Panel | Building secure wallets for all | ETHDam III - 2025

Welcome to Easter to Isa to Isa to Easter. So to join on this panel, I'd like to welcome and I don't want to butcher it, Constantine and also Greg to the stage. Please give a round of applause. I'm going to sit here. Um yeah, can take a seat.

Um now I want this to be as interactive as possible and this is really for you. So, please do scan the QR code and after we get f through the first 15 minutes or so, we'll bring up the slido and go from there. Um, it's quite funny. Everyone's been looking for Greg. We didn't know whether Greg was here and then suddenly Greg was sitting over there.

So, we're really happy. So, welcome Greg. Um so just to kick things off um please could you just give a very short introduction of yourself and um why why you're working for a wallet or building a wallet company. It should if you Hello. Okay.

Um a quick introduction myself. Um I've been um in the blockchain uh space for a really long time personally. Um, this is my first gig in a crypto company and I'm super excited. Um, I went down a big rabbit hole um, and learned to code. I learned a whole lot of languages.

Um, and yeah, so I'm I'm super excited by it. I've always been interested in in the philosophy um, for a very long time. Um, and now I'm kind of like living the dream if you can call it that. Nice. And how long have you been there now?

Um, I've been with Trust for about six months. Okay, cool. Um, previous to that I worked in fintech and a lot of kind of like innovation type space um and startups and I had a few of my own startups as well. So nice done everything. Thanks for sharing.

Thank you. Constantine. Uh, I work for Cake Wallet. They hired me due to my open source work in wallets in general. So, uh, yeah, I'm now getting paid for it.

Hey, nice. Uh, hi everyone. As I kind of introduced myself earlier, but Paul Puy, CEO and co-founder of Edge to precisely answer your question, JJ. Um, I got involved and built I helped build this company because in when I got into Bitcoin in 2013, it was insanely hard to hold secure and secure Bitcoin and obviously other cryptos. And I foraw a way to simplify it using existing technology and tools, but all had to kind of get packaged together.

And so I was really excited by hiding the complexity of self-custody and privacy to really be able to drive it to more of the users that I want to on board. Um and so like that key management is probably the core piece that you know brought me interest in building something in this space. Cool. Thanks for sharing. And I guess so I'm looking for a short answer from this from each of you in quick succession which is in your own words what is the role of a crypto wallet like what is it really supposed to do in your eyes for the users other than keeping your crypto secure like that's that could be the answer right I'm just interested in what your quick take would be quick take is I hate the word because it doesn't actually tell you what the app is supposed to So, I if I could not have the edgewallet Twitter account and have like an edge something else, I would.

Um because yeah, I'm I'm starting to hate the word. I'll be honest. It doesn't tell you what you're supposed to do there. Everything from a USB stick to a piece of paper. Yeah, that's fair.

I see it as an entry point into the broader ecosystem of cryptocurrency in general. Would you would you then agree with the word like super app some or like those those big like those Yeah. those big apps that have all the functionality or because it is a it's a yeah kind of kind of. So I would say that's the danger of um our ecosystem of wallets is that because unlike banks you have all this regulation. Here's what you can do here's what you can't do.

Software can kind of do anything and then you're tempted to do everything. um in essence, hence why I don't like the word because I think some quote unquote wallets specialize and they should they're better at some things than others and you should use what you need for that specific purpose. Um you know some are really good at being key signers for DAPs. Others are really good at payments like they're really good for being able to go and pay other people really quickly um or the functionality of a specific chain to do governance. And then others such as what we focus on is being able to buy, sell, and trade.

So like really focus on that. So they're different apps and so hence it depends. And some are good at being the onboard, you know, as Constantine said, and some are terrible at that, but they serve a different purpose. And Greg, yeah, I think I kind of agree with that. I think um definitely we finding a lot of people use different different wallets for different specific functions.

Um we definitely have a strong ambition to be a bit more of a super app. um in particular, you know, the way I see it is is there's a whole group of people who are not deep into crypto who need to be who are interested in the the utility value of what uh web 3 can offer. Um but they they need a sort of a web two experience and so for us that's kind of where we're heading giving this really easy to use experience using web 3 rails so you kind of get all of the features of that. Um I think in a more philosophical sense um you know I'm I'm a big believer in in the libertarian values and I think the wallet is the facilitator to that. It's about freedom.

It's about owning your own assets. It's about sort of maintaining control and you know self-s sovereignty is very important. So that's where I see the the the purpose of a wallet uh in in my my view. Cool. Thanks for sharing.

Um, so the the key thing here is obviously this topic is about building secure wallets and one of the reasons why I wanted to ask that question is obviously the more functionality you add inside of an application maybe you end up losing focus on some of the core things about security and so I'm interested to hear kind of um from you kind of what are the key things that obviously you you you are doing to keep wallets secure for the users um but also on top of that like what are any trends that you see in security or wallet security that maybe are either brand new or are going to be coming over the coming years. Should I go first? Um, in terms of what we are doing in terms of um, wallet security, um, we're moving towards as, as I said earlier, we we're trying to focus a little bit more on UX. Um, and and there's obviously a little bit of tension between security and UX. You know, you want to make something easy to use but also very secure.

Um, and the path we are going down is what I would call just in time security. So we we very much care you know number one thing for us is all about security. Um and that's always it's always been like that and will never change. Um at the same time we like to we like to provide security at the point in time that it makes sense. So for example when you have an empty wallet there's no reason to back it up because it's empty.

So things like that small things to improve the experience. Um the same thing with security warnings. um only when I'm starting to do something should you be given warnings about you know this is potentially malicious etc. Um so we're very much focused on like how do we provide guidance um and security at the right point in time um which also helps to give people a really great user experience as well. So that that's kind of how we are looking at the world uh in terms of security.

Think about it. Yeah. Um I think the thing that we've been actually no the thing that we factor in when we think about security and what are we going to implement is how are people actually getting breached right look at the history of especially cryptocurrency because it's a whole new paradigm where data is money so how are they getting breached um and what we've primarily come to that's in our control is the human mistake factor. And so how are people imperfect? We're all imperfect.

we admit that I'm imperfect, whatnot. And so what are the things that people are usually messing up and then build technology to address that? Uh I think that the industry has been a bit too and this is super controversial. I think people are going to hate me for saying this, but the industry has been too focused on securing people against a thing that has not been what they've been losing money from. All right.

um which is the really like nuanced like malicious virus on device that can read memory and you know read disk space from other apps and get access to the keys. Now that can happen in theory majority of how people are losing money is they're just mismanaging the keys. They're putting them in insecure locations. They're writing them down, putting under the keyboard, putting them online on, you know, Google Drive unencrypted. And so fundamentally that is what I think needs to get solved.

And the way we've decided to solve that is to have people just simply not do that, right? Don't actually have people see keys. And like I said, this is controversial. Many people that would onboard people say the first thing you do is you back up your key. Well, you've just taken the most dangerous thing that you could show to the user, put it on the screen, right?

which then potentially could get you know screen you know screen captured or then have them write write it down you hope they write it down as opposed to typing it somewhere else so I think more focus needs to be done in that and we've been doing that for 11 years as focus on removing the actual private key from visibility from the user still having them access to it if they want it if they really want it sure if you can be careful with it but to the masses and to the average person it's it's a ticking time bomb Well, I have nothing to add. I I I'll just add to that. I kind of agree with the the the principle that most most of the the people that tend up being defrauded quite often it's it's sort of social engineering where the risk comes in. That's where most of the issues happen. So I I kind of echo that.

I think um generally you know the ecosystem is pretty secure in terms of in terms of like robustness in terms of security protocols. it's always ends up being some really crazy social engineering, someone phoned them or you know those kinds of um exploits are where we're finding the most issues particularly for for you know a well-established wallet that's got really good processes and and security protocols in place. So I think that's that's the thing we it's how do you how do you fix as you were saying how do you fix people very hard social engineering is always going to be first have to admit you're broken. Yeah, it was crazy. Last year, my dad actually ended up transferring all of the money out of his bank account because someone from Microsoft called him and it was just like, how have you done this?

Um, but I'm sure other may have other people may have experienced that. There was an interesting uh panel yesterday where they also really highlighted the point of using multiple devices for managing your keys versus like your work laptop or your personal laptop. Just out of interest, does anybody in the audience use more than one device or has airgapped their device or use another phone for all of their crypto transactions? Put your hands up if you do. Okay, that's actually quite a few people.

That was more than I was expecting. I'm happy with that. We are at a privacy hackathon focus conference. So, you're in the right place. Cool.

And well so moving on from that um kind of we we touched upon it in terms of what you're saying in terms of the social engineering um obviously as more apps uh build or wallets build more functionality in where they for example show all of the tokens that are available or all of the DAPs that you can now integrate with or different tools that you can use. Where do you see the line in terms of your responsibility as a wallet provider of protecting users of what they can interact with with their wallet versus kind of enabling them to go and interact completely freely and potentially lose their funds or uh fall victim to some form of attack. Yeah. Um and and the way we kind of look at this is very much um provide guard rails um but also allow people to do what they want to do. So um we have a lot of like um security warnings risk um assessments on the various tokens.

Um for the most part we don't block any um D5 or D app connections um unless we are know we know that they are very malicious or outright malicious then we will we'll blacklist them. Generally, we let people do whatever they want um with some pretty robust warnings um around that depending on the the risk level um because I mean that the point of self custody is you know it's your choice and so you need to have to you have to live by that principle as much as you want to protect people from themselves um you also need to you know give them give them their liberty as well. Yeah. And I'm interested in your guys because I believe both Cake Wallet and Edge they're more privacy focused wallets versus Trust Wallet. I'm not trying to call you out.

I'm just putting it how it is. Right. And so I'm interested in in yeah your guys' take there. Well, we mainly let the user do what the user wants to do. We don't want to compromise the UX too much.

M I mean yesterday there was a panel and the guy from Polygon said the most important thing for your privacy or safety is just read the warnings. And if you don't read the warnings then there's only so much we can do besides blocking you from doing something which is something we do not want to do. I I'll attest that if people don't read the warnings um but we need to put them there. So um at least to be able to show and you that do listen. Um so I think twofold answer that question.

Microphone check. Here we go. Twofold answer to that question. A the warnings and so no I agree with Greg there. We need to identify what could be very highly likely malicious.

And that's, you know, not super impossible in the DAP world because it takes takes money to spin up a DAP and a a protocol. And so once you detect it, you can you can blacklist it. Um, but I think a secondary method that I think the world needs to get to, the crypto world needs to get to, I hate to say it, but I really cannot stand the separation of the key manager, I won't even call it a wallet, your key, your signing key manager and the DAP. That separation is just begging for security issues and it also presents a terrible user experience. It's like you know we said oh there's a fine balance between user experience and security.

This one fails at both and so at the end of the day those features that are in the DAPs should just get baked into some of the wallets that fit that definition of their use case. So, a case in point example, one of our staff members lost a few hundred dollars connecting to a bridge that was malicious and fake that can easily get integrated into most wallets. We didn't have it integrated at that time. So, I almost feel bad if we had it integrated. She wouldn't have had to use the bridge.

It would we would have already selected the protocol, right? We would have whitelisted it by virtue of it being in our app. It would have been both a a better user experience and there would have been no way for her to connect to a malicious bridge. So, I think integration into the different wallets is the long-term solution and in the meantime, it's what Greg said, we've got to warn people, identify, blacklist the ones that we are 100% sure, you know, are malicious. And was that a third party bridge or one that you developed yourself?

The one that stole the money, we did not develop. No. Oh, you mean the one that we could we integrate? Yeah. No, we can integrate third party bridges.

Yeah. And so that goes through a whole auditing process, goes through a process and a few different companies will go through that audit like we would the bridge and dex aggregators that we connect to as well. Um, and we know that there have been at least successful transactions over some period of time in comparison to going to a random website you Google um, and hitting the wrong link, you know, because they happen to have good search results and then there you go, you've connected to a malicious bridge which has not sent any money, you know, they've just been 100% malicious. Yeah. That's fair.

And um when there are also like third party providers like for also transaction or transaction simulation and kind of these sorts of things. There's also a company called Blockade I believe. Um do you guys use any third party providers for any of that analysis? I guess maybe you would maybe you guys not so much. I'm just interested in kind of like what are the players that you see that can help wallet builders secure their wallets and or secure the users actually is better.

We are not currently using those but we've evaluated several. Yeah. Um and just trying to find what we think is the best like there's another mic here. Okay. Mic check.

Cool. All right. And so we're looking at one that has pretty good identification rate because some of the ones that we've looked at um didn't actually flag uh some known scam contracts, tokens, and whatnot. Um but I think it's it's an important piece and we definitely do want to have that integrated. Um yeah, we we have a number of different um providers that we work with.

um for depending on what we're looking at whether it's um a token specifically or contract um or de apps just depends. So yeah, we have a range of different providers. We're constantly evaluating as well and um not only that also looking at like our you know how often we check these lists etc. That's one of the big issues is quite often um you know the providers you're still only as good as your the speed of your lookup and and you don't hit the API every second you know you have to have some sort of a time limit um on how quickly you hit these security APIs um so we are continually evaluating that and improving it u but it's always a challenge just in terms of the volumes of data you know we because we have 100 chains and like millions and millions of assets um to pull all of at every second is just like overload crazy. Yeah.

So, um we we were doing it on sort of I believe was sort of like a 24-hour basis. Um but we've chopped that down to like minutes now. So, which is good. Yeah. You have your own token list as well, I believe.

Yeah, we do. We do. Yeah. We've just done a whole lot of work improving our improving our token lists. Um previously, it was all manually managed.

Um which meant we were quite slow on on new things coming to market. Um but we've just done a big piece of work to automate the whole process so we can pick up hot trends as soon as they hit the market. Um again the issue there becomes security. You know how do you protect people from something that's a hot trend um that hasn't been scanned yet. So that's the challenge we haven't I don't think anyone's solved that just yet but we definitely I think doing some good work in that space.

Do you guys use any token lists or anything like that or verified DAP lists at all? Well, we do have our own token list as well, which is public in our GitHub, but nice. Yeah. Yeah, we're currently manual, so we whitelist tokens, although people can add them manually. So, we can always bypass our our list and the list is is growing.

Our goal is to not have to whitelist and basically allow people to add anything that they want. And of course, the challenge with that is exactly what Greg said is how do we keep them away from what would eventually be kind of malicious tokens. Um but that definition of malicious is very very vague. Yeah. Uh could we put the slido up?

I want everyone to try and vote on this poll really quickly. Um and the poll basically is would you rather have scams blocks scams blocked? Yes or no? I'm interested in kind of what your what your thoughts are there. Um and maybe we'll come back to it.

But I I it is it's a very touch it's a very difficult um um topic. and then we'll come around to any questions uh on this as well. Um, so I guess like my last question that I have is kind of over the next let's say 5 to 10 years when it's coming to wallets like what do you think is is going to be kind of the big things around or big trends that you want to see or what are maybe your goals for the next 5 to 10 years would you say? Onchain privacy obviously. Yeah, that's our main focus with Cake Wallet.

M we try to pioneer that in other chains as well. For example, we I believe we are the first wallet that added Mimblewimble for Litecoin. We were one of the first wallets who added Payjoin as well which is something we are rolling out over the next days for Bitcoin and we also have silent payments for Bitcoin as well. So we are trying to improve onchain privacy. This is mainly our own focus and I very much like that that comment that you said about maybe integrating some of the features a little bit more natively because I think it's so difficult to go and find out verify and you want people to go do that work for you in a way obviously comes with an associated risk but it's about the usability.

Um in terms of in terms of where we're going um we have lots of big ideas obviously trust wallet um one of the things we're definitely exploring a lot is the idea of a super app um and how we can become sort of the the financial asset center for an individual's life. Um in addition to that we are also looking at like um one of the things we're talking about again this is very far into the future um is like how do you start to um use your wallet as sort of your your self identity um how does that then tie into things like government services things like that. So there's some really interesting um ideas in that space around like how does this become your your sort of like the holder of not just your assets but also your identity to some extent. Um and I think that's kind of interesting. Um, and so that's kind of like the the areas we're exploring for sure.

Um, I think the other thing that for me in the in the in the shorter term that's super interesting is we have a really growing audience in what I call utility users. So people who are using our wallet not just for speculation and trading but actually for like real world use cases which is amazing. Exactly. Yeah. So um and there's a big there's a big audience particularly in like the remittance space um particularly in in developing countries um and we seeing a lot of growth there um for me that's super exciting because it's like the holy grail of crypto is you know actually using it for stuff not just you know the the the trading and speculation which has kind of been really great and founded everything.

Um, for me that's, you know, that that's what kind of got me super excited about trust as well is, you know, people are really using it for doing real things, you know, which is kind of pretty exciting stuff. What we want. Quick answer, Paul. I see the suite of apps, not the super app. And so I look forward to when there is a suite of different apps for the different purposes.

As I'd mentioned, not having to connect to anything I want to use via a wall connect or um like a browser extension. So that's number one that I I think I really want to see in our future, right? And I think we definitely want to be able to build that too. Not just one edge app, but several for different purposes. M um and then second privacy.

But by default, I don't want to see a single switch in setting that that's necessary to enable an advanced level of privacy. I just want any user that just turns on the app, opens it up, makes a transaction to get the highest level of privacy that we can deliver. And so that is one of our missions is everyone gets it, not the people that are willing to jump through the hoops, enable this, run your own node, this, you know, whatnot or mix this after you send a transaction or after you receive money. Yeah, cool. Thanks for sharing.

And I'm very happy to hear that about the privacy part, like we need more of this. It's about time. Um, we got quite a lot of questions, but first a review of the poll. It was a 50/50 split. It got taken away and I took it down, but it was a 50-50 split, so everyone's right.

Um, so the first question we have is, uh, John John, is it John McAfee? John Fee John McAfee thought wallets on phones was a bad idea security wardens privacy. What's the argument for mobile wallets today? And is it possible to get the slider on the back screen so the guys can see it? So the question was, John McY thought wallets on phones were a bad idea.

What's the argument for mobile wallets today? I I the argument today I think is just um ease of use. You know, everyone's life lives on their phone. No one carries around their laptop to do their banking uh for the most part. Um and so it's makes sense for the for you know your crypto wallet to live in the same place.

Um yeah, I don't I guess there's some element of risk but you know it's the same could be said for your your you know your banking. people get their their phones taken out their hand on the street and then someone hacks into their banking account. So, it's no different to that as a principle. But I think always uh ease of use is going to trump um is is there some element of of risk in it? I would say access because we quite have a lot of users that don't even have notebooks anymore.

[Music] So cutting them out of the the ecosystem would be just unfair to not provide them with the same level of privacy. I think you mentioned your direct feature. So um on bitcoin.org there's a list of wallets. They actually have different flags for different features of different wallets.

One of the flags is it runs on a more secure um environment. That's specifically for mobile, not for desktop. So even as far back as when that website was built, they recognize that mobile devices from the viewpoint of the devices that you could own are actually the most secure because they were the devices that at the time of launch were already launched with what's called app sandboxing, meaning one app can't access the data of another, whereas desktops by and large were not. And so um and I still fundamentally think this is true all the desktops are improving. So while no digital device is perfect um aside from purchasing a much more specific use device like physical device such as a hardware wallet or a dedicated phone um there is no more secure device generally speaking than your actual mobile phone.

Digitally secure now physically you might lose it sure but digitally secure it's actually your most secure device. So, I think yes, I I don't want to just re regurgitate the usability cuz that's that's kind of known. Um, but I think it's overblown that that's an insecure device as opposed to it actually being your most secure aside from once again customuilt hardware. Cool. Moving on to the next question.

Um, does the lack of a proper decentralized forgot my password functionality block mass adoption? Disaster recovery seems to be a big mental barrier to adoption. if you have any words of wisdom on that. So, first, I don't like the word decentralized because I don't actually know what that truly means when you've accomplished it. Um, we actually, so really it's not a decentralized forgot my password, but a self-custody version of forgot my password.

Um, and that actually does exist and we built that 11 years ago. So there's an ability to do password recovery without edge having access to the actual keys or the actual password like using some you know novel uses of split key cryptography. So I think we have to really ask ourselves what with the tooling we have can we build a lot of that original web 2 experience and I think if you think hard enough with some sharp people you can um and solve exactly this kind of problem. Well that's what also account abstraction led into as well right? it it can although it's so limited to well ETH and so if you are and once again if you're an ETH wallet and that is what you're trying to support then great if the fir the the purpose of the app is you know multiasset multi-chain trading then that doesn't really solve the the problem as a whole so it depends on your purpose yeah any other thoughts otherwise we'll move on cool um warning customers is good to two security is most important but most users is this a question or a statement I think it's a statement and there's another one after that as well another statement around the poll question around definitely labeling something as a scam it's a very good point I take that as good feedback for the poll um what is I think there's there's a point around the whole scam you kind of mentioned it earlier is that um there's varying degrees of scam and like how do you how do you classify at what point something is a scam or not, you know, and I think the best because of that kind of slightly greyness to to like contracts and how they might be executed.

Um, I mean, even some people called hacks, you know, it's not a it's not a it's not a hack as such. I'm just using the protocol in a novel way, you know. So, um, the the way you qu you quantify it is quite difficult. And so, you know, some some level of like warnings is the best is the best you can do. I think like blocking stuff or outright saying this is a scam is you're just never going to be able to do that cuz how do you how do you actually um define what exactly what that is?

It's almost impossible. That's fair. I think there's an interesting overlap then with especially like uh the MA regulations coming up for better or for worse. All right, don't shoot the messenger. But ultimately it's giving a label to something that is a definitive this is not a scam one way or another.

Um cool. move on. Uh AI, does AI pose any kind of threat from your perspective? Yes. Okay.

Why? Well, mostly because it makes social engineering easier. And like like Paul mentioned earlier, social engineering is the main reason why people lose their funds. So you can also automate social engineering with AI. So yeah, I I think the biggest threat is AI is going to make all of us expect AI in all our apps and the last place you want AI in is your wallet.

Um, so we've already gotten people like, "Oh, can I just go and talk to my wallet?" And you know, um, say it say like send over to this buddy of mine and that all everything you're doing and saying likely without the invent uh advent of much much stronger hardware, we'll have to go to a cloud and so you're giving up both privacy and security. So, part of me wants people to not get so used to AI and apps so that you can we we don't have to expect that out of our wallets, but that's not going to happen. Um, and so I'd be wary of any app, especially a crypto wallet featuring AI functionality unless it's on device. And that's one thing we're looking at is ondevice AI is just super hard to do and very limited.

But once we can do it crossplatform, then we'll integrate those kind of features and be very proud that nothing leaves the device, you know. But that to me is like one of the bigger dangers is just our expectation out of of out of AI. Yeah. Cool. Yeah.

The ondevice a part looking forward to and I've been spending a little bit too much time in the Salana ecosystem and there was this whole wave of AI agent stuff happened earlier this year. It was absolute chaos and security out the window. Um, last question I'll take. Uh, and this is for Trust Wallet. Maybe you can't answer this, but will there ever be a Trust Wallet name service?

Not sure. I don't know the answer. We kind of we do comment. We do currently have trust handles which is um I think it's through fear um which is sort of the same thing. We don't have a plan currently on the road map.

However, um just in terms of direction of travel, we do want to um definitely make a bit more of this feature. Um so potentially that's something we might look at. Yeah. I just don't want every single app to have its own name service. We need to use common name services like ENS or open alias, which is something native more native to the Monero community, but it is blockchain agnostic.

So I mean the most blockchain and also exactly what Greg said is already in trust and also in edge is FO right it's a blockchain that unlike the other name services is intentionally trying to be used for nothing other than name service. So it's not trying to compete whereas ENS you know is on Ethereum which is trying to be the global computer competes with other protocols and even the ones on private chains like Monero's and and Zano's and Dash's name service. So, by trying to be almost like the Switzerland blockchain and not trying to upset anyone with its functionality, it's actually been a a really good solution for crosschain name name resolution and I'm really happy with the service. It also let you do messages, encrypted notes in your payments, and so um recommend people try it out. Cool.

Well, thank you so much gentlemen. It's been an absolute pleasure. Thanks everyone for asking question and engaging. Um so, please give a round of applause and uh that's that.