Les Amis d’Ethereum France #13 — Comment travailler dans la DeFi

[Music] We're going to start this Bobbit UP on decentralized finance. Thank you for coming in such large numbers to participate. I'm Patrick, president of the Paris Blockchain Society. I'd like to say a few words about the association if this is your first event. We're an association of young Web 3 professionals, so it's generally aimed at people with between 1 and 50 years of professional experience in the sector.

The goal is to get together to deepen our knowledge through expert conferences like tonight's and to develop our network. We have two main types of events: conferences like tonight's, which are free and open to the public. We organize two types: highly professional conferences, like tonight's, which focus on how to work in a specific sector, and then you'll have small booths where you can interact directly with the participants; and thematic conferences, for example, recently on asset aggregation, where we bring together different experts to discuss a topic in a... More in a debate format, and then for association members, I invite you, if you're interested, to join us at the end. There will be a short presentation with all the information if you'd like to join.

We also organize weekly webinars with experts, reserved for members. For example, we've had the opportunity to meet the Global Adobe Trading Manager from Galaxy Digital, and soon, the CEO of Banane France. This is also our anniversary event for the association's members, and it's exceptionally open to the public. You can participate on the Blue Chain Society LinkedIn page. Just leave a comment with your question, and we'll select the best ones for this exclusive exchange.

Before we begin, I 'll give the floor to Nathan S'Exer, Vice President of Ethereum France, since we've partnered with Ethereum France for this event. He'll tell you a little about the association. Finally, I'll give the floor to [name missing], who are kind enough to welcome us once again to these magnificent premises. Thank you all, and have a very good evening. Just one more thing...

A word about the association: we're a non-profit organization (under the 1901 French law). We've been around since 2015. We organize regular Meetups like this one. We also organize the big conference called "ity," which takes place in July. I'll talk about it a bit later.

We write content for our website and blog, and in France, we translate technical articles. The Ethereum White Paper, in particular, has been translated. We have all our social media accounts where you can find us: our Minecraft page, our Discord server. We're currently relaunching memberships, so that will be happening in the coming weeks. We also have a YouTube channel where we're currently recording the Meetup.

It's transcribed live, and you can find all six recordings on our channel. We did a Meetup on CFOs, specifically on Web 3D, last month. We speed-updated with PVS, and we're doing a developer-focused Meetup on June 1st with the Kairoslawn and Angle teams, which you'll hear tonight. Then there's our big conference I mentioned in July, called LIGHT (here and there, yes). So, a quick word about that: we'll be reselling tickets on June 1st.

It seems pretty soon in general, but there's a big wave of access, so I encourage you to note that date. We've just launched the Itissywick website. It's generally a week or ten days of events surrounding the conference, and there are 150 events, sometimes 150. In any case, last year there were 150 events taking place around here. If you're organizing an event with your company, you can submit your event, and we'll accept it.

There are already about twenty events listed. So, that's all for me, and I'll hand it over directly to Comète. Thank you very much. [Applause] Hello everyone, I'm David Dufour from Comète, and we're delighted to welcome you to this very, very nice event organized by the Lazy Blockchain Society with Ethereum France. So, I was just briefly talking about Comète.

At Comète, some of you may already know, we started developing games on the blockchain, and then little by little we began supporting companies interested in developing their Web 3 applications. So today we support companies like Française des Jeux, Lacoste, and Live Beyond. And today is a rather special event for us, as we announced the official launch of Alambic and our Web 3 development platform for developers to help them develop their Web 3 applications and games quickly and easily. Our platform covers the entire inventory and technology stack necessary for this, from simplified longboarding facilitated by Act abstraction, to the infrastructure side as well, with features we're working on that allow developers to launch their applications extremely easily and at extremely low cost, which is essentially the goal. It's a whole middleware stack that covers everything from Marketplace APIs and rental systems to NFTs and more.

Indexer and relay API, so we 're happy to announce this today. It's a good time for us, and it puts us on a path of extremely strong progress for the company. So I'll leave you in peace, I'll hand it over, maybe to Patrick. Thank you very much. [Music] We can now start without further ado with Nicolas.

So, high up, Bubble Map Crypto, entrepreneurship in the storm. Hello everyone, my name is Nick, I'm the co-founder of Bubble Maps, a tool for analyzing and visualizing unchained data. So today I've come to share my experience as an entrepreneur, since 2018 I've created two crypto startups: one that didn't work, called Cryptonavia, and the other, which is doing very well, called Bubble Maps. From these two experiences, I've learned several lessons that I've come to present to you today, starting with Bubble Maps, just briefly. So, Bubble Maps is a new vision of one-chain data, a vision that we wanted to be very interactive, fun, almost addictive, with colorful visuals and Intuitive, and our goal is simple: to bring as many people as possible back to One Chain data.

This is a goal shared by many blockchains, as evidenced by our official partnerships with Polygon, Arbitraum, Avalanche, and Phantom, which have obviously greatly encouraged us within their ecosystems. I also wanted to mention something important: Bubble Maps has become, over time, a trusted intermediary on data/ unchain topics. We publish some excellent investigations on our networks that generate a lot of discussion, particularly this one. We managed to show that, in a single story, an investment fund—let's say Henry Moritz and Sixtinzie—had significant control over governance. Through multiple wallets, they can unilaterally make decisions.

So, what you see on the screen, to put it simply, is the bubble map of a story. All the holders are represented by bubbles, each bubble is a wallet, and when the bubbles are connected, it means there have been transfers between wallets. The wallets know each other, and you can see very clearly that the wallets are not independent at all. In reality, there are political stakes, power struggles in this governance, and therefore entities trying to control it. We even talked about it on Twitter; I remember it caused quite a stir.

Binet Easy's SIO even started to express legitimate doubts about this governance. Okay, now let's go back to the beginning. I'm here to tell you my story, so we'll start at the beginning in 2017. That's when I really started to get interested in crypto. At the time, I was short-stacked and had a few engineering collars.

I'd been working in consulting for a year, and 2017 was a bull run. Everyone around me was talking about crypto, Bitcoin, Ethereum, etc. And I remember, in my free time, I started researching and I stumbled upon this YouTube channel. (I'm putting a picture of a cut-off hand that no one had stuck in the room.) Ken the Crypto was a mini YouTube channel with...

He had 10,000 subscribers, and his goal was to find the next crypto unicorn, the one that would become One and DeX. He kept saying it, and I remember looking at myself in awe because every time he talked about a crypto, the price would explode in the following weeks. I'd think, "Who is this oracle? Who is this fortune teller I've found?" But actually, no, we were in a bull run, and everyone was right, but I didn't realize that.

So, I invested with him, of course. In 2018, when all the prices collapsed, my investment dropped to zero. He shut down his YouTube channel, and we never saw each other again. But I took away one thing from that experience: his investment methodology, which I found interesting. Before investing and looking at several options, he would look for cryptos with a low market cap, therefore higher growth potential.

Then he would look at the team behind the project to see if they were experienced people who had already done things. He would look at the GitHub code. And the GitHub activity to see if there were any new features or functionalities from an active technical team, and then looking at social media: was the team regularly sharing updates, and how was the community reacting to those updates? So, in 2018, we were in the crypto market, the beginning of the 2020 market, and I thought, " Okay, I want to do something in crypto. This market interests me.

I'm going to transform this methodology into a product." And with my two partners, we launched the famous Nadia crypto dashboard. So, for investors, you can add these metrics that I just mentioned: GitHub activity, news feeds, social media sentiment, etc. We worked on it for two years, and it was a complete failure, a real flop, zero users, tragic. And I'm going to try to share the lessons I learned from this and why it didn't work.

There are four main reasons. First, the product's market fit— specifically, is there a market? Are there people interested? are willing to pay to use your products, and it's very important to do this research beforehand because if you go down the wrong path later, at least you haven't wasted time, energy, resources, etc. And crypto, life...

we could have realized very quickly that it was too simple for the experts and far too difficult for the novices, and therefore not intuitive enough. So we didn't know how to find our place in the market and a good way to make a product that fits the market. Obviously, ask around you, but honestly, especially for B2C products, if you were a regular user of your own tool— we weren't, sometimes I was just too lazy— then a clear and concise value proposition... today I don't see it. There are a lot of pitch decks where there are lots of ideas at once, lots of product ideas, lots of features, lots of business models.

It's often interesting to refocus because it allows you to concentrate on the essentials, and at the same time, users know where you're taking them. So it really has several advantages. Virtues and the same goes for crypto, life. We had plenty of building blocks that had some value, but not one that was essential. Third point: quickly release a Vitaa MVP beta as soon as possible with crypto lavias.

We spent a year developing before releasing anything, and there was no one to receive it once we released. It was a mistake, and what's more, I realized afterward that we added cosmetic elements that weren't necessary at all. For example, we had a tab system so people could organize their dashboards. We made the modules dynamic in size so you could adjust the size, but nobody cares about that. The goal really has to be to create value, and then you can afford to add cosmetic elements, but not before.

And finally, build a community. That is to say, today we are in an attention economy where having attention is almost more important than having a good product. That's why, by the way, all the influencers Those who launch products that are a hit, whether it's drinks, clothing brands, etc., it's not because the products are better, but because they manage to convert attention into business. And indeed, that was something we did wrong with Crypto Nadia, and I really advise you, if you want to start, to jump right into the attention arena.

Don't wait, don't tell yourself, " I'll build the product and then I'll see about Twitter." No, do Twitter and Telegram right away and co-create with your first users. Well, it's now 2020, we're coming out of this experience and wondering what to do now. For me, it was hesitating about whether to go back to a normal job or not. There was family putting a bit of pressure on me, and it turns out that something pretty crazy happened: the 2021 Bull Run resumed.

So, after three years of cardiac arrest, the first signs of life from the pool run patent and... That time was also the peak of the diffusion, so blockchain and Layer 1 were exploding, especially the banana smart chain which was just starting. There was one site that managed to stand out in the Marchienne basket, which is why some of you might use this glorious site, Bouconne, and why it climbed into the top 100 most used sites in the world during the 2021 trading run. You get it, it was absolutely incredible. And why there was a promotional section where they highlighted tokens?

It was simply advertising, and we realized that as soon as a token was highlighted, its price skyrocketed. So we created the simplest product you can imagine: a card alert system. We send alerts, and it's a huge success. And there, the four criteria I told you about before are well validated: product-market fit, everyone should make money, it's simple, plus you'll reinvest 5 times each time for a month. It was truly incredible.

Second point: the value proposition. You buy our alerts. Third point: get out quickly. It took us a week to develop this, whereas Crypto Nadia took us a year. We're clearly not on the same scale.

And last point: build a community immediately. A Telegram channel, a Twitter account, and beforehand, we already had 5,000 people following us. Obviously, the problem with these alerts is that we didn't see a really long-term future for them. For us, it was more of a game. So we had prepared ourselves, we had already anticipated, we were thinking about how we could concretely move forward with this.

And we found ourselves in a rather interesting period where we had a community but no product, whereas with Crypton Avia we had a product but no community, which is actually a better position. So we thought about it, we took a deep breath, we remembered that in Crypton Lavia, perhaps the most valuable component was already a bubble map that we called the cluster map, which showed how Cryptocurrencies were correlated, they took off, and so, without hesitation, we adapted it, recycled it, and very quickly, two weeks later, we launched V0, the MVP of Babel Maps, which was far from what it is today. With a blockchain, it couldn't click on bubbles and there wasn't much else, but we quickly released it to gain market exposure. Our community was very enthusiastic, and we started generating revenue, also because the tokens were curious and worried about seeing their own, probably. So that was quite practical.

And then, a year and a half later, Bubble Map has changed a lot. Today, it's a tool that works very well and has made some great discoveries. We played the attention game to the hilt, so today we have 50,000 followers on Twitter, analyses that often go viral, we've partnered with blockchains, integrated with Open6, Cogneco, etc. And it's only now that we've almost exhausted this initial value proposition that we can transform Google Maps into a dashboard and diversify. So now there are some very big...

News coming to Bubble Maps, some in this room know about it, others don't, but it will be announced soon, more or less. And to conclude, a word to wrap up: entrepreneurship is difficult, entrepreneurship in crypto was even more so, by the way, the title " Entrepreneurship" is a bit lyrical, "Entrepreneurship in the Storm," that's really it. You've seen that my journey hasn't been linear at all, there's been a lot of trial and error, a lot of hesitation, but always with that energy and dedication, we wanted it, we wanted it to work. Today I don't regret it at all, it's the best experience I've had in my life and I really enjoy what I do every day, so I really wish you the same. Thank you very much.

So, hello everyone, great. So, my name is Paul Frambeau, I'm here at the Morphologist department, and what we do is we develop Morphologist, it's a protocol, a protocol that allows for a decentralized financial service. So we're focused on one segment, which is the landing. So basically, crypto allows you to exchange money with each other. It allows you to lend and borrow without a bank, in a way.

So our job is to create an algorithm with no more than 1000 lines of code, which we'll deploy on Ethereum for people to use. It could be individuals, it could be other protocols, etc. Today, there are almost a billion dollars deposited on Morpho that are borrowed by other users who want to finance themselves, do financial actions like leverage. Etc., etc.

, we'll come back to that a bit, and today what I wanted to do in 5 minutes is give the keys for those who might be interested in starting a protocol. It's quite exotic in terms of entrepreneurship, it doesn't really connect to other segments of Web 3.0, so we don't care about that here. The goal is to code, so it's really very, very precise, very focused on something. And so, strangely enough, it's going to be quite different from the presentation just before, even if there are some common points.

And I see three main steps. The first is what we call a White Paper. So the goal, once again, is to come up with a protocol, like the protocols we had on the Internet, like HTTP. So it's a very precise sequence of instructions that allows you to perform a very simple service: transferring a web page or SMTP, transferring emails, etc. And it's something with which you can't afford to make a mistake.

So if we make a mistake on one line, well, indeed, the billion dollars can end up... So, we'll have examples later on the other security panels. It all starts with a vision and work around a protocol. The second step is fundraising, which is quite the opposite of the other approach, much more Silicon Valley-style, where the focus is on product-market fit, user needs, and so on. Here, we can't afford to iterate quickly because every mistake can be very costly.

Therefore, development cycles are much longer. It's much more important to unite around a vision; you need to find people who believe in that vision to secure funding because even the smallest change to our code can cost hundreds of thousands of euros. So, it's crucial to raise funds at this stage. Finally, deployment is simply all the engineering involved in coding the smart protocol and deploying it. These are non-trivial tasks.

You need to try to assemble the most talented team possible on the subject and, again, be very focused and concentrated on it. Okay, so to deploy the protocol, and then if the primitive It makes sense, so we can market it, get it adopted, etc. So in practice, you do need to iterate with users before deploying, but since development cycles are much longer, we tend to plan what will be deployed on the pipeline several months in advance. That's about it for my part, and Merlin will talk a bit more about Anti-Ify security practices because, as you'll understand, it can have a significant impact. Hi everyone, so I'm Merlin, co-founder of Morpho with Paul.

So I'm going to explain how not to die when you launch a vulnerability project. It's quite a challenge, actually. So we're going to review how you personally can defend yourself and how to protect your protocol. So, roughly last year, there were 3.1 billion hacks—3.

8 billion, I think, actually—so 82% of those were simply Anti- Ify protocols that were hacked, so that's a considerable sum. And in Skam attacks, we have about 6 billion dollars that were... What are you personally fighting against, anyway? So there are scams everywhere, whether it's on Telegram, Twitter, or by email. Here you have a friendly meta- mask asking you to freeze your site—don't do it!

Your favorite project, NFC, suddenly makes an announcement, but it's actually just one of the founders of Board Yacht Club who got hacked, and I don't remember how many millions were lost. That was two weeks ago. I detailed a great airdrop from Lance, and here's an airdrop from PP and emails from Camille. So all this to tell you to be careful, keep a sharp eye out every time you receive a notification that seems a little strange. Really pay attention; if in doubt, don't do anything.

It's an environment where you're much more likely to be the target of attacks, especially from... So here it's Paul, not really Paul, who sends a message to Tom from Morpho. Luckily, Tom didn't lose any money, but... as a founder Perhaps in a future IFy protocol, you need to keep in mind that potentially your employees, or you yourself, will be impersonated, and someone will try to hack you. So, to defend against that, what can we do?

Funerals... I have links at the end so you can look them up. So, what can you do? Use a password manager. I'm going to say some basic things, but if you don't have one, it can still be detrimental to you because when you have a protocol worth, for example, 500 million dollars or a billion, and it's managed via a multi-signal, the slightest vulnerability can cost you dearly in reputation, in money, or in money for all your users.

So, a password manager with more than 12 characters, using DEFA, so any offensive factor, no SMS, otherwise if someone steals your phones, kisses, don't use too much cologne, especially on your mobile, it's much easier to crack, and don't wear large... We're also good at something, which is not putting your siraille on a note, for example. It's nice of you to remind us that a pretty good setup is a multi-SIG of 2 or 2 or 3, for example, with two under Wallets or one hardware and one hot. So you can learn a Ledger, despite the controversy, it's perhaps one of the least bad and safest for multi-cig, which is quite dry when you go on a protocol. Pay close attention to the links; there are a lot of scams.

Generally, if you type in, for example, a story, the first links that come up won't really be a story; they'll be ads. So check links, a stad app. You say, "Be careful, check that the link is good." As soon as there are signatures, check that what you're signing is correct as much as possible. Try not to do it blindly, and so if in doubt, there's no doubt, don't do anything.

Really, take your time; you're not in a hurry. And if not, well, pray. So once have that, that's you for your Personal setup, but then on the protocol side, for example, Morpho, we have almost a billion dollars invested, so the setup designers, etc., need to be top-notch. Also, the protocol is open source, so basically everyone, even the best hackers in the world, can see the code and exploit it.

So, that's a non-exhaustive list of things to do. First, design the protocol in the least complex way possible; it needs to be as simple as possible. Then you can do attack modeling, which means trying to analyze all the attack vectors. For example, we'll focus on AV. We'll need to study attack vectors to understand how this protocol works and what the implications are for your protocol.

So, on the one hand, but then it could be all the mechanisms of your protocol and what the consequences might be. Now, once it's designed, you have to build it. And there, you can't just do a little something—I don't know if there are any developers, but when you're in school If you don't do too much testing, it breaks, no big deal, you just restart it and it works. You have no choice, you have to do tests, and you have to do as many tests as possible and try to break your code. So, unit testing, fuzing, formal verification, you have to go for it, as Paul said earlier.

Just one broken line in your code is enough, and it's over, and your reputation is tarnished. So, once it's built, do as many internal reviews as possible. If you have a development team, you have to allocate time. Okay, let's take a few days, and we'll stop stealing anything. We'll spend time just checking that the code we've copied is perfect.

Even better, you should have an external team, or at least a team that hasn't implemented the code, to look at it. They'll have a fresh perspective on the code and they'll tear it apart. You can also call on freelancers; it's quite useful, and you can follow checklists. Then you... You can move on to external audits, so that's where it starts to get expensive, costing hundreds of thousands of dollars if you want good auditors.

Then you have a fixed, optional step—it depends on your protocol—but it's good to have a monitoring system. Monitoring means that you check that everything is going well, and if there's a problem at the start, normally at the launch of your protocol (at least in Landing Space, there aren't many protocols that are easily imitated), it's good to have a bot to warn you if something strange is happening and to be able to place contracts in a timely manner. Then, IRP is like Cydia; it's basically what you need to do when things go wrong. It's a basic checklist, but when everything falls apart, you shouldn't spend three hours figuring out how to place the contracts, who contacted which authorities, who the people are who. So, you connect.

That could be... The National Gendarmerie, if you're in France, has quite a few contacts. You have to deploy it once everything is ready, and if you're not sure, if you have any doubt, well, go back to step five and start the loop again. You do fixes, an audit, and bug fixes too. It's a good way to ensure that hackers, instead of accessing your contracts, actually use them.

You pay them once, once they've discovered something, and the last step is you pray. So I call this the mental model because, in fact, each layer of protection helps you a little more, and the more layers you add, the more secure your protocol is. So the values ​​needed to launch all this, because there are a lot of processes, it's quite intense. There are some pretty intense situations, like a PEC hack, things like that, pretty much every week or two. So you have to be very rigorous, especially for code reviews.

To steal perfect code, you need a lot of resilience because it's long, very, very long. There might be some code that has a vulnerability when you're just about to launch it, so you'll have to redo the audit or redo the stats you saw earlier, and a lot of patience is required because development times are quite long. It's rather difficult to test the solution in a real production environment; we're not talking short timeframes, especially at Bolloré. An audit can take me six months, so you risk the audit being rejected, and in fact, you have zero lines of code. So you have six months to do pretty much everything.

Here are some cool links. Actually, I had put a QR code, but it's no longer there, so take a picture. It's after... oh yeah, super smart. So you can...

I'll put the slide back up, but basically, you have blogs, actually, for good funerals for you personally, and then sequences of Nelson, who has a jacket background, and on the transmission security, which is a very well-known definition. And there you have a few others. I'll put them here. Thank you very much, cool, thank you Merlin. He gave a short presentation on what to do to prevent things from going wrong.

I'm going to try to explain what happens when things go wrong and give you other contexts and ideas for how to react appropriately when things go wrong, and also how to limit the occurrence of these incidents, focusing on two aspects: the security of smart contracts. So, let me introduce myself. I 'm Pablo, a co-founder of Angle. We're a decentralized stablecoin protocol, a bit like Maker, which makes DICE, except we make a euro stablecoin called Giro. Giro is currently the most traded euro stablecoin, accounting for 60% of euro transaction volume.

So, if you've already traded with a euro stablecoin, but I'm changing it on disk, it was with Giro. It's a protocol deployed on many chains and is collateralized, so we don't use Bondy or Luna. That's a whole other aspect of how to design it. Protocols with a meaningful economic mechanism would be worth discussing, so I'll go back two months. Angle was a protocol that had 60 million in VAT when we launched.

We thought, well, we wanted to improve the way stablecoins existed by creating new ones. It was pretty cool, though. So we launched a euro stablecoin, Goro, and we also wanted to improve how decentralized stablecoin protocols were built. So we launched a new way of making decentralized stablecoins that try to improve the stablecoin trilemma: a more decentralized stablecoin, but one that remains collateralized, and the stablecoin itself remains stable. Anyway, to be able to do this, the protocol invested part of its reserves in other protocols.

So we had 60 million under management, 40 of which were invested in Campan. And so, on March 13, 2023, at 10:00 AM, Angle had $17.6 million on Uller, which Bacquet agents... So, it was protocol assets that backed up labiability, which was at 0 at 10:02 AM. Angles' 17.

6 million USDC had disappeared because a core was stolen. It was the Euler protocol in which Angles had money, so we had lent money to someone who was being robbed. Well, we were lucky, we didn't have as many early warning systems as Morpho, but we were still quite quick. So, 3 minutes after the hack, no, we realized what was happening. We understood that something was wrong.

We had an internal checklist to follow when things went wrong. Also, the first thing to do when there's a hack is to lay everything out, meaning I don't know what the damage is, I don't know what might continue to happen. I have to skirt the system so that no one can profit from it, so that I don't get emptied even further. In fact, we at Angles had managed to accumulate a certain cushion. So, for security reasons, we had a surplus of 7 million.

That is to say, we had, on all the people who had a Clem on the collateral of the protocol, if they all loved it, in the end we had 7 million left. So we were robbed of 17 points, 6 million. We have a 7 million surplus, but we still owe 10 million to other people. Now, if everyone is aware, well, it creates a leak, a bangkron, to use the English word. That is to say, it's a bit like first come, first served.

And the fact of filing this allows us to break this dynamic and ensure that, well, everyone is treated equally in light of what happened in the protocol. So we filed everything, we put the protocol into liquidation, and then our responsibility was communication about what's happening because we didn't want there to be, said an insider, people who see, oh well, the Géro is no longer in the bucket there. I have euros, I owe some pegs, we didn't want people to be able to bet on the Giro and its intrinsic value is 70 cents, while others say, well no, their intrinsic value is 50 cents, given that there are reserves, so after laying everything down, after putting the protocol into liquidation and removing all the liquidity that the protocol controls here and there, there are other elements that I wo n't go into detail about, we did what we call a protocol whale so that even my grandmother would be able to understand what's missing and how unfortunate it was for us, what happened, well, that's for the first two days, it's something you shouldn't hesitate for a single second, that is to say, the moment it happens to you, you shouldn't have to ask yourself questions, how do I place it, what transaction do I have to make to place it, which contract do I have to call, everything has to be prepared in advance so that it's executed within a minute, we have The chance that a transaction had already passed through the protocol by the time we reacted and set up the entire system was there, and we were also lucky—we were so fast that Angle was deployed before Air was deployed. They didn't have the mechanisms in place to deploy the whole system; it wasn't planned. So they were able to continue being exploited after the hack began because people continued to steal the remaining liquidity on Air.

After that, once you've communicated, once everyone is aware, there's not much left to do. The question arises of liquidating the protocol, or is the protocol bought out? Do we get bailed out to reinvest and restart the machine? We were lucky. Merlin said earlier that failed hacks accounted for three billion dollars in 2022, so hacks of 200 million dollars like the ones we're having now happen, but hacks of 200 million dollars...

the hacker Repayment happens, it never happens, and we're incredibly lucky, a bit like miracles, because the hacker repaid all the stolen funds. Angle, who had millions stolen, recovered 19 points and 2 million because the hacker bought ethers, which actually increased in value. So we came out of the hack in a better financial position. We would have liked to avoid this because, well, we have a decentralized token that lost value, we lost a lot of assets. From a business perspective, it's not good at all; we took a big hit to our reputation, but we're coming out of it in a better financial position.

So now there's something that's failing us. In a question we often ask ourselves, there are many protocols that are hacked. The conditional probability of a protocol being hacked, knowing that it has already been hacked, is it considered higher than the conditional probability that a protocol to get attacked knowing that he's never been attacked before, okay, our theories are that we don't know how to do it, okay, but yet we were tormented by this experience and now we have gadlines on how Reducing our exposure is limited, so it depends a bit on the protocol, actually, on how they react. Will they get hacked more if they've already been hacked, or do we consider it even more likely since they were bad once? Are they bad a second time?

Well, we can assume that's the case. So how do we prevent this kind of situation from happening? Angle, for us, is a protocol built on the fact that Diffa is composable. That is to say, in Diffa, there are several types of protocols. There are protocols that operate in a closed loop, that won't interfere with other protocols, and protocols like Angle that think, "Well, this is great!"

There are protocols that build yield mechanisms, there are protocols that build mechanisms for exchanging tokens. We'll try to connect to them all to try to collectively contribute value that is greater than what each of us contributes individually. Protocols, like a story... For a protocol, like a story, it works in Vasco. Your Nice hen isn't going to talk to other hens.

She isn't going to talk to other hens. Campan, balance, if you're on balance, it's not exactly the case because they go into liquids. You look at the kouti when you put an ether on the cookies to borrow lusd. It's a function, it has to work on its own, it never talks to anyone else. So when you launch a protocol, you have to be aware of which category you fall into.

For us, it's been clear from the beginning that what interests us in the industry is the composability of the failure. So for us, it's a value to be able to compose with other protocols, to be able to interact with other protocols, and that comes with risks. That's what found us involved in the hacker. So once we have that in mind, what is the way to avoid interacting with a protocol that makes it a pariah or that risks becoming one? Paying up for the first really simple thing, well, it's pretty obvious, it's about only interacting with blueships, protocols that are well-established, whose team you know, which was the case for now, which has existed for quite some time and has already managed billions of assets, suggestions, or hundreds of millions.

We were pushed for a long time to do things on Cahier Soir, for example, which has a protocol, a protocol, a bit like a story in which we added liquidity, after the hack we withdrew all our liquidity, and after the hack we said we would never again put liquidity into protocols that aren't the most reputable. It turns out that a week ago, she realized we had found a critical vulnerability where people who had added liquidity, or all the money we had put in, could have been taken. So even protocols with a good reputation, protocols you trust, only interact with people you are sure of. After you have identified the The SCOP of people you want your protocol to interact with should n't be lumped together. What we were criticized for was that 40% of the protocol's VAT was in ULR at the time of the hack.

I still consider that to have been poor risk management on our part. Not everyone agrees with that, but Angles' money was n't all up in the air. And if there had been 60 million in an hour, well, then the people who had invested in the Angles protocol, instead of getting their money back if it had been hacked, wouldn't have been reimbursed 60 cents for every euro invested. They would have gotten back 20 cents, so it would have been much worse. So you have to try to diversify the protocols you invest in.

The other thing is that even when you diversify, even when you only interact with cheap blue networks, you have to assume that all protocols will crash sooner or later. And that's a very good security principle. For example, Angle is a stable protocol. A stable protocol is only valuable if it's available everywhere, if people can use it easily. The simplest way to make something stable is to make it available on many different chains.

And making it available on many different chains means interacting with bridges. Bridges are historically the most common way to prevent failures. If you allow mining your stable, allowing bridges to mit your stable, well, that means you're exposed to the risk of all possible bridges. So, Han's bridge mechanisms, which were built before the hacker, are based on the idea that we support multiple bridges, thus diversifying our risk. And because bridges will break, we put hourly limits and global limits on what can be mined on a chain.

So, what can be mined on one chain and another in Algeria, so that breaks the US a bit. It's a bit of a compromise, actually. We gain in security, but we completely lose ancillary features. But at least we're sure that if a bridge breaks—as we have today, it's mainly the R0 bridges that make Stargate, which is the bridge used for Giro— we know it will break one day, but at least on the day it happens, we can sleep soundly because we know that our relative exposure to a bridge hack is rather limited. So, that's how you choose how to interact with different protocols.

There are other things, I think, that are also valuable for trying to limit your exposure to hacking. I have two other ideas, but only one comes to mind for the moment. The other idea is that the protocol exposure must be very transparent to everyone. That is to say, you'll have your internal metrics that you look at on H8, exposed in time to the risk of ether, exposed in time to the risk of touch wrap on my landing protocol, if tomorrow the Lido Finally, the viewer of the Idopet, well, my risk extends, but in fact, that's what you're looking at. But collective intelligence is always stronger than what your team is capable of seeing, and that's something that's super, super important and quite neglected: having very high-quality analytics on the protocols so that the risks are transparent, not only to you, but also so that you can judge your risk in a second, and in an automated way.

So, for example, we have scripts running that give us our exposure to different protocol assets, telling us, before adding collateral, for example, what deviation we can tolerate before a protocol is considered risky. We have internal nomenclature, but we make sure that everyone can really see it. And because we've already had feedback from people in the community saying, "Look, I don't know if you've noticed this, maybe we should increase these risk parameters because, like some protocols, there are different designs, but it's not The purest design of a protocol—we design economic systems that we want to scale and that aren't as easily imitated as protocols, like a story—means there are thousands of parameters we can play with, and we can't see each of those thousands of parameters at any given moment. But before that, I have analytics or transparent data made available to everyone. We encourage and allow the community to contribute, with some pretty interesting feedback.

So, what I was adding to what Merlin said about the checklist to have when you launch your protocol in production is that you really have to make sure that all the protocol data is understandable and transparent to everyone without needing a PhD in protocols. And that's something you shouldn't delegate to the teams because you yourself, when you build your protocol, are best placed to know what the economic risks are to which your protocol is exposed, or the composability risks via other brands, contracts, and What are the important metrics to measure? So there are many companies that offer to delegate certain components. It's important to raise these from these companies, but ultimately, you must retain control over how you present your risk to your users. So, that's how we now approach our few key points to prevent the hack from happening again.

I think we've been somewhat inoculated; we've been lucky. We're the living dead, and especially since we received the money, we're practically dead. We saw ourselves dying, and so now we believe more than ever in the idea of ​​a traditional, transparent finance system. Therefore, we must embody these values ​​of transparency and transparency at the protocol level, even with the risks involved. And everything is a question of risk in the end.

Hello everyone. So, I 'm Anastasia, I work at Kaiko. Kaiko is responsible for providing protocols to all companies, whether they are in the crypto sector (i.e., protocol providers), in the traditional finance sector, or even regulators.

Data on everything happening in crypto markets, whether centralized or decentralized, and personally, I handle all the Defy data. So I joined Keiko four years ago to make sense of all the data Keiko had at its disposal. At the time, it was all Dataify. And for the last two years, especially since Defight Summer, we decided with ICO to launch our Defy data masks and complement what we already had in this very promising sector. Our goal is to bring transparency to what's happening in the crypto world, who its users are, and to best support companies that want to enter this ecosystem.

So today, I wanted to share with you the importance of Defy data, not only for the IFY ecosystem (the Dataify protocols), but also how other players can use this data. We'll start with a brief overview of what Defy data is, because it seems obvious, but there's actually much more to say than you might think. Diffa has two main categories of data: market failure data and user failure data. Market failure data includes things like how much volume is traded on a deck, for example, how much has been borrowed on Avion or Campin, and how much liquidity has been achieved. This relates to markets in general, specifically failure data.

User data includes things like the sustainability of a user's pre-borrowing position on Avion, or the concentration of liquidity on a DEX, in terms of the exchange provider. So, on one hand, we look at the markets, and on the other hand, we look at the users. I'm going to show you how we use both types of data by combining data from Keiko and Aleno, another data company whose founder is here today. As a reminder, all this data comes directly from the blockchain. Diffa involves applications built directly on the blockchain that allow any financial transaction to be performed directly on the blockchain.

Therefore, Keiko and other data providers that do this connect... directly on the blockchain, and we give meaning to some of the transactions that happen on the blockchain. We'll move on to the interesting part, which is all the use cases—or rather, some of the use cases—of blockchain data in general. The first and most obvious is how, thanks to blockchain data, we can find the best investment opportunities. One way to do this is to look, for example, at the borrowing cost, to compare the borrowing cost of a cryptocurrency that interests us across different platforms.

As long as you have access to a database that gives you access to this borrowing rate on several platforms, it 's very easy to create a graph and see where it's least expensive for you to borrow, and therefore potentially optimize your strategy. Another fairly basic database you can look at is prices. Prices, well, the price of BTC, the price of any cryptocurrency, isn't the same on all platforms; it differs a lot between centralized and decentralized platforms. But they differ greatly between the various decentralized exchange platforms, so there's a difference between a petabyte story, for example, and it can be huge opportunities for companies that focus on arbitrage. The graph on the right shows how to compare and have total visibility on the price of all cryptocurrencies in one place.

This allows you to directly identify arbitrage opportunities and potentially profit from them. The last thing I wanted to say regarding identifying the best investment opportunities in savings accounts is that the Defy data shows you the concentration level of a liquidity pool on the DAX. The liquidity in a liquidity pool is held by liquidity providers. The problem is that while there are many liquidity pools, there are fewer liquidity providers, and generally, the liquidity is held by a small number of liquidity providers. This means that if you are a liquidity provider within a protocol, you are using part of your liquidity to hold it.

To make it profitable by being a behind-the- scenes provider, somewhere along the line, if that's your job, you need this data to make wise decisions about where you put your money. So that was the first use case. There are many, but the others will go a little faster. Another use case for a very different type of data is the one that allows you to leverage data on Alify protocol users, and in particular their profiles. This is useful not only if you work in a protocol to better guide your feature implementation strategy, for example, but also if you are a regulator or a financial institution and you need to understand who is using Alify in order to better propose regulations, potentially so that financial institutions can enter this ecosystem and bring more liquidity and adoption.

Regarding the current state of the default, the data I used here is data from the Morpho protocol, which was presented earlier, and it's user data. So we took the users who use Morpho and Aleno. The focus is on labeling these users, and we've seen that most users are people who use leverage. This allows the Morpho protocol, and it could allow any other protocol, to better orient its strategy. Another use case for thread data is better preparation for failure-specific attacks.

Defy brings many innovations and is also heavily inspired by traditional finance, but it has also inherited some of its downsides, such as attacks. In particular, it has inherited all the attacks from frequency trading, for example. But that's not the attack I'm going to talk about today. I wanted to talk about a very specific failure-specific attack that doesn't exist in traditional finance. It's a legitimate attack, but it must be identified because without identifying it, if you have significant liquidity, you gain absolutely nothing by providing liquidity to a protocol.

This is what we call just-in-time liquidity. Very simply, it consists of providing a large amount of liquidity just before a trade to Capturing all the Fizz transactions of this trade and immediately removing this liquidity involves enormous amounts. Just by looking at the liquidity inflows and outflows on a V3 story, which seems to be the only protocol to date that offers this opportunity for some "goodies providers," we can clearly see that it's possible to identify what proportion of these operations are from Justin Liquidity and what proportion are organic liquidity events within the protocol of a V3 story. [ Music] Another rather striking use case that I like to present during conferences because it's something that has affected all of us in crypto, I think, is the one that allows us to anticipate crises by using Fail data. The biggest crisis we've experienced recently, I think, was the crash of Terra Luna and Akaiko.

By cross-referencing CIFA and Dify data— because ultimately, that's our job— we were able to identify the crash of Terra Luna, or I don't remember exactly, just by looking at the volumes. Trades in the default and civil markets, basically what happened was that what was happening in the default market, nobody knew what was happening. Basically, there were a lot of STs being put up for sale, except that in CIFA, the reactions happen one to two days later, so without a combination of CIFA and DIFY data, nobody could have seen it. Actually, we saw it, but we don't know our job is to predict crises, so at least this will potentially allow us to have this data and potentially anticipate future crises, hoping there won't be any. We'll stick to market data, not just defaults but in general, but talk about a slightly different subject, a slightly different use case, which is to anticipate, or rather, to better guide Tokyomi's strategy, and by anticipating the potential price drop linked to unlocking some investor liquidity when you participate in ICOs Generally, you have a quantity of tokens that are unlocked over time, and we've seen by looking at market data that once a portion of these tokens are unlocked, and you can sell them on the secondary market, the price drops.

However, the larger the total supply of tokens, the greater the price. You have allocated to investors the more the price drop will be following the first unlock that you make we will talk again about ify users but this time to identify and understand the psychology of the markets linked to a certain token when we look at the blockchain we have a lot of data including ify data but especially data on who owns what and who does what and what the kit owns we can see it by A7 and here I chose the governance token of Ave and I looked using Alleno data what the proportion of holders of this David token remains after one year and after two years of holding this token and it is quite surprising I expected much lower percentages but we see that the proportion of users who remain and keep the token after one year is 61.5% and the proportion of users who keep the token after two years is 33%. This is a fairly high retention rate, however, it's quite debatable because when you look at address activity and the proportion of truly active addresses, which therefore hold the "a" (presumably referring to a specific value or value), 50% of them were considered inactive at the time of the study. This would be interesting in my opinion—well, I don't have that information, but it would be extremely interesting, in my opinion, to do the same thing with the users of the IF (Identity, Identity, and Security) protocols.

This would allow us to know the true adoption rate of an IF protocol, and in that case, it would be to give it a pure and simple Defy (Defy) rating. Here, it's the same kind of data that's used, but it's precisely to evaluate the level of decentralization of a protocol's governance. So, generally, when a Defy protocol offers governance, it offers a government token. Those who hold this token can vote for future decisions and many other things related to the protocol. However, many stories in crypto, many bad stories, are linked to the fact that there's usually only one person, the Finder.

which holds more than 50% of the governance tokens and has almost total power over the decisions made regarding its protocol, and today, due to the availability of the blockchain data and the Defy data, because ultimately it's enough to read the blockchain and add a lot of intelligence to deduce things from this availability, it's possible to see for all the governance tokens what the share of each user is in the whole blockchain. When we take the Yuni token, for example, which is the governance token of a story, we see that almost 50% of this governance token is held by the community, and there is another part held by a contract that belongs to a story that is quite large, but it's a contract that handles governance, and there is another fairly large part, shown in yellow, which corresponds to all the tokens that have not yet been tested. Finally, to end on a positive note, the divine data is especially important for everyone. The world, on protocols, regulators, and financial institutions, to show the potential that, in the end, Kaiko was the first company to show that a story actually records the same trading volume as Coinbase, and this for almost a year and a half. We were also the first to show that, in fact, almost all the liquidity, meaning the number of tokens available for trading, was in decks rather than in centralized exchanges.

And this data, what's positive is that we sell it to financial institutions, to banks, to hedge funds, which certainly act and use the loophole today, but who could use the loophole tomorrow. And when they see this, they realize that there are not only many opportunities because there is a lot of use, but that they themselves, if there were a framework to easily enter this ecosystem, could do so, and that would be beneficial because that's where the best opportunities and the best market conditions are. Thank you very much, good evening everyone. So, my name is Hugo, I've been president here for a year and a half. It 's called White Loop Capital, so we invest, among other things, in centralized finance protocols, and something that's very important to understand in order to explain behaviors in general, including unfying ones, is game theory.

I'll try to give you a brief introduction [Music] In a few words, what is game theory? It's the study of decision-making and behaviors. It's important to understand what it is and what it's used for, because it's what will allow you to drive certain behaviors, insofar as you, the users, your contributors, will always choose the solution that satisfies them the most, and that's really at the heart of it: the incentives, whether positive or negative. Game theory, as I was saying, is present in your everyday lives. I'll take an example: apart from default, maybe you play sports.

That's good if you do it because you want to be in good physical shape, you want to be good mentally. There's a... there's A quote behind it says " fault," well, maybe you're going to bring liquidity because you hope to make an economic gain, so in that case, these are positive incentives. You can also have positive incentives and penalties if, for example, you decide to quit smoking because the price of cigarettes is increasing. It's because, on the one hand, you want to avoid the price increase, and on the other hand, the penalty.

So when you have a price increase, and also to increase your purchasing power, we're back to the fault. Maybe you're staking land, you're doing it because you hope to make an economic gain, and you're not malicious towards Node.js if you have one because you know you'll be penalized through slashing, for example. In short, game theory is really what will incentivize, and you must use it to promote the results you expect from you, your users. It's really important to exploit them to achieve objectives and promote engagement.

To better understand this, we need... Defining several types of games that apply to centralized finance, there are more than four. I suggest we focus on C4; in our case, these are the most common. The first is the zero-sum game, in the sense that one person's gain equals another's loss. I'm excluding platform fees, but for example, if you go on GMX, for instance, and you're a trader and you make a loss, those who provided liquidity will make a gain.

Conversely, you have non-zero-sum games where everyone wins, and the sum of the gains is greater than what was initially invested. Typically, if I take the case of liquidity, it benefits everyone on the network. It allows for more tokens that are stored, making the network more secure and increasing liquidity in the market because the staking tokens can be reinvested in the market. You have the games Sequential games, which are pretty much true for all games, are a game where actions happen one after the other. It's a game that, yes, as I was saying, is almost entirely different.

If you see someone else take a big action before you, you might change your behavior or make a decision based on that action. For example, if there's a hack on a protocol, for instance. And finally, you have cooperative and non-cooperative games where you cooperate for the benefit of a group, or conversely, you might be, in a sense, malicious towards the group and prioritize your own personal interests. This is a fairly common case within the DAO (Digital Asset Organization). I'll give you two typical examples: people who contribute to the DAO need to be paid, and these people will have an interest in getting the highest possible pay, which is done at the expense of funding other projects.

This translates into... Runway lower, for example, and so another example will typically be the token holders who will push for governance choices that favor short-term interests. A topic that comes up quite often is whether a portion of the tokens should be returned. Recently, with the hard drop of Arbitraum, several high-level participants had the question: " Okay, should we return the tokens we received to us, the holders?" A selfish choice that doesn't favor the DAO in the long term.

So the question we need to ask ourselves is how do we achieve, on the one hand, engagement and cooperation, but also maximize the overall and long-term benefits of the group? I'll illustrate this with the case of liquidity mining in its most classic form. Ideally, we would want it to be a cooperative game in the sense that the interests of the platform will be greater than personal interests because you believe you have users who are convinced of your platform and who won't act opportunistically, and who want to see your platform succeed. This translates to a zero-sum game since it generates an economic gain for the token holder, and it's supposed to appreciate. You have a competitive platform because you have more liquidity, so you, the users, can make swaps with larger sizes, for example.

And as always, it's a sequential game. In reality, we realize it's often a non- cooperative game insofar as your liquidity providers will still see the token. This is called a "farm hent," meaning they will automatically sell the governance token you provide to increase their API. This also translates to a zero-sum game because if one side wins, another side wins. So, generally, we have a token that is subject to seller pressure, not cooperation, because in this case, your users will come as long as you have incentives, and the day there are better incentives elsewhere, they will simply leave the platform and migrate their liquidity to another platform.

This is called mercenary capital, and generally speaking, your DAO is rather losing out. So, to create a gaming environment that encourages cooperation, it's done in two steps. I start with the negative, by penalizing, in general, the behaviors you want to avoid. I think, let's take the case of GMX for example, when you store the token, the longer you stake it, the more points you'll get, which will increase your income. If you stake it, you lose this advantage that you took so long to obtain.

Another example: in the case of A to V, when you want to unstake your tokens, you have a 10-day cooldown to avoid big panic movements. If we take the case of liquidity mining, there's the VVE model, which was developed by [unknown] and is now quite common in the default, which literally prevents the sale of the tokens, which are therefore locked, and there are obviously rewards behind it. And finally, generally speaking, in the default, when protocols grow enough It's also important to think about financing, about financing the good things we have, by the way, I don't know why it's there, it's more of a positive point, so I'll reiterate, it will also be important to finance and reward good behavior, even retroactively. I'm thinking, for example, of the Optimism layer, which understands this quite well, since they have a fund that will finance and reward actions that contribute to the common good, including retroactively, as I was saying. So, rewarding good behavior means, for example, giving incentives for the token to be held for a long time.

I'll take the case of GMX again, as I was saying, the longer you hold it, the more multiples you'll have, and therefore higher income. Aligning interests is also important, both between token holders and liquidity providers, and for that, the Ve and ve33 models, a little more recently, have aligned their interests quite well. And to conclude, I'd like to end with a quote that's not at all made up: the only path that offers some hope for a A better future for the failure is that of cooperation, that's for your attention [Applause]. So, so, if my prison, I'm in charge of the structuring and product side at LG Forge, and I quickly forged it. It's a subsidiary of Société Générale, as its name indicates, and we're not really failures, so please continue listening to me, but in any case, we hope to be able to cooperate, like we did before, to advance the ecosystem and also to find business opportunities that make sense and to have all this innovation that you're working on so that we can also use it.

So that's kind of the objective of LG Forge. We're primarily focused on security tokens; we've been doing this since 2018. We've done four issues and we have two products: bonds and structured products. So we allow Société Générale to offer institutional clients today different formats than traditional finance, and therefore to be able to provide parents Structured projects in token format, and we've also been trying to expand into crypto for the past year and a half. So we have business cases that we're trying to push.

We've also done experiments with Maker, with whom we work quite a bit, and we're also working on quite a few DAOs. I think we've already had some discussions, and so we continue to push these things. We hope, we hope, we've rented resources, and we're convinced that we'll be able to find collaborations, even if the constraints are different and we're not as happy as you. But we can learn from each other. Unfortunately, being an institution, so perhaps how do we define a succession plan for us?

Well, I'm not starting from a crypto-native institutional perspective, meaning that our clients are typically traditional institutional investors, like tier 1/3 banks. We work a lot with insurance companies and also with private banks, so mainly on the crypto side. So we're not Not good with native crypto, so you have to keep that in mind. When I say institutional, I mean more like this: I'm including both the by- side side, which is what I just told you, and the issuance side, with which we interact a lot, mainly this general side. But we've already done issuances for the BU and we've worked with big banks like Goldman Sachs on everything related to syndications.

So, these are pretty big players to move, it's difficult to do projects in that environment. And to give an example, Maker took us two years to push, so you can't be worn out, you can't be tired. And luckily we love crypto, otherwise we would have stopped a long time ago because it requires a lot of effort. And the biggest effort, well, that's the one that, how can I say, is very, very tiring and that greatly hinders the arrival of institutional investors in the crypto world, and therefore the default. So, just getting hold of Bitcoin is complicated, but the default is...

Okay, the most useful challenge, and in fact, it's made up of lots of contradictory injections. For example, AML and all that, so that's pretty classic. GDPR, that's funny because basically AML likes transparency, so it's an argument that works well. On the other hand, GDPR likes privacy, so it works less well. We already saw that we have problems.

It's a bit like the debates we have with our people on the side of complacency. I have pairs, in any case, I think the CNIL (French Data Protection Authority) found a good idea to consider that public blockchain data is private data; you shouldn't put it on the blockchain. So yeah, it's a bit like that kind of thing we have to debate, and it's not necessarily obvious. So sometimes I want to hear your solutions. It's pretty cool.

So, I'm not at all legal compliant, and so I had to navigate all that to try to find arguments and to be able to move projects forward. Another point that I like it, it's there today, how to say it, the incentives need stability, they need to have regulations that are more or less clear to be able to start seeing things move forward, and we have, how to say, in Europe we've been doing pretty well, that is to say, we've had Mika in place for some time now, which is, well, all the regulations around crypto assets, so everything isn't perfect, but overall it allows us to lay the groundwork, and for a social, that's quite important, and everyone is waiting for Mika to move forward, unfortunately, but not long ago, so there's a text at the European institutions level, so we have three European institutions that have agreed to be able to implement a text on prudential regulation at the level of banks, something that comes up regularly, and they said why not add the crypto asset part, it could be good, it will be able to start regulating it, except that the text should appear within a week, so we saw a bit of the version, not a final one, but a bit of a draft on which we will try to bounce back to avoid the worst, but in any case, in the worst-case scenario, a bank or institution—mainly banks, but it seems to include institutions—would be taxed. Well, it's not taxed, but basically, what we call the risk weighting would be at 1250%, which is enormous for a bank. This means that, basically, it cuts off all business for the bank. To put it simply, if tomorrow a bank is involved in shale gas and has one euro worth of Bitcoin, it has to put one euro into equity.

So, basically, that's it. It's penalizing, and it's a pretty strong message and symbol to show how sensitive and politicized this subject is, and that's why we're trying to curb the various types of institutions in this world as much as possible. Obviously, we hope that the worst-case scenario won't happen and that we'll be able to have more granularity, to favor liquid assets, to favor stable assets potentially, but we'll wait for the final version. This is to show how much everything is done to slow things down, and I'm not just talking about the United States, which scares everyone. We, by default, even put money discs on our financial securities to indicate we're not US investors.

So these are small financial companies, so it's still quite complicated. For example, when we worked with AV We have a branch in the United States, which we opened last year to promote devices, security tokens, and more. There's a rule that was issued in the US that tells banks, "You don't have, basically, you haven't used the public blockchain. There's no public blockchain. You're going to look at something we like called Demol and then play around with it."

So, we're more into public blockchain. We discovered it, we started in 2018 with the rise of, well, technologies we know, so we're trying to push the Twin Chain. However, for the US, we have to look at things that are a bit different, so that adds complexity. The other point—this is for the crypto side—but if we go further into the pure vulnerability aspect, I think that for institutions, we're certainly among the only ones today, I think that in European banks, we must be the only ones pushing these vulnerability issues for one simple reason: We don't want to wait for Mika, we want to learn and see what we can do to grow it and do business with you. We're working on the onboarding process today, and we understand your specific needs and accept them.

We also want to, and hope you'll accept, another approach. We don't want to adopt a traditional governance model; that doesn't make much sense. We need to be comfortable with your governance and ensure you implement good decentralization. If you're not decentralized, I'm interested in your Anastasia project on governance decentralization, as these are things we're also looking at. We're willing to accept this governance model based on digital principles.

So, we're reviewing the... We don't know the protocol here, but it's more of a different boarding process. There will be something to consider, but it will be more expensive between us. In any case, these things can work, obviously, if it's not decentralized, but otherwise, and it also allows us to address the whole problem today that arises at the level of European regulators, which is that they want to address the default. There are many groups talking about DAOs and who are starting to look at these issues, and they notice that there is still a lot of centralization in these players, which is normal at the beginning.

But for example, Maker, hoping to give a good example, although it's true that with their range I don't understand everything, they lost me a bit, so we have to review that, we have to look at it, but in any case, it doesn't work, and in any case, there are still Huskas that we can push in the short term. In any case, that's what we are on our side because, as I said, there The classic part, which is everything you call in the ecosystem, is failing, the Real Warda 7, so we're pushing this, we're pushing this, that's Huska, and I think it's a bit trendy lately, there are quite a few fintechs doing it, I'm thinking of Hondo on the back, I remember, as we say, but also all these players who are trying to organize fund shares to be able to provide traditional guidance to the crypto ecosystem, we went into a slightly different world, that is to say, basically we natively liked the security of someone on the blockchain, we have access to the whole range of light structured products, so, pff, and we are able to provide, well, to DAOs on behalf of their protocol but also their treasury, I do n't know, traditional guidance, drivers, some more exotic things if we are able to do, but that's kind of what we want to push quickly in the short term and the longer term because it implies It's all about refinancing, financing, it's, well, it's still, how to say, traditional infrastructure relies on refinancing, and it's something that's very important. However, it requires liquidity on our collateral, which isn't the case. So we hope that with the other regulation called the pilot scheme, we'll be able to provide that, but it will allow us to provide, well, the experimentation, the experimentation we did with Maker, to be able to do it with, well, with other protocols that do the same thing on the screen, for example, do the same thing. These are things that we could, we could do rather from 2024 onwards, but these are things that we're looking at, we're looking at carefully.

For us, it's quite easy in terms of, well, to push these topics because already, these are new business models, new players, even if the bank is, well, hesitant, they like to have new distribution channels, so these are things that are quite, that are quite, well, manageable and that They just continue the skills aspect, and everything we've been working on for some time now is the stablecoin. So today it's more of an institutional mandate, as you know. However, our symbol is striking, so for the issuance, we need to implement the code to attract liquidity to the euro, but also to provide settlement on our securities. Because today we have a base that's in the traditional world, a base that's on the Monte token, so we need this to optimize our operational model, to pay the chain costs, to be able to run a whole chain. So that's something we'd like to have, that's why we need something that works and starts the same way on the payment side.

So we're going to do this with LG, but in any case, there are cross-border concepts between subsidiaries, things we can look at that could have distance in terms of cooperative operational model. So, there were some bugs in the system, as you all know. We're going to try to re-filing it fairly quickly, going through the audits that you're conducting around the world. In my experience, we're also going to try to streamline the Smart Contract. We had quite a few compliance constraints that we needed to approve, plus the harassment methods had to be doubly approved.

So, we're in the final stages of everything. We're only going to do whitelisting initially. Our goal is to proceed step by step. There's been quite a bit of criticism about this; we knew we'd be attacked for centralization, and we accepted that. But we're going step by step.

That means that at the beginning, we'll provide this to traditional crypto exchanges with traditional market making, to traditional crypto brokers, and tomorrow, perhaps we could integrate with the flop, perhaps have some lighter things, things like Circle, for example. So, not through whitelisting, but blacklisting. That's what it's about. What we have is something we have as a long-term ambition, but not right away. And the other point is, the added value of this, this table-con, it's not unique, not that.

Well, we have constraints, but on the other hand, we still have the rain, which means we 're capable. We've structured a stable system that we hope can be a bit more protective of what exists on the market today. That is to say, today when you take your coffin or your breast, you have no recourse on any ground. So basically, you just have a comical, well, sort of gentleman, he comes back with Circle or with your torts to convert your thing into euros. If tomorrow they don't want to, well, they don't do it.

There's no legal obligation. We go through a structure that's still quite classic at the bank level, which allows a token holder to become, at the African level of the site, and damn, look at me, I have tokens, I'm going to get my euro back, and that, and that, that's something. that we were able to put into the balance, so there you go, there are still advantages to being able to hold, I hope rather a site on this table, so hello everyone, I'm going to be very quick because I'm aware that I 'm a human barrier between you and cold beers and other things outside, so I'll be quick, so my name is Ernesto Petit, I'm co-founder and Scipio of Kill Notre, our goal is to make access to staking very widely distributed in the world so that everyone who holds crypto assets can easily store them because we believe that by storing, we can participate in the creation of value from everything that happens on the blockchain, for example, if we make a deposit in an IFI protocol, there is a daughter transaction that is used, that is paid, sorry, and a part of these daughters are distributed to all the people who were there, so it allows us to share all this value, so what is staking? We're talking about the staking protocol This is the mechanism for securing blocks on blockchains, so basically all of them except Bitcoin. I think it's very quick to grasp this because if you're here you probably already know, but basically it's the act of depositing collateral to earn a reward.

By depositing this collateral, you have the right to run a validator. Validators are essentially the servers that validate the blocks proposed by other servers on the chain and that propose new blocks when they are chosen. So it's kind of the foundation of blockchain infrastructure. We facilitate staking through four products. Basically, the basic suite is at the bottom: the validators.

We run infrastructures, so for example, roughly 16,000 Ethereum validators, but we're also present on about twenty other chains. So our clients pay us to keep these servers running smoothly with SLS, good performance, and insurance. Let's talk about it a little. On the left, we have an Unchain product, which is smart contracts that we... We deployed a dashboard to facilitate Ethereum staking, which we launched with Ledger, and now we're launching one with other partners.

This is fairly standard for monitoring positions, etc. It connects to an SDK, which is a code library that allows integrators—for exchanges like the robust Diane at the Wallet—to easily integrate staking across multiple chains with a single library. It also allows them to easily collect chain performance data, specifically the performance of staking positions. So, as I was saying, we're working with about twenty protocols. We started with Pocket Network, which didn't take off too quickly.

After Ethereum, that was kind of the takeoff point for us. We were very fortunate to be on Lido, which is a major protocol. It listens to King quite early on, and then we diversified into many other chains because we often see that our clients did n't want to integrate 25,000 providers. It's important for them to have a one-stop shop with multiple protocols. That's where the Unchain product I mentioned comes in.

It works. Right now, since the approved chapel, when Wavejols became possible, so unstaking and an Ethereum position, we see the derivative rise, so we have more people staking, so we're very happy about that. And if you have a Ledger and you want to store, well, to summarize, just about us, then I'll talk a little bit about advice, I'll go very quickly: we have approximately $1.5 billion in assets. These are n't really assets under management, it's because we're not a central bank, so we don't have the key to the funds, we can't leave with them.

We just have the right to validate transactions with clients' money, but we don't have their private key. That's very important. We're about 3% of the inner network, a little less than 3%, and well, for the moment we have very good performance in terms of optimization on these variables, and we invest a lot in this: large infrastructure teams, many different types of validation clients. This is basically the open-source code that we run on these servers. We use a lot of it to be able to switch if there are any performance issues.

And slashing is basically the biggest risk of staking if there's a configuration problem in the infrastructure and we run the same aviator twice. There are some chains where we can get part of the collateral burned, and that's really not good. Here are some logos we've worked with. I just have three quick pieces of advice if you want to work in Diffa, especially on the institutional side. The first, for us, is that the market for the "big players," at least for our go-to-market strategy, is absolutely critical.

We go through the "big players." These are basically the companies where people secure their assets. There are several types. On the left, these are companies where you basically pay them to keep your assets safe. We use them for private investments.

I'll get to that later. It's a kind of service. Okay, one and to the right, it's more about technology so you can better secure, for example, your Ledger. It's more on this side to the right. So, in fact, we now partner with about half of these companies, but that's what held us back a lot at the beginning.

It's because these are the companies through which people will make their staking transactions, so it's essential to be well integrated with them. It's a market you need to understand. Secondly, certifications and insurance are very important for this market. So, for example, we're SOCK2 certified. It's a security compliance thing; it's quite a long process, but it's super important to give customers confidence in our service and operations, and also to provide insurance.

So, we're insured with Chainproof, which is a regulated insurer reinsured with Munich. So, they're reputable, and we offer this to our customers, plus the option to buy additional insurance with Nexus Mutual. who really has a strong presence in insurance plus an insurance IFI project and with a chain, so the message here is that if you're launching in this market, it's going to be super important to be extremely compliant with all these things, to get the certifications, and in fact, that really speeds up the sales cycles with clients if you can show this kind of certification. And the third and final message is about the chat, our favorite chat here, which, when we personally read, we read different things than if we work in the institutional market. I think we're not all on Twitter and Discord and all that, but I found that working in this part of the market, it's important to read, for example, Blockchain.

That's really where all our clients are. Maths, the vines of Bloomberg, they have a really incredible newsletter on finance in a more global way, but for me, I loved it, it's excellent. And the Unchaine podcast, third recommendation, so I think if you want to work in institutional crypto, well, I think it's interesting to To change your diet and eat carrots, OK, and finally, it's still an event about how to work in Defy, so we're also hiring if you're interested in quite a few roles. We're in Paris, we have about forty people at the moment, and if it's not to work with us but you also want to come during THC, for example, we're going to do quite a few events, so there's a code for that, but otherwise you can find us, who's funding us, or who's on LinkedIn. We're going to do quite a few events, so if you want to come, sign up, and that's it.

Thank you. [Applause] Thank you very much, Ernest. We're coming to the conclusion of this evening. I would like to extend a big thank you to all the speakers. Honestly, it was extremely diverse, extremely practical.

I think you brought a lot of value to the participants, so really, thank you. Thank you also to Rome France and Commir for hosting us here. And finally, I'll keep you for a second longer: if you ever want to know a little more about the association, if you're a young professional, why not join us or simply stay informed? You have all the details of our upcoming events here. Have a good evening and enjoy your cocktail!

[Applause] [Music]