In this conversation, Stephan Livera and Kevin Loaec discuss the challenges and advancements in self-custody and multi-signature (multisig) wallets for businesses. They explore the reasons why many businesses are hesitant to adopt self-custody, including regulatory constraints and the current limitations of available tools. 

The discussion highlights the evolution of multisig, the risks associated with centralized custodians, and the introduction of Liana Business as a tailored solution for businesses looking to implement multi-sig setups. They also dive into the importance of defining roles within a multisig framework, the mechanisms for recovery, and the future of recovery paths in Bitcoin. 

Kevin also shares various aspects of Bitcoin custody, focusing on time locks, key management, and the importance of recovery pathways for businesses. He emphasizes the need for user-friendly solutions that mitigate risks associated with key loss and the complexities of managing multiple wallets. 

The discussion also touches on the role of insurance in Bitcoin custody, the potential threats posed by large custodians, and the challenges of maintaining privacy while reusing keys across different setups. 

Takeaways:

🔸Businesses often avoid self-custody due to regulatory concerns.

🔸The tools for self-custody are still not user-friendly enough for many businesses.

🔸Multisig wallets have evolved but still require technical knowledge.

🔸Centralization of funds in custodians poses significant security risks.

🔸Liana Business offers tailored solutions for business self-custody needs.

🔸Recovery pathways are crucial for ensuring access to funds.

🔸Insurance options can mitigate risks associated with key loss.

🔸Employee turnover can complicate key management in businesses.

🔸Replay protection is essential to prevent loss during forks.

🔸Future corporate custody solutions may involve hybrid models with banks.

Timestamps:

(00:00) – Intro

(01:06) – Why aren’t businesses doing self custody right now? 

(02:55) – Evolution of multisig and Hardware Wallets

(07:51) – Are there centralization risks with custodians?

(10:24) – What is Liana Business?

(13:11) – Multisig configurations with Liana

(15:24) – Is Liana business optimal for businesses of all sizes?

(17:31) – How does Liana Business address role-based governance?

(25:03) – What are some of the recovery mechanisms in Liana?; Recovery paths 

(29:04) – Understanding Time Locks and Wallet Management

(31:02) – Sponsor

(32:00) – Who are the key holders while setting up Liana business?; Insurance in Bitcoin custody

(37:06) – How should businesses manage key loss?

(38:47) – What are some of the threat vectors in Bitcoin custody? 

(41:07) – What are the privacy concerns with hardware key reuse? 

(48:23) – Seamless key management for businesses

(53:40) – What is Liana business’s USP?

(55:27) – What is the future of corporate Bitcoin custody? 

(58:49) – Balancing convenience and security in Bitcoin custody

Links: 

• https://x.com/KLoaec 
• https://x.com/lianabitcoin 
• https://lianawallet.com/business/ 
• https://x.com/wizardsardine 

Sponsor:

• CoinKite.com (code LIVERA)

Stephan Livera links:

• Follow me on X: @stephanlivera
• Subscribe to the podcast
• Subscribe to Substack

Transcript:

Stephan Livera (00:00)
Hi everyone and welcome back to Stephan Livera podcast. Today we’re going to be talking a bit about Multi-Sig, Mini-Script, Self-Custody as well as in the business context. Now rejoining me on the show is Kevin Loeck from WizardSardine. Now you may know them for their product known as Liana or that’s the name, it’s called Liana. And I noticed you guys recently brought out ⁓ Liana Business. So I thought this would be a good opportunity to chat because…

This is kind of a known thing in the industry where, you know, it seems a lot of people are not doing self-custody right now. So maybe let’s start there. So in your view, why aren’t businesses doing self-custody right now?

Kevin Loaec (00:39)
Yeah, thanks for having me first of all. ⁓ Yeah, why our business is not doing self-custody? I think…

So of course, different business have different reason to not do self-custody. Some of them it’s for regulatory reasons. And so these ones, you know, it’s very difficult to get them. So for example, in the US, if you’re a publicly traded company, you might have to use a custodian. And so of course, like that just remove the option, but then it’s not the case everywhere else. So different places have different regulations. And sometimes it’s just, oh, we see the companies in the US are doing this. So we’re going to do the same, like keeping our funds at

Coinbase or something like that when they don’t necessarily have to. Just sometimes looking at what the others are doing and you might think it’s the same applies to you, but it doesn’t necessarily do that. ⁓ I also think the other reason and probably the main reason is actually that the tools aren’t really great for businesses. So, you know, we started having multi-sig wallets a couple of years back or a few years back, but still like for many the multi-sig is still using multi-sig on one computer, maybe having different device.

but then just as one individual, not really as like a group or a team. So seeing things like coordinators are pretty new. How do we make sure even like transaction labeling and things like that are actually coordinated between users. So of course it’s simpler when it’s a custodian and it’s just a server somewhere to keep track of everything. But yeah, the game of course in Bitcoin is trying to offer self-custody. So trying to the tools for that.

Stephan Livera (02:12)
Yeah, I guess let’s just kind of walk through. I’m just going to kind of loosely walk through a little bit of the history, at least as I recall it and understand it. You tell me, I mean, you’re closer to this than I am. But it’s like, let’s say Mike Belshi from BitGo, right? He had some interesting innovations back in the day on multisig, right? I don’t think he invented multisig, but let’s say he made it really practical to use, I think P2SH multisig. And that’s like maybe 2013, maybe 2014 around there, something like that.

that was like not really practical for everyday people unless you were like a wizard basically, unless you were like super technical. And then, you we’ve had this innovation of hardware wallets. We’ve had things like, you know, Electrum has been around for a while and nowadays things like, you know, Sparrow and, you know, Nunchuck and now your product Liana and Keeper and various solutions are out there, but it still seems like you kind of have to be a bit techie to use it. It’s gotten better though. And I think that’s, it’s fair to say it’s gotten a lot better in terms of like,

the way we coordinate things like let’s say you’ve got different hardware wallets that form a multi-seq setup, let’s say a two of three or a three of five, which would be like a typical thing. ⁓ But I guess as you said, it’s regulatory reasons for some people and maybe it’s just the tooling is not quite there yet. How are you seeing that? Like, do you agree with that summary? Where do you think I’m missing pieces there?

Kevin Loaec (03:33)
No, I think you’re doing great. I really think Electrum was like the big thing that really started the multi-sig on the business side of thing. A lot of people are actually still using Electrum for that reason. Although right now, yeah, you hear a lot more about Sparrow, just like the feel of the software is much, much better, I feel as well on Sparrow. And then yeah, new tools like Liana, we’re trying to make it even simpler. So if you use Sparrow and you move to Liana, of course you have a lot less ⁓

kind of bloating in the interface, right?

It’s not really built for the advanced user as let’s say Sparrow would be. yeah, I completely agree with that. There is also another part though, which is the hardware wallet, as you mentioned. Their UX also improved drastically. It’s still not the simplest thing, but at least now most hardware wallets support PSBTs, support multi-sig in this sense, not just blind signing and having bad security. So it’s also like a whole thing of the industry moving forward and making new stuff.

MiniScript is another one, OutputDescriptors is another one, I’m using technical terms here, but it’s like small steps we did that actually make multi-signal really usable, not just like a hacky thing like back in the days with Electrum.

Stephan Livera (04:51)
Yeah, I see. as an example, think another example that relates to this is like registering the multi-sig quorum, right? So back in the day, like maybe that wasn’t as much of a thing or the hardware devices didn’t all support that as well. Whereas nowadays that scene is like, this is just a part of your normal multi-sig setup. So when people, know, for listeners, if you’re not familiar, like the basic thing is like, you might write down your 12 words for your, you know, that’s your private key of your cold card or whatever.

But then let’s say you’re registering that into some kind of multi-sig setup, it’s like that’s how the cold card, in this case, knows the public keys of the other signing devices. And so then it knows how to generate those addresses in a way that’s a bit safer than what was before, right?

Kevin Loaec (05:36)
Yeah, and it’s actually critical for security, not just for the receive address, which is already something, but also for the change address. So as you know, and maybe the listener doesn’t, but when you use Bitcoin, when you spend some coins, when you spend UTXOs, you are actually destroying the UTXO, so the coin you have, and you are creating new ones, some for the recipient and some back to yourself as the change. ⁓ And the real thing here is that if your hardware

wallet ⁓ is not aware of how to generate addresses, but not just for deposits, also for the change, then one of the kind of attacks that was performed back in the days was actually that the software would lie about like the change and it actually didn’t go to you, it would go to the attacker. So your transaction looks normal, the funds are actually going where you’re trying to send them, but it’s just that every leftover you had in your wallet would also go and to the attacker without displaying it to you. So this is why it’s extremely

Stephan Livera (06:16)
Right.

Kevin Loaec (06:36)
important for your hardware wallet to be able to generate the change address so to be aware of every other keys because otherwise like your wallet could just be emptied without you noticing.

Stephan Livera (06:46)
Right, and that’s why the better hardware wallets nowadays, they do like a change check, right? They check that the change is coming back to my address, so to speak. So to make sure it’s not, know, because that’s a hacking vector. And in some cases, either literally just going straight to the hacker or in the case where in some circumstances, it was like a ransom attack where it would like, they would send it to like some really weird place that they, you know, the hacker could then leverage that against you to say, hey,

you better pay up or otherwise you’re not going to see that coin again. And so that’s obviously something for people to understand. So, yeah, as you mentioned, like, it seems that there’s been a lot of, at least for now, there’s a fair amount of centralization into some of the big name custodians, right? Like the big ones, Coinbase Custody, Bitgo, Fidelity Digital Assets, Anchorage, and maybe a couple others, but there’s kind of this bracket of, let’s say, tier one kind of well-known custodians that almost all the big players go to.

Kevin Loaec (07:18)
Thank ⁓

Stephan Livera (07:43)
Is that a centralization vector? What do you think?

Kevin Loaec (07:46)
Yes, centralization not necessarily in the sense of like governance, right? ⁓ I don’t really think these people have a lot of power over the network. We kind of saw that as well with the big blockers versus small blockers where you had companies like Coinbase potentially supporting the other side. ⁓ And still, they didn’t really have… ⁓

bargaining power in this sense. But there is a risk though of centralization, which is actually for the hacking risk or for the threat of a motivated attacker. Because of course, if you know you can break into Coinbase, you will get billions, if not trillions at some point of dollars. That’s actually terrible, right? And this is what we’re doing right now by really centralizing a lot of money into this custodian is that we are like increasing the ⁓ incentive to

break into them by a lot. so of course now with ETFs and things like that, like also doing custody at Coinbase, for Coinbase it becomes a huge risk of course. Like how are you going to, if I was Coinbase, know, how am I going to enforce a level of security that could be resilient to multi-billion dollar motivated attacker, right? And this is extremely hard. So of course there are, you know, nation state level attackers trying to break into Coinbase and equivalent.

So yeah, that’s your centralization risk here is that the cost to defend such an amount of money is just insane and doesn’t even exist. You always need people at some point to be able to access these coins. And how much do we need to break these people somehow?

Stephan Livera (09:26)
And even to be fair, I don’t know, I haven’t looked into detail, but it could be that some of them are using more like omnibus style, you know, custody and others may be like, no, it’s segregated per large customer, right? And so maybe it’s a little bit less of a risk if it’s like segregated per customer and they’ve got like different setups for whatever this large company’s coins versus some other large company or some other exchange that they’re doing the custody for this kind of thing. So it mitigates it a little bit, but it’s still, yeah, maybe that is something there.

⁓ So let’s talk a little bit about Liana Business. I know you recently launched this, so give us an overview on that as well as what’s the pricing.

Kevin Loaec (10:03)
Yeah, so Liana business is basically a complete offering around the Liana wallet. So as you know, the Liana wallet is just an open source software. can go to our website or go to GitHub, download it, run it by yourself. You do your own setup. You use it yourself. Support is kind of just offered on a, you know, best effort basis. You can contact us on Telegram, email or whatever. But of course you can’t expect us to just spend hours on fixing your problem if you’re not a paying customer. We already build the wallet for free, you know, that’s

That’s quite enough. ⁓ For Liana businesses, know ⁓ we have quite a lot of business users using the open source version of Liana, or at least the free version because the Liana business is also open source. The main needs they have are around figuring out their policy, so how to set up the wallet and what kind of conditions to use. Liana is not like your typical multi-sig where most of

vendors are using like a two of three or a three of five in Liana it’s really custom and and we encourage you to pick exactly what’s what fits your actual governance in your structure and so of course not everyone is a security expert so part of the ⁓ of the onboarding sessions of Liana Business is really to help you figure that out so we are going to map out your normal ⁓ business operations so how do you usually do a payment do you need to have like two three people authorizing a transaction it’s just

Is it just one? How do you deal with issues if someone leaves the company, if someone is on holidays, whatever, things like that. So it’s really about helping you map the highest level of security you can without disrupting your operations. And with Liana, yeah, we can also have, of course, all of these recovery layers. So you can have recovery layers internally or externally. That’s what we call the safety net. So yeah, quite a lot of things we can help with on the design of things. Then on the technical side, Liana Business also

now brings the notifications. So you will get notifications if there is let’s say a specific deposit to your wallet. If somebody is crafting a transaction and is waiting for a co-signature you will get a notification. If some of your UTXOs, so some of your coins need to be what we call refreshed so they haven’t moved in a long time and it would trigger the recovery situation you will actually get a bunch of email before just to let you know hey you should actually access your wallet to prevent the recovery too.

happen. So this is part of Yana business and then a bunch of other things like safekeeping your encrypted descriptor on physical media and things like this.

Stephan Livera (12:42)
Okay, so let’s just walk through like a typical like I don’t know obviously as you said it’s custom and actually sorry one other question while we’re in the custom multi-seqs know a while back people used to say like Stay to the well-known multi-seq things like two or three or three or five if you go to the large ones You might like if people start doing crazy, whatever like 15 or 15 or 20 of 20 or some, whatever some 15 of 20 That can be difficult or you might risk not paying enough fees or things like that

Has that been resolved or is that kind of dealt with inside your app for the user?

Kevin Loaec (13:16)
It’s

pretty much resolved even at the Bitcoin layer, right? So with RBF being basically on by default for everyone now, you could resign transactions. It’s still very annoying if you have a 15 out of 15 because RBF means you need, again, 15 people to sign, but you have things like CPFP that works as well. So the recipient could just increase the fees and things like that. ⁓ So I believe from that perspective, it’s On the other side, there is also the way that wallets were ⁓ kind of dealing with multisig.

back in the days. So when you didn’t have output descriptors, ⁓ you would actually kind of…

use a hard-coded script so you would know that every two of three would use this specific script and you would order the x-pub in this specific order etc etc. So all of this was kind of like a very hacky way of making things work ⁓ because there is a lot of ways to do a multi-sig on bitcoin script and just you know doing the same script for everyone is not optimized in terms of fees in terms of ⁓ yeah a lot of things like script execution and things like that so things like mini-script ⁓ really optimize this and

and output descriptors are also fixing the problem of having to standardize everything for everyone. So this is pretty much solved. Not every wallet is still implementing Miniscript, as you know. Even output descriptors, still not everyone does it, but pretty much now it’s becoming a standard. ⁓ I believe every wallet that is used today would still tell you, hey, you should save your output descriptor. It’s not going to assume it can rebuild the script with your Xpubs and things like that.

Stephan Livera (14:54)
Yeah, yeah, it’s interesting to see how that’s evolved. So yeah, let’s walk us through just an example, just so people can get an idea. Like let’s say it’s a small business with a few founders or key employees. What would that look like in Liana business?

Kevin Loaec (15:07)
Yeah, great question. it really depends as well on the size of the business or what they do with the funds. So that’s a great example. ⁓ We might actually go down to some kind of recommended templates for the smaller businesses. So, know, the two of three with some recovery options or what I prefer, which is a two of two that then extends to a two of three with a third key. ⁓ So again, depending on the business, right. So for what we’re using at WizardSardin, my company, it’s a two of two requiring me plus

another employee in a different country that then rotates to another two of two, which is me again, plus a different employee and that then keep rotating including me being removed from the set in case something happens to me. So we have ways to do this in a kind of pretty simple way to explain on paper, but it’s just that you need to think about it. So it’s not going to be just two of three with one recovery key, although we could. It’s usually kind of better depending on who are your typical people that need to cause

sign. To clarify this maybe, so let’s say if you go to something like Unchained, right? ⁓ They might give you a 2 of 3 where your business has two keys and Unchained has one key. The normal way of signing is that the two people that have a key on your side should be signing. If one of them is unavailable you could ask Unchained for a signature on their side, right? So in this matter it means that for you, you should have always these two people available. It’s okay there is a recovery if they are not but you have these two people that are

required. On Liana, we can do the same with this two of two, and then you can have a third party or a third person that’s there only in a recovery option. And I think that’s strictly better than just a two of three, where any two of these three could be signing, so it’s a little bit more difficult to ⁓ modelize your threat model around that.

Stephan Livera (17:00)
Okay. Yeah. And so let’s just talk about typical examples then. So how, how can it work in a scenario where maybe you’ve got employees, managers, and maybe a board, like should the keys be distributed across those different kinds of people? how can you walk us through the different roles that might apply in setting up a Liana multisig? Yeah.

Kevin Loaec (17:21)
Yeah, yeah, absolutely.

Absolutely. really think ⁓ this question is on point as well. Again, it depends how your business works, but in a typical business, the people who actually do the transactions are not always, you know, the C level. You might have actual people that are in the, I don’t know, accounting ⁓ or payment teams that are going to be controlling the keys or some of the keys could be multiple wallets as well, right? Like kind of a spending wallet and a holding wallet. ⁓ So yeah, you need to define these different roles. ⁓ Typically, the higher you go,

or like the further from the day-to-day business you go, the lower it will be in the recovery scheme. So as you were saying, ⁓ what about the board? I typically believe the board has a very important role to play here, which is to have one or multiple of the recovery keys because they are ultimately the people who don’t want the funds to be locked ⁓ if something happened, right? So you want them as a backstop that if the CEO leaves the company and there is a board on top of him, ⁓ if the employees are leaving, if there is a disaster in the company,

still want the board to be able to recover the funds ⁓ through a disaster recovery thing. But the board should probably not be there controlling the wallet on a day-to-day basis. You don’t necessarily want them to have access to the funds. They are just here as an oversight and as a backstop, right? So it really depends again, but I would say in this setup that you would have the employees ⁓ that are dealing with the payments day-to-day on the normal path, like the path that’s always active. Maybe you want someone from the management or someone from finance to be co-signing that. ⁓

And then you can have the other layers where maybe the management should be enough by themselves to be recovering and then adding the board if something is really wrong. Potentially as well like a management conflict that could be in a small startup where you have two founders. What if the two founders are disagreeing and you need both to sign and they just don’t agree so they don’t want to sign? It’s good to have the board or the investors maybe having a kind of a backstop there as well.

Stephan Livera (19:19)
Interesting. And so as I’m reading you then it’s more like it’s moving away from just kind of a simple straight, like two or three or three or five. And it’s more like have different spending paths encoded using mini script, using, you know, the mini script compiler to create these different conditions and spending pathways, let’s say.

Kevin Loaec (19:41)
Yeah, absolutely. And it’s also easier from a kind of a no-date perspective to know exactly who can sign when instead of having quorums ⁓ because like two of three, we can say, you know, there is three groups of people that could sign. If you go three or five, it’s a lot more combination, et cetera, et cetera. So if you have insurance for your business, doesn’t have to be insurance on your Bitcoin, but just generally business insurance and things like that. They want to be able to map out things. If you have to deal with the regulator, they really want to know

who is in control of what. So you have a lot of different things that can be affected by just having a clear who can spend at which time.

Stephan Livera (20:20)
I see. And so to be clear, these Bitcoin protocol level spending pathways, they at least currently cannot like rate limit, right? So it’s not the amount, what we’re talking about here is just who can sign in ⁓ a given setup. So I guess just for people to understand the difference, because there are certain things that Bitcoin’s protocol L1 cannot enforce.

Kevin Loaec (20:40)
Correct.

Stephan Livera (20:46)
And so what we’re talking about here is just only the things that are possible to enforce that way.

Kevin Loaec (20:46)
correct.

Yes, at least in Liana right now. ⁓

There are ways to do weird rate limiting in Bitcoin, but that’s outside of the scope of Liana. So typically Revolt was able to do that. ⁓ But the easiest way to do actual rate limiting today is to do a co-signer. So that could be a specific machine, an HSM or a company that would just sign or co-sign if you are within the limits. And you might also have seen that ColdCard now has a ⁓ of a co-signing mode on their…

device that is now compatible with Miniscript. you could actually have your signing device itself refuse to sign if you go over a specific limit, which is quite interesting, right? It’s not on-chain, but we are pretty close to something you can’t bypass normally.

Stephan Livera (21:30)
enforce the conditions.

Yeah.

Interesting because I can imagine a lot of listeners like if they’re kind of coming from you know, maybe not as deep technical world They’re just kind of used to I don’t know a bank or some kind of You know interface that has its own rate limiting But these policies are enforced by the custodian or by the bank it not you know something done in You on chain in Bitcoin per se

Kevin Loaec (22:03)
Yeah, absolutely. ⁓ It’s just something that Bitcoin doesn’t have. We would need something like covenants and maybe some specific type of covenants to be able to enforce limits. It’s not something simple. One of the reasons as well is because the way Bitcoin works, we are spending UTXOs, we are spending coins, kind of like the coins in your pocket. They are not related to each other. They just happen to be in the same pocket. And this is what a Bitcoin wallet is. It’s a lot of coins that are not related to each other. They just happen to have

Stephan Livera (22:23)
Right.

Kevin Loaec (22:33)
the same spending conditions. That’s how we define a wallet. But from the blockchain perspective, ⁓ they are not linked to each other. yeah, we can’t really count. They don’t know each other exists. So it’s really hard to make sure there is a rate limiting there at the scale of a wallet where it’s different things not related to each other.

Stephan Livera (22:40)
Right, there’s no accounts, it’s a UTXO model.

Yeah.

So I guess following on from that, the natural way a lot of Bitcoin people do it is they segregate their wallets, right? So as an example, even for individuals, they may keep their life savings in the deep cold multi-sig that they rarely ever touch, maybe a warm setup for somewhat big but not really big amounts, and then a phone wallet just for day-to-day buying a coffee level of transaction.

So do you see a similar thing applying at the business level then that they, where they will have, because they can’t have rate limiting in the sense that we were talking about, they naturally just do this thing where they keep like, again, a deep cold multi-storage for the really long-term stuff and warmer setups.

Kevin Loaec (23:37)
It depends. ⁓ Currently we don’t see that actually often. So…

Either they use Liana as their cold storage and then they have like a spending wallet that could be an exchange or something because what they need is actually to sell Bitcoin to get fiat more than payments. ⁓ Or you have businesses that actually use their Bitcoin and they don’t necessarily segregate. So in this case, it’s still good to have multiple people because the people are also acting as these cosigners, right? So if I know and my employee know we’re not supposed to spend, I don’t know, one BTC ⁓ per month or something, ⁓

are both going to check the condition of the transactions we’re signing. So we do act as a backstop to each other. ⁓ So yeah, we don’t necessarily need to segregate everything, you know, at the protocol level in different wallets, we just can be humans and check actually the conditions we’re signing, which is the point of co-signing, it’s actually to check each other, right?

Stephan Livera (24:35)
Yep, gotcha. And so when it comes to recovery, then let’s talk a little bit about the recovery side of things. How do you set this up in a way where it’s secure and also not leaking early, this kind of thing.

Kevin Loaec (24:51)
Yeah, so the way recovery works in Liana is there is no automatic movement of funds, ⁓ On chain, like anything on chain is not happening automatically. It’s just like you still need people to actually sign transactions to trigger something. So the way Liana works is that you have these multiple paths. You have the normal one that can spend at any time. And then if the funds are not moving for a specific amount of time, the recovery path or the recovery keys are able to spend.

But it doesn’t mean the funds are going to a different wallet automatically or things like that. That’s not the case, right? So even in the case where a business would not move their funds for a specific amount of time, for some reason they forgot or they didn’t have access to the keys or something, ⁓ the recovery method is just becoming active, but it doesn’t mean they have to spend, right? So ⁓ as long as the primary one still recover their keys, they can still do a transaction and everything is just reset.

normal way. there is no, I mean I’m not sure what you meant by leak but at least there is no leak in this in this way where the funds could go away by themselves. ⁓ Now there is

Stephan Livera (26:00)
Yeah, I guess what I was

referring to is more just that the recovery pathways are not used early, right? And that’s again enforced on chain by Bitcoin’s protocol.

Kevin Loaec (26:07)
Yes, again, yes. On show.

Yes, so yeah, we do use something called relative time locks. So it’s every time, well, I’m simplifying a little bit, but it’s every time you do a transaction that would push this delay again. So in a typical disaster recovery situation, we might have a long time lock of like a year or something. And so we would expect the business to do more transaction than just once a year. ⁓ Simplifying a little bit, but this is how it works. So you don’t have to think too much about specific days or specific dates. It’s more like a regular use means that there is no

problem so the recovery is still moving forward to the future. Another thing that’s quite interesting here actually is that let’s say it’s a third party key or it’s a key that’s less secure maybe and that recovery key or keys are somewhat leaked. That could happen, Like a key is breached or an employee didn’t secure their key properly and that just happened to be the recovery key. That’s pretty good because it means, well, it’s not bad because it means that

In a normal case, these keys are not valid. So as long as the employee or the company notices, we can still migrate the funds to a different wallet with the primary bill.

Stephan Livera (27:18)
Right, and you would migrate out of that

setup into a new one because you know that if you stay on the current setup, you’re eventually going to be in a situation where that key is now valid and it could spend the coins, obviously.

Kevin Loaec (27:29)
Yes, and this is particularly good for custodians because currently custodians have access to the funds of all of their users. So if something leaks, they might actually leak all of the funds they are managing. But you could imagine that now these custodians just become a recovery key on their users’ setup. And so even in case of a breach of the custodian, ⁓ typically their users shouldn’t be in a situation where the time lock expires.

So if the user is actually normally using their wallet a breach of their recovery provider doesn’t mean anything too catastrophic for them They would just receive a notification saying hey, there was a problem. You should rotate your funds elsewhere I know in a different wallet and ⁓ and yeah, that’s that’s much better and that goes back to your first question about the centralization of risks ⁓ It would also lower a lot the incentive to attack such providers because you know getting access to their keys does

mean you get access to any funds so that’s you know that’s helping a lot on the attack defense.

Stephan Livera (28:32)
because they would only have the recovery key for like

a year from now or whatever. And while we’re on the whole recovery thing, now we’ve been talking a little bit about relative time locks and I can’t remember the exact, think we did on a prior podcast, I think you mentioned the limit is like a year and a half or a year and three months, something like that. And now are there discussions about longer recovery paths that are maybe encoded in an absolute way because we can’t do relative time locks for that long?

Kevin Loaec (28:37)
Absolutely.

Stephan Livera (29:02)
What’s the thinking on that?

Kevin Loaec (29:04)
Yeah, for Liana business, we are able to do this. We’re still not doing it on the normal free open source version of Liana. The reason for this is that we believe it will, well, users will make mistakes if we let them do this kind of stuff.

Stephan Livera (29:21)
They could like foot gun themselves and lock their coins for 20 years

or something or a hundred years or yeah.

Kevin Loaec (29:25)
multiple reasons, yes.

One of them is that the maximum time lock you can do in an absolute time lock is 9,500 years, so that’s quite a long time. Maybe. The other one is that it’s also absolute in the script, which means that you cannot

Stephan Livera (29:34)
Only Brian Johnson will be around for that.

Kevin Loaec (29:46)
push it in the future. You have to create a new wallet, technically. So that means new backups of your descriptor, new registration of your descriptor, all of that. And from a UX perspective, it’s not great. We also know that a lot of users and businesses are reusing addresses. They should not, but they do. And so what happened, you know, if you actually mistakenly send funds to your previous wallet, which has the time lock, the absolute time lock expired and you rotate it to a new one, and maybe you didn’t keep a backup of the old

Stephan Livera (30:12)
⁓ you see.

Kevin Loaec (30:16)
And so this is where it’s becoming tricky to manage such wallets. It’s not impossible. Some wallets are doing it, you know, kind of correctly like Nunchuck or Keeper, but it’s still something that I don’t feel a normal user should have access to. So that’s why we’re going to limit it to business users for now. I don’t want the average user that doesn’t have our support to be able to do these kind of things.

Stephan Livera (30:38)
Right, and they could easily make a mistake or yeah. So what would that look like in practice? So let’s say, Leanna, some business wants to sign up with you, they do their normal setup, you know, the normal stuff we’ve spoken about, but they want to have some kind of deep recovery key for five years out or 10 years out. Who would actually be the key holder in that scenario? Are they going to like, you know, a big custodian to be that key holder for the 10 years out case? Or how does that work? Or are you going to be the key holder for that?

Kevin Loaec (31:07)
It’s not going to be us because for now we’re still based in Europe and we cannot do anything with keys. ⁓ We might at some point open a subsidiary somewhere else and be able to offer these kind of services. That might make sense when we have enough demand. But yeah, for now it’s going to be third parties. The process though in Yenna business is a little bit different. The business is not going to do their setup themselves on their side. We will have a discussion with them. We prepare their kind of call it a template, but it’s just for them, right?

It’s a unique template. And then when they use their Liana business software, they will actually have a different onboarding flow that just asks them for their specific key. So each of the participants in the wallet is going to be registered with their email and name and things like that. So on each of their computer, it will only ask for their specific key. So it will ask them to plug in their hardware wallet, right? When I say key, it’s not your mnemonic, never enter that on the computer. And so the setup is really kind of like foolproof.

have to change anything, they don’t have to set up time locks, everything is going to be prepared for them and they will just ask you know to put their keys in the right place and then to confirm everything is correct. And so we are going to do these things even with the absolute time lock when needed on our side. So on their side, on the software side, they are not going to be modifying any of these settings so they know it’s going to be what we discussed with them. For the recovery keys itself, it could be them again so they can choose to have one of their

their own keys that they have maybe held with the board as we were discussing before, or held in, I don’t know, like a safe deposit box in a bank where they actually have kind of an old-school multi-sig where two of the founders need to go to be accessing this safe, things like that. ⁓ So yeah, it really depends. And if they need a third party, they can bring in their own third party. So that could be a custodian they deal with already. ⁓ Or we could kind of recommend them some that we know of. ⁓ But yeah, at this stage,

we can’t do it ourselves, it has to be third parties.

Stephan Livera (33:09)
I see. Do you see that becoming a common practice though? Like having, let’s say, mini script and like, you know, your normal stuff, but then also one deep recovery pathway for five years out or 10 years out just in case. And that is held with maybe some big custodian or some, you know, professional whose role it is to do that.

Kevin Loaec (33:30)
Yeah, possibly. I still personally prefer my relative time lock stuff, so I always prefer to be able to push it further, ⁓ but I don’t have a problem with very large or very long recovery path with a third party. The problem is more like, you know, if you lock your key for like 10 years, is your custodian still going to be there in 10 years? It’s starting to be a long time, right? Even for a trusted party. So these kind of things are a bit difficult to gauche and to figure out, you know, how

long is too long and yeah and how long is enough. I do think it’s going to be very common to have proper self-custody where only you can spend at any time. There is no cosigner, there is nothing else, there is no authorization to ask anyone else. It’s just you and maybe your team right when they say you is like the organization. But you would have a recovery option in case something go wrong and I really think that should be the nearly the default for normal users of Bitcoin because we know loss is a real risk and

Stephan Livera (34:02)
Yeah. Yeah.

Kevin Loaec (34:30)
it’s not fun when it happens. yeah, that’s a good way of doing it. And something we haven’t covered yet, and I don’t know if you want it to cover it, but it’s insurance. So of course, when we start talking about third party… ⁓

Stephan Livera (34:39)
Right, yeah.

Kevin Loaec (34:43)
What we were describing until now were basically a technical insurance. It’s not an insurance from a regulatory perspective, but you kind of know a third party can access your coins if something happens. But now we can talk about the actual insurance, where an insurance company in this specific setup can just insure the recovery ⁓ party. So if it’s a, I don’t know, a trusted custodian that’s here only as a recovery option, if they are covered with the insurance company actually saying, you know, if their key is lost,

will cover the user’s funds. That’s starting to be interesting as well because for the insurance company, the risk is really low. The amounts are not like crazy as if it was all of the funds of Coinbase, right? It’s only affecting this one user if something went wrong with that key. And for the user, it’s like nice because they know that no matter what happens, even if their third party disappear or lose their keys, there is an insurance coverage against loss, of course, right?

Stephan Livera (35:41)
Interesting. OK. ⁓

Yeah, so we covered a couple of things. So one other area, I guess we were touching on this lightly, but just to kind of spell it out. So let’s say in a scenario where there’s an organization where keys are lost or revoked, maybe it’s employee turnover. How do you maintain that recovery pathway without introducing a custodial risk? Is it just about having like another spending pathway that opens up to account for that fact that let’s say there was some turnover?

Kevin Loaec (36:07)
Yeah.

So yeah, forgot to mention that, but yeah, of course, you don’t have to have just one recovery path. You don’t have to have just one recovery option. So you can actually use them and abuse them ⁓ as you would in a normal multisig. So when we were talking about the two of three, a two of three is three different groups that can sign, right? A or B or B or C or A or C, right? And so depending on who is supposed to sign when on who is the normal signer and who is there kind of as a backup,

because it’s always something like this. When you have a two or three, it’s always like one of the keys kind of thought of as a backup. You can actually do that in your setup. So that could be key A and B as the primary. And then you could say, I don’t know, after like three weeks or months or something that seems okay and would still let you recover the funds fast enough, especially in a business environment, it would move to maybe A and C. And in case there is a problem with A after like six months, it goes to B and C, right?

do this kind of stuff. It doesn’t have to be the same key. You could also say something like, we have these two people, and B, that can sign now. And in case something go wrong, we have a different group or a different business unit that’s like ⁓ C and D that could sign, for example. Yeah, so we can have as many layers as you want. Yeah, yeah.

Stephan Livera (37:23)
That can be the backup, yeah, gotcha. Okay.

So zooming out a little bit, are there any threat vectors in Bitcoin custody that you think the ecosystem is undervaluing right now?

Kevin Loaec (37:38)
absolutely. ⁓ I really think it’s what we were talking about before. It’s the amount of money in very large custodials, all large exchanges, ⁓ namely Coinbase right now, because Coinbase is also dealing with most of the ETFs. And it’s just an insane amount of money. I don’t know how many millions of Bitcoin they are ⁓ custodying right now, but it’s an insane amount.

So what’s the risk there? I don’t even know. What happens if ⁓ something breached there? ⁓ Is it the entire amount? Is it segregated somehow in smaller groups of keys? I don’t know. How do you get access to this? Do you threaten people? Do ⁓ you have bad actors that are already working there and trying to get access ⁓ as employees? It’s very difficult. And so what would happen? Would the attacker actually dump

coins? That seems kind of unlikely but it’s possible. ⁓ Would there be some kind of government influence? Like what would happen? Because we know the ETFs are mainly pension funds at this stage so it would make sense that if all of this money is gone or locked or whatever, it could just be locked or burned. ⁓ It would kind of make sense that let’s say the US government actually start putting pressure to unlock this fund somehow.

the miners, maybe things like that. we have no clue how deep this could go ⁓ because of the amount of money. It’s just Coinbase is kind of like too big to fail, but Bitcoin is not designed for too big to fail. And so yeah, I think this is really the risk is the complete unknown of what would happen if such a large custodian would have issues. ⁓ Yeah, so I really think this is the threat. ⁓

We need to fix it somehow. That could be through something that actually is being worked on in the US, which is to have multiple custodians. So let’s say two or three of different custodians instead of just all your money in one custodian. So things like that might be helping, but it’s still a very scary risk to me.

Stephan Livera (39:47)
I see. ⁓

In terms of hardware key reuse. Now this is something you were talking about. I saw your bit of your talk in Bitcoin Japan, the dev day you were talking a little bit about, I guess this concept of using the same hardware key, but across different setups. And I guess you were saying, actually, people should be able to do that.

that you could have as an example. Let’s say I have a cold card that’s used in one of my personal setups, but it’s also used in like a family setup and it’s also used in some kind of business setup. But the question and the challenge is across the ecosystem, how is this coordinated in a way that is secure, redundant, ideally private, but maybe not, I don’t know. And let’s say compatible across different kinds of setups. So do you want to just discuss a little bit on that because I think that’s probably going to be interesting for listeners too.

Kevin Loaec (40:41)
Yeah, yeah. I think so one thing we know is that most users, most Bitcoiners have or had multiple wallets over time. They change software, they change, know, they go from a pay to script hash to pay to W script hash, like to witness script hash with SegWit. They might move to Taproot. It’s lot of different things, maybe wallets as well in terms of software, maybe single-sig, multi-sig. And so the traditional way of dealing with that was to

to kind of just generate new keys and generate new backup, generate new mnemonics. And you end up with like a list of mnemonics. I don’t know how many, but you might have a bunch. And that was just not practical. It’s also bad in terms of, yeah, you don’t know which is which. You don’t want to re-import all of them, et cetera, et cetera. So there have been a lot of progress to be able to reuse the same key across different wallets. That can be done with what we call the account number. So in your derivation path,

you

can choose a different account number. It’s just not user friendly, but anyway, it works. You also have things like BIP-85. So BIP-85 is that you keep one mnemonic, but from there you can generate different mnemonics. So all you need to keep is just one master backup, and from there you can create all of your other mnemonics that you had. So these kind of stuff are very useful, but my discussion in Japan was really about how do we make this in a way that the user, especially a non-technical user,

doesn’t have to think about it and cannot make a mistake. So the main risk is not really security on the, you know, it’s not really a problem security wise to reuse the same exact X pub across different wallets, but it’s terrible ⁓ from a privacy perspective because you are going to reuse also what we call public keys. And so every time you spend, you will reveal all these two wallets were connected and that’s just not great. So yeah, we need to make sure users use different account numbers, but we

don’t want to ask them every time they create a wallet, ⁓ which account number do you want to use? Because they don’t know what it is. They don’t know how many they used in the past, et cetera. And so, yeah, the whole discussion is that we’re trying to find a way to make it transparent for the user in a way that they just can’t reuse an XBub, but they would still use the same mnemonic if they want to. I’m not saying, and I think it’s an important point as well, I’m not saying users should just have one mnemonic and use it for every

I think they should be able to, but of course, I do still segregate my personal keys and my business keys, but I want to be able to use the same one in multiple wallets if I need to. So for the business, for example, we have different wallets and I’m using the same key ⁓ for the different ones. And if you look at the blockchain, you will never know it’s the same business or it’s the same key.

Stephan Livera (43:35)
Yeah, and actually there was an interesting point you made which people might not be familiar with it nowadays, which is that? Especially in earlier years. ⁓ maybe even now there’s this element of people could be reusing a key on a shit coin and Then that key could then come back and docks you on the Bitcoin chain if things hadn’t been coded correctly so can you just explain a bit of that dynamic for people because that could be a bit surprising or Not countering. It’s kind of counterintuitive for people

Kevin Loaec (44:04)
Yeah, yeah, yeah. So now we have this thing called, you know, derivation path. So HD wallets, etc. But anyways, technical terms, but you’re using one secret, which is like a master private key equivalent to your mnemonic. You enter it in your hardware wallet in your signing device. This key is actually going to already be derived. You’re not using the master one, you’re going to derive it to do some derivation. So some depth to generate what is going to be

used by your wallet. So in Bitcoin, we use a specific derivation path. There is also a coin type ⁓ field, which is like a different number depending on which shit coin you’re using. But again, that is just something implemented on the software perspective by the hardware wallet or by the software. But if you don’t do it, it will work. You can use your Bitcoin key to create a whatever wallet. And of course, if you do that, when you sign a transaction, could be also

reused on the Bitcoin network. The thing is that it would need to have the same kind of format, so the same UTXOs, etc. So the risk is only really there when it’s about a fork of Bitcoin, much more than a completely different shitcoin. ⁓ And so yeah, the risks are there when we’re talking about, you know, like Bitcoin Cash, for example. ⁓ It was kind of scary at the beginning when the discussions were there about like, ⁓ it’s going to be a fork, but we really need to make sure ⁓ users are not going to

move some funds on Bcash and then someone there could just broadcast the same transaction and move their funds as well. Absolutely. so typically, the attacks were kind of weird, but you could be an attacker like a merchant or a peer-to-peer exchange where you would get someone to sell you their Bcash because it was worthless. So they send their Bcash or their Bitcoin from the other chain to you.

Stephan Livera (45:36)
Right.

And that’s why they’re talking about replay protection,

Kevin Loaec (46:01)
But then you replay this transaction and you also get their actual Bitcoin. And so this is where the risk really was. So it’s something important we need to do when there is a fork or any kind of split. We need to make sure users can’t lose their coins like this. So sometimes it’s not something the user can do anything about. It’s really on the developer side of things to think about this fork risk.

Stephan Livera (46:24)
Right, yeah,

and so I guess maybe that could be, especially in the age where now people are doing AI and vibe coding, maybe someone doesn’t really know some of these nuances and they just try to one-shot, you know, vibe code an app and it doesn’t really pay attention to some of these nuances, especially if, you know, it would be one thing if it’s like just kind of spending wallet, whatever smaller amounts, but it’s like another thing altogether when it’s like, no, this is for like your hardware wallet, like serious security, serious money kind of stuff. So I guess just things that people have to…

sort of understand that’s why the security focus and the review and the technical competence is important, especially for these kinds of discussions and products.

Kevin Loaec (47:05)
Yeah, yeah. And another thing I would like to add on this. So all of these discussions about wallets is not just for individuals, right? It’s also because we’re talking about businesses. And so I don’t want to have to change my key, my mnemonic every time I have one of the employees that’s part of my multi-sig setup that move out of the business or that lost their key or things like that. I want to be able to keep my backup, my secure backup that’s somewhere in a place that I don’t want to access right now. Things like this, right? We really want to

make it as seamless as possible to rotate keys in a wallet when one of the key, maybe not even mine, is swapped out and we need a new one, ⁓ it’s important that my ex-pub also change, but I don’t necessarily want to have to create a new mnemonic for that. So this is really where the discussion is. ⁓ there are actually two wallets we mentioned already in this call that actually use this, well, we don’t have a proposal yet, but already have this issue, and that’s Nunchuck.

Keeper because they do use the absolute time locks. So they already ⁓ rotate wallets in the background for users when they reach the time lock limit. And so by rotating this, they need to change the xBub. And of course, they are not going to tell their users to generate a new mnemonic, et cetera. So they have to do it in the background. And yeah, that’s a real question. Like, could I import my Keeper wallet to Nunchuck? And currently the answer is not very clear because you don’t really have the same coordination or the same generation of xBub.

the ⁓ rotation because we don’t have a standout for it.

Stephan Livera (48:36)
Wow, yeah.

Right, and I guess it’s not an answer to say, but you have your output descriptor because what we’re talking about is the xpub that goes in that output descriptor. That’s the thing that’s changed. And as you said, so let me summarize as I’ve understood it, just make sure I’ve got it right. And then maybe also for listeners. So just in Bitcoin in general, what we’re protecting really is like this massive, massive, massive number. That’s your master private key that can be represented with

12 or 24 words, right, the typical BIP 39. But then what we’re talking about is that going from master private key to master public key that can shift based on account numbers and this derivation path that we’re talking about. And so what we’re talking about here is if the software has been coded in a different way where they’re maybe they’re incrementing the account number in that derivation path, it’s going to result in a different X pub. And then that X pub, you know,

because the idea is we would like to keep the same, let’s say hardware wallet with the same 12 words with the same master private key, but the X pub might’ve shifted for good, and it could be for good reasons. So it’s just like, there’s this technical and maybe social and economic coordination that has to happen there. ⁓ And yeah, maybe, I mean, I don’t know. I think what I got from you is that there’s not really a solution here, at least yet.

Maybe in the future, maybe there’ll be some kind of BIP to help coordinate this, but even that is a challenge across like, okay, so are people just gonna do like Google Cloud and Apple iCloud backup, but then lose the privacy, but at least make sure they haven’t screwed up from a redundancy point of view? Or is it more like, no, it needs to be even better than that, that’s not good enough because that’s gonna dox privacy? Open question, right?

Kevin Loaec (50:25)
Absolutely

and yeah loss of coins is always the know the bigger problem ⁓ But privacy is also pretty high and so keeping it like it is today is just not good enough for privacy So we need to do better without

Increasing the risk of loss and this is really the tricky thing. It’s like we know there is a problem privacy wise today Because all of the bips for let’s say multi-sig ⁓ if I already use my key in a multi-sig and I use it in a completely different multi-sig It will by default use the exact same xBug and this is a problem. We don’t want that and Yeah, it’s happening today So how do we make sure this does not happen? In a way that the user is not going to be stuck because I don’t remember which software I use

or things like that. We really don’t want that.

Stephan Livera (51:11)
Right. And so then in practice,

even if you’re an advanced user, it’s a pain because you’re having to keep track of, ⁓ which account number, like I’ve got the same master private key and the same 12 words, but actually which account number and which derivation parts have already been used for other setups so that they don’t cross contaminate with other new setups that you might do on that same hardware wallet and same 12 word seed or 24 word seed, whatever.

Kevin Loaec (51:36)
Exactly. Yeah, and you might also have to think that for the other participants. So one participant leaking their xBub on one side could actually, you know, kind of compromise the wallet from a privacy perspective. And so, yeah, you really want to make sure every user did their thing right and chose the right account number, et cetera, et cetera. And you might not really have control over their keys, over their setup. So that’s why the idea here of the whole discussion we had in Japan, and I’m going to have it at ⁓ BTC++ in two

weeks as well in Taipei ⁓ is like how do we make sure this is kind of automatic, completely foolproof and there is no way ⁓ the user can fuck this up.

Stephan Livera (52:17)
Yeah, I don’t know if there’s any easy way, but yeah, let’s see. So I guess zooming out a little bit back to just kind of self custody generally. Well, talk about Liana business. So why do you think Liana business succeeds where other efforts have not at getting businesses to self custody? Like, what is the USP, the unique selling point here with Liana business that you see?

Kevin Loaec (52:40)
Yeah, it’s the entire recovery option thing, Like having, knowing that no matter what happens, if you have a correct setup, you will not lose your coins. ⁓ This is really the main thing. If you ask people about why they don’t use self-custody, it’s usually because they are afraid they did something wrong. It’s not really about the physical threat or things like that. Of course, we need to mitigate that as well. But the main fear is that they don’t want one day to just be like, oops, I did something wrong and I can’t access my coins. So this is pretty much

solved with Liana in general. Now Liana business makes you know that you don’t have to think about how to do this properly. We can help you think. We can help design ⁓ it the best for your use case and make sure there is really no mistake. Sometimes it’s just about also having the convenience of knowing if you have any technical question, if you need any support, we are here for that. It’s about knowing that you can have an encrypted backup with us of your descriptor. It’s about knowing that you have very high

SLAs that your wallet will always be able to spend no matter what, no matter if CloudFlare goes offline or something like that. It’s a lot of things that we can help with. so USP for me is that it’s really built for businesses, ⁓ with businesses in mind. But the actual USP of Liana in general is like this entire thinking of recovery. So no matter what happens, you will not lose your coins.

Stephan Livera (54:04)
Yeah,

okay, gotcha. And so you kind of answered a little bit there, this next question, but around the future of corporate Bitcoin custody. What does that look like? And where do see Liana business fitting into that?

Kevin Loaec (54:16)
⁓ Regulation will have an impact on all of this. ⁓ I really hope the regulators stop thinking like everything should be a bank because I think in Bitcoin that’s not the case. It should be the opposite. should be like nothing should be in a bank. ⁓ In Bitcoin, really should be the control should never be ⁓ completely held by one third party. There is no way. So I believe this will change and we will at least force the custodial deployments to be multi-

of different entities and we need to make sure they’re not the same people controlling these entities. ⁓ Self-custody is great and having recovery with custodians ⁓ is also a very good idea. So I don’t think it’s about destroying the value proposition of a trusted third party. It’s just about making sure they can’t steal the coins ⁓ if things are going right. And also something that is maybe less discussed ⁓ is the role of banks for this. So I was talking about banks in the principle of like nobody should

be controlling all your coins and keeping them just for you. But traditional banks are also a thing and they want to have exposure to Bitcoin now. Bitcoin is becoming really a topic with Trump and ETFs and all of that. So their clients are coming to them and they are asking for Bitcoin products. ⁓ So what does a Bitcoin product look like for a bank? ⁓

Right now, kind of the easy way is just letting people trade some kind of exposure on Bitcoin. So ETFs is a great way to do that. So the bank doesn’t have to custody anything. It’s just a ticker. You know, you can buy and sell it and that’s some exposure to Bitcoin for their clients. But what about, you know, imagining again how we could position Liana for this. What about the bank letting you have your own self-custody wallet, maybe within their app. So you are in control, you have all of the keys. But if you lose your keys or if anything happens,

to you, the bank has a recovery key after a time lock. So that also fix kind of the inheritance problem from the traditional side of things where if you disappear, your family would get access to your Bitcoin through the bank, but after this time lock expires. And I really believe that’s a very good way for a bank to do Bitcoin the right way. So the bank would not be a custodian, but they would still be there to assist their users if the user lost their keys, basically.

Stephan Livera (56:36)
Interesting. Yeah, so it could be sort of like maybe in the future apps can actually be self custody But the user doesn’t necessarily have to even know as much about that But it’s actually the keys are there but I guess there will still be concerns around that like as an example Let’s say we’re dealing with like if it’s a high net worth customer Is he going to be comfortable having the keys to let’s say millions of dollars worth of Bitcoin, right? At today’s prices, let’s say 10 or more BTC on a phone. He might not be right

Kevin Loaec (57:05)
Correct. ⁓ We still have this discussion around how do we deal also with physical threat. ⁓ We need to make sure the user isn’t going to feel like they can fuck up really hard. ⁓ And so, maybe multisig is part of that. Maybe co-signing is part of that. Yeah.

Stephan Livera (57:15)
Yeah.

Yeah. Yeah. And, maybe

it’s like a threshold thing. Like the user is self-custodying, but actually above a certain threshold that the apps like pings you and says, Hey, get a hardware wallet to put this into that. You know, but, I guess bottom line though, it is going to be a kind of a battle there between convenience, right? Because convenience and security are almost there at ends with each other. And so people, there’ll be users who want easy ability to trade Bitcoin or borrow against Bitcoin, or maybe even the fiat interaction side of it. Now, of course,

Kevin Loaec (57:40)
Yeah.

Stephan Livera (57:48)
We’re all maxis, we want the hyper Bitcoinized future, but we live in the fiat world today at least, and it’s gonna be a while to get there. So there’ll be a lot of users who want this swapping functionality. Now maybe that can be built in, like maybe it’s Liquid or Arc or whatever, something that it can be sort of built in in a programmatic way that is a little more Bitcoin and cypherpunk-ish, even if it’s not perfect.

But I guess that’s the trade off, right? This convenience and security balance.

Kevin Loaec (58:19)
It’s extremely hard. It’s really, really hard from the technical perspective as well, because we don’t want to take this kind of compromise for people. We don’t want to reduce security. But at the same time, if we don’t do it, or if at least we don’t simplify things, we know they are just going to give up and go for a custodial app or a custodian or leave their funds on, I don’t know, Revolut or whatever. And this is kind of the things where the fight was

It’s like when, it’s not really us to decide, for us to decide what the users will use. We can just build tools and hopefully they will pick the correct ones, the ones that kind of made the right decisions for them. ⁓ But it’s extremely hard and it’s the same in every industry. Sometimes you just have to build.

what the customer wants and hopefully what you’ve built is still good enough. I don’t want to take shortcuts, some shortcuts for now. For example, I don’t believe in blind signers. I really want users to verify on a signing device that has a screen and buttons, their transaction. But of course the user prefer to just have a hotkey on their mobile phone. And it doesn’t mean a mobile phone is bad. Maybe a multi-sig of two mobile phones is better than one signing device where they don’t

read what’s on the screen and they just press next, next, next. But it’s a really hard thread to go through. It’s just like, yeah, it’s really hard. So we’re trying to find the solutions and at some point we will crack it. But yeah, we need to make sure they are not all going to ETFs and things like this where it’s just they don’t have any control.

Stephan Livera (59:51)
Yeah, okay.

Yeah, it’ll be interesting to see where things go. interesting and you know, I wish you well with what you’re building and hope more users do try to use self custody for at least like even if they do use custodial stuff to at least have most of their stuff in like the self custodial side and only put, you know, a smaller amount into the custodial platforms. That’s at least a healthier balance maybe. But yeah, listen, just check it out. It’s lianawallet.com that’s spelled L I A N A lianawallet.com and then

leonaworld.com slash business for the business side. Kevin, thanks for joining me and yeah, all the best.

Kevin Loaec (1:00:37)
Thanks for having me. Bye.