not about evading responsibility with regulators. This is not about refusing to comply with the law. This is about keeping your competition, whether they're another company or somebody who wants to see what your portfolio looks like.

This is about preventing them from just snooping in on all of your business transaction activities. So we have a quite substantial long-term roadmap for these technologies and we are starting to see what I think is very exciting and meaningful adoption. So the first network that deployed Nightfall uh just a few weeks ago is the Plume network. You may know them as a real world asset network. They've been getting a lot of traction.

Their test net is up and running, right? And you can actually find the data in our open source repository. And today I have even more kind of amazing news which is I get to share with you that our second Nightfall network has launched and it's Cello and yes exactly I am I'm very very excited about this because whereas Plume is a real world asset network is a global payments blockchain network that is getting a tremendous amount of traction all over the world. People are starting to use Cello for their personal transactions. But one of the most interesting things about networks like Cello is they are really in many cases B2B networks.

They are small business networks but they are B2B networks and our goal with Cello is to bring uh businessto business payments on chain. So really really pleased to see this thing happening. Um it is it is very exciting. It's big for scale. Uh, and I could not be more thrilled with Cello as a uh an implement here.

When we set out thinking about what would it be like to build out sort of public domain assets, this is the kind of adoption that we were hoping for. Believe it or not, as I mentioned, we've been working on this for a while. 2026 will be uh 2026 will be our 10th year uh of working in blockchain. We started the core engineering back in 2017 when we'd been around for a year and we started to realize, wow, privacy is incredibly important. Um, I showed the prototype of Nightfall at DevCon in 2018 and our transaction cost was a $100 in gas.

And by the way, this is when ETH was going for $250, right? You want to talk about computationally intensive, right? We have moved ourselves to the point where we've gone from a $100 a transaction to about 5 cents. We are currently on Nightfall version 4, which is a ZK ZK rollup, which means you have instant finality and full privacy. Uh we've also been developing Starlights in 2019 and uh in 2026 we will have uh incremental versions of Nightfall, incremental versions of Starite and full interoperability and composability between them.

Now if you're thinking gosh this sounds like a very Ethereum talk, it is. I will just straight up admit to you EY is heavily heavily Ethereum aligned. This is kind of our big north star. We have made a strategic bet around Ethereum that we believe that this is probably going to be the winning public blockchain. And we have always been very clear public blockchains are going to be where all of this growth is.

And in fact, it's because all of this growth is going to be on public blockchains that you really fundamentally need robust privacy technology. I am pleased to say I wrote a book, it's called Ethereum for Business. The US Food and Drug Administration says it is safe and effective if you're having trouble sleeping at night. It's not that great. My mom says it's not too bad.

Um, uh, I'm also the chairman of the Enterprise Ethereum Alliance. I had been working with Vitalik since 2014 on blockchain technology. So for me, this has been a long long trip. Before I wrap up here, I would like to all of offer all of you a proof of Paul NFT minted on EY blockchain infrastructure. Uh so if you'd like one of these, now is your chance to get it.

I can't match Cello who are actually giving away Tether $10 Tether uh uh tokens this morning. But I can give you an NFT. Eyeal would like me to remind you that this has no cash value and you definitely should not be selling this on OpenC. like that would be that would be very bad. Um I would get in trouble for that even though it's a special uh token 2049 NFT.

So with that, I'm going to say again how incredibly excited I am that the folks from Cello are on board. Like this is a huge milestone in terms of privacy adoption and uh have a terrific token 2049. This is the era of mainstream privacy. Thank you.