Hello everybody, welcome to the Juan Galt show. Today I have a very special guest. We have
Jameson Lopp in the house. He's a man that needs no introduction. He is an OG, a true
OG. He has created some of the most important cyberpunk self-custody works with Lopp.net.
He's got a huge set of resources on that website. And he's the co-founder of Cast Out Hodel,
which has certainly broken ground in terms of self-custody, multi-sig, user experience
and inheritance planning. So I'm very happy and proud to have Lopp on this episode of
the Juan Galt show with BitcoinNews.com and their support. And yeah, we get into a full
range of discussions about the whole spectrum of self-custody. We go from why you need it
at all and why you should do it and why not just stay on a Coinbase. I mean, BlackRock
thinks it's fine. Why shouldn't you? Right. And we go all the way to the edges of self-custody,
past multi-sig and inheritance solutions into what is being developed with Cashew and FettyMintz
in community and collaborative custody models, which is pretty interesting and complex. So
yeah, I mean, I love this show. I think it's going to be an epic. And I hope you enjoy
it too. Let us know what you think and see you in the next one.
Welcome, Jameson Lopp to the show. Thank you for coming on. I can't wait to hear from
you. I can't wait to get into some of these topics. And it's so fitting that we're going
to be talking about self-custody because at the same time as all this sort of upheaval
we've been talking about is going down, the European Union with Lagarde is talking about
doing a CBDC. There's all these stories about banks refusing cash withdrawals to customers
unless they give the bank a reason like why do you need the cash? Right. So, you know,
this is, you know, the CBDCs let these banks extend their reach and their control over
money so that they can micromanage the collapse and keep people from doing a run on the banks
and be able to punish their enemies and reward their friends with your money. So Bitcoin
really is the path out. But in order to use Bitcoin properly, we need to learn how to
self-custody. So Jameson, it's great to have you. Welcome to the show.
Happy to be here.
So Lopp, I'm a big fan of your work. I really appreciate a lot of what you've done. Lopp.net
is an incredible resource for anybody that's looking to, you know, take the deep dive on
the Bitcoin rabbit hole, right? Chase that orange rabbit down the, all this content and
all this, all this really interesting information and cyberpunk philosophy and all these tools
that we have available to us. So Lopp, you have a really great resource there. And you
also co-founded Casa, Casa Hodel, which is in my opinion, really breaking ground in terms
of self-custody. You know, you've solved some significant issues with, you know, the user
experience of self-custody, especially at higher levels of wealth. And you've even come
up with a fairly strong solution for inheritance, which is a very tricky problem in the Bitcoin
world. So I'm definitely looking forward to, to getting into that. Is there anything, anything
you want to, let's say announce or say before we kind of jump into some of the big questions
that we have for you today?
The big announcement for today is I'm now a wizard. Hooray.
Are you minting NFTs, Jameson?
No, I'm quite ambivalent about the whole NFT thing. Technically I have control of, I think,
three NFTs now. Two of them were gifted to me because they were made in my likeness.
I've only ever purchased one NFT. And that was actually something I'll probably write
a blog post about because it was an internet meme that was early 2000s, like pre-Bitcoin
meme. But the meme itself is kind of an NFT. So it's kind of a meta NFT, if you will.
Which NFT is that? What meme is it?
So there's this meme of the, it's a seven legged spider. And it's basically this email
thread back and forth where a guy gets an email from his utility company claiming that
he owes them money. And this was like in 2003 or 2004 or something. And he basically said,
here, let me pay you with this really crappy MS paint drawing of a spider that I made.
And it's a back and forth story.
Sounds like an NFT for sure.
People who are, I think, grew up on the early internet might recognize the meme. But I think
these days it's not as popular.
Right. I just had to Google it. I think I had seen it at some point. But yeah, that's
the kind of stuff that's going on. And let's start off with a story that was a bit controversial
and just to get it off of the table. Casa HODL recently announced that they're implementing
MultiSig for Ethereum. Right. It seemed like a lot of people had expectations that Casa
was a Bitcoin only or a Bitcoin first company. And so when they launched this announcement,
it kind of triggered an uproar of some sorts. So I have to ask you, Jameson Lopp, you know,
in the theme of seven legged spider NFTs, do you personally like Ethereum?
I mean, I'm still not that interested in Ethereum. I have used it for things. Like I said, I
guess I own two NFTs on Ethereum, but I only purchased one and the other was gifted to
me. You know, the vast majority of the functionality that's being built on it, it's just not interesting
to me because I'm not a trader. I'm not like a DGN person who's trying to find get rich
quick schemes. But you know, my personal opinion of it is kind of irrelevant when it comes
to business decisions. It's also not my first rodeo. I think a lot of people probably weren't
even around when I was actually working on not only Ethereum, but also Ripple and Bitcoin
Cash and Bitcoin Gold and all of these different alt coins back in the day at Bitco. When I
started at Bitco in 2015, it was Bitcoin only. And really it was a function of market demand
of either you decide that you only service Bitcoin stuff and you limit your addressable
market there or you try to expand and have more offerings to your clients. So the thing
that really becomes irksome at a business level is when you hear from your sales and
marketing people about all of the deals that you're losing because you don't offer all
of the services that your potential clients want. So that's basically what it comes down
to. Of course, people say, "Well, it's a slippery slope and now you're going to end up supporting
everything." This is, of course, a gray area of trying to decide which things will actually
have a positive return on investment for the company. There's obviously a very long tail
distribution of coins and tokens and stuff out there. And it certainly becomes untenable
at some point to be able to support them all because there is not only startup costs, but
ongoing maintenance costs, knowledge costs, so on and so forth. So moving from Bitcoin
to now also Ethereum, where there's a ton of activity and demand, it became a no-brainer
for us because we saw the volume of requests and lost deals continuing to increase over
the past few years. So I know it's difficult for some people to understand that there's
Casa the business and then the various decisions that have to go into that. And then there's
me personally and I am not the dictator of Casa. There's a lot more going into Casa than
just my personal opinions.
Right. And what can you tell us about the result of the decision from a revenue perspective?
Any insights you can share on that? Have a lot of people taken up Casa on this Ethereum
multisig, which I know you guys were working on back in the day, but there was no real
clear standard for multisig in Ethereum. So it seems like you guys have actually brought
a very strong product to market. How has the market reacted to that?
Right. Well, I think we only really launched it like a week or so ago. So definitely a
bit early to say, you know, ask me again in a year and we'll see. It was a year long effort
of research and development to actually get it to the point where we were pleased with
the quality, where we felt like we could offer a secure enough solution. And kind of to what
you mentioned, we technically supported, I think, single sig Ethereum for maybe the first
few months after Casa's launch and we ended up removing it. That was back in 2018 or so.
And one of the big reasons around that was just that there wasn't a sufficient standard
for multisig that we were confident in. We didn't want to write our own smart contracts.
In general, one of the things that we abide by is we don't want to roll our own crypto.
We don't want to do anything that's highly novel because there is, you know, viruses
and numerous strength in numbers that applies to many different aspects of security. Most
people think of that with Bitcoin in context of the cryptography and the math that is securing
the network. But it's also true for security models in general. If you're using a security
architecture that is implemented and used by many people and is securing large amounts
of value, then just like open source projects in general, the more people that are a part
of that project, the more eyes that are on it, the more scrutiny it has, then the less
likely that there's some hidden vulnerability that could be catastrophic. Point being that
we didn't want to roll our own smart contract. That was a process I had gone through at BitGo
back in 2016. It took us basically a year and three different third party audits of
that contract. Every time multiple critical issues would come up, it was a really scary
process. What have we done? Well, we continued observing the ecosystem and we saw that Gnosis
Safe and their contracts had developed a multi-year history of not having security vulnerabilities
and of securing billions, if not tens of billions of dollars worth of assets. So we felt like
implementing a solution on top of Gnosis Safe was sufficiently safe enough for our standards.
Gotcha. That's really interesting. And it really goes to show that Bitcoin has the best
wallets, it has the best security. Ethereum is so complex and there's so much action and
so much money in Ethereum, but the fact that they haven't had very strong multi-sig offerings
until fairly recently. I think the fact that you guys are jumping on this technology is
a good testament to that. But when I look at Bitcoin and compare it to all these other
coins, the quality of wallets is just not even comparable. Bitcoin has very strong architecture
and infrastructure on that front. With this multi-sig offering, do you see, are we going
to have more than just Ethereum on multi-sig going forward? With Kasa, are we going to
end up doing things like, I don't know, NFTs or maybe liquid level volcano bonds? Are we
going to just be self-custodying all kinds of stuff in multi-sig in the future?
There's no technical limitations. Gnosis Safe itself supports all ERC-20 tokens and probably
a number of the other token standards as well. It really becomes more of a question of, once
again, ROI, of what is worth our time and engineering effort to add support for and
then continue ongoing maintenance for. I'll say this is a function of demand and the by
far most demanded thing is stablecoins. I don't think that should be a surprise to anyone.
There of course has been some demand for NFTs, but not nearly as much as stablecoins. Since
technically the contract supports any tokens, you could use your multi-signature safe that
you launch through Kasa to manage anything that you want through the Gnosis Safe UI.
We have no control over that and that's by design. That's also how you would recover
your funds if Kasa went poof and ceased to exist because that's a core tenet of our architecture
is we don't want any single points of failure, including our own company. It's always possible
to route around Kasa to be able to access other tokens or functionality that we don't
support directly through our mobile app.
I also want to point out that I don't particularly like Ethereum either. I think it's mostly
Ponzi schemes and pyramid schemes. It's a very sketchy space. There's very few highly
valuable things that I see in it. Nevertheless, as somebody that also worked teaching people
and helping people secure their funds, I'd much rather that they hold their shitcoins
securely than that they don't. If you're going to have crypto at all, whatever that means,
it might as well be done under good security standards so that the day that you do decide
to buy Bitcoin and hold it long term, you have those skills to secure your property.
This is kind of why I'm kind of sympathetic to the decision by Kasa, not just business
wise, but I think adoption wise, people should be understanding multisig and really integrating
these systems into their lives. However, Ethereum has a lot of really weird scams, right? They
have a lot of sketchy, like fake token scams and other things that you don't really see
in Bitcoin because Bitcoin is not, let's say, as complex. Can you tell us, is there any
story that you can share? Have you guys had any weird issues with niche Ethereum scams
so far?
Well, it's definitely too early to have any stories for you, but it is something that
we're aware of. For example, one of the reasons why I would generally be against us allowing
anyone to add any arbitrary token they want, at least through the Kasa app, because like
you said, one of the many ways that people get scammed in this ecosystem is people create
a token that looks very similar, if not almost exactly like a real legitimate project and
then trick you. A lot of the stuff that does go on in the system ends up being various
types of phishing attacks or just general tricking the humans into interacting with
contracts in ways that aren't safe. So limiting what people can do has actually been another
core piece of Kasa's architecture for a long time. There's a lot of advanced functionality
that we consciously don't offer for Bitcoin because we believe that the potential value
that can be gained is actually offset by the risks that a less advanced user might shoot
themselves in the foot by applying something incorrectly. So that same thing is going to
happen across the board when we're adding any sort of new feature. So trying to save
people from themselves is a major part of what we're doing in our offering. I'm sure
that people are going to get scammed at some point. You can only help people so much when
it comes to preventing them from sending their money in places that they shouldn't. But there
are certainly some low-hanging fruit for stuff that we can try to keep out of the platform.
And kind of going back to what you were saying earlier, it's not just the business decision,
but even from a higher level perspective, one of the things that has really pained me
greatly is when I see people who have, you could call it a diverse portfolio or people
who own a lot of different altcoins, end up deciding to keep all of their assets with
a trusted third party instead of implementing self-custody because there aren't self-custody
solutions that basically will support everything in their portfolio. So it's so much easier
to just leave your funds on an exchange that already supports everything that you've bought
than it is to possibly have to go out and find multiple different wallets to self-custody
tokens for different networks and so on and so forth. So that's another thing that we're
trying to push back against. And like I said, the thing that has pained me is when I see
people deciding not to self-custody their Bitcoin because they also own these other
tokens and they obviously for simplicity want to keep everything managed in the same place
if possible. And so if that desire for convenience leads people to choosing third party custody
over self-custody, that's when I start getting upset because this trade-off between convenience
and security is basically the battle that I've been fighting for the past decade.
Yeah, no, totally. I mean, I think that makes sense. And that's kind of why I'm a supporter
of this effort on CAS's front because, yeah, I mean, what I see in the ground is a lot
of people are in Ethereum, a lot of people like to play with these tokens and we might
as well teach them how to take self-custody so they're ready when they need to actually
put on some more, let's say, risk controlled and secure assets on them.
Now I also have to ask you this. We have these stories about the El Salvador bonds, volcano
bonds, I believe a billion dollars were raised. There's thoughts about putting them on liquid.
I'm not exactly sure what the cutting edge of that is right now, but is there any efforts
or plans to support securities of these sorts or bonds in CASA? Are you guys looking at
that side of the market and considering launching, integrating that?
As far as I know, we have had zero requests for that stuff. So I think that kind of answers
your question. We go by market demand. Now, of course, since those bonds haven't launched
yet, that would also keep the demand pretty low. But we're not against it. One of the
many things that I've been canceled for over the years was actually being involved in a
security token. And so I think it should be clear to people who have been paying attention
that I'm not anti-STO either. I think that this is something that I kind of grapple with
from an anarcho-capitalistic standpoint of, on one hand, I think that people should be
free to basically do whatever they want and issue whatever tokens they want and deal with
the consequences of the market. But on the other hand, we have seen that there are certainly
a lot of problems with the fact just that the vast majority of tokens that get issued,
they're not very well defined. Essentially, they provide no rights to token holders or
they don't even really try to describe very well what they are offering to the token holders.
One of the big issues around that, of course, is often with sort of token distribution and
lockups, tokenomics, if you will, that are often not very transparent and in many cases
gamed to favor the issuers and often done in ways that are kind of opaque and so that
people don't necessarily know what they're buying. And then, of course, there's also
governance issues that it may not be clear that even if the attributes of the token look
good right now, that they may be changed in ways that you don't anticipate. So there's
a reason why securities law exists and why there's so many countless pages of regulations
that basically describe how you should be managing a company that has equity that is
privately or publicly traded. And I think there's certainly some lessons to be learned
that could be pulled in related to that and just tokens in general and giving people a
better idea of what the hell it is that they're actually buying. So that's kind of why I'm
in favor of supporting securities tokens, because while you may be against all the regulatory
stuff, it can certainly be a pain. At the very least, you know, a security token is
following a lot more best practices when it comes to actually giving or describing the
rights of the token holder and what they can expect.
Yeah, yeah. From an anecdotal perspective, in theory, everybody should be free to issue
their coins. But on the other hand, fraud is also a violation of the NAP, right? So
we always see a ton of fraud on this side of the space, not just in Bitcoin and crypto,
but also like pre-crypto with all these Ponzi schemes that have been very popular before
this era. But yeah, no, I mean, I think I like the idea of being able to hold a bond
like an Apple bond or a Tesla bond in paper in self-custody and not take, let's say, the
T, you know, like the risks of like fake bonds on the market that have been reported by,
I think, the CEO of Overstock sort of had some big stories about these sort of revealing
that there's a ton of fraud and fake bonds or fake securities in these brokerages, right?
That they weren't necessarily synced. And so I'd rather just have my own and make sure
that I own it, right? And that's kind of a cool aspect. So hopefully we can see something
like that in the future.
Yeah, I mean, these things, obviously, you're not going to be able to get the same type
of security model that you'd get with a completely permissionless token. You know, they're always
going to have a central issuer, but you can at least make some improvements on the existing
system by having that level of transparency, you know, the auditability aspect.
Yeah, it does seem like the system could be improved. I mean, again, people are used to
using Tether now, you know, they use Tether all the time. It functions like a physical
asset in the sense that, or like a self-cospited asset, right? You have it on your wallet,
you can send it, you can receive it. But it also has this layer of administration that's
kind of invisible. And it's certainly centralized and managed by a third party, you know, at
the end of the day, right? A kind of federal bank, right? With Tether. And so I think people
are already in a way used to what this technology can do to improve the experience of that.
It's just a matter of like, you know, implementing actual securities, like integrating this technology
with securities laws, which, you know, maybe El Salvador will be the one that leads that.
We'll see how that plays out.
Yeah, I mean, that is actually probably my highest volume use of the Ethereum network
is Tether and USDC. And, you know, sure, you know, they are centralized issuers. I wouldn't
recommend using them to store long term value. But I tell you, I would highly prefer and
I do highly prefer using these Ethereum based stable coins over having to go to my bank
and do wire transfers any day of the week. Like it saves me so much time. You know, sending
a stable coin, you know, basically takes less than 60 seconds. Whereas if I'm going to go
do a wire transfer, I have to fill out so much paperwork and often have to get on a
phone call with people and, you know, jump through so many hoops of security bullshit.
You know, it can easily take me half an hour to an hour, depending on how complicated the
wire transfer is. So, you know, from a utilitarian standpoint, I am certainly a fan of stable
coins as, you know, a medium of exchange.
Yeah, that's the crypto dollar. The stable coin is a topic we've talked about here before.
I agree. They're very useful and they're going to continue to be quite useful going forward.
I would love to keep talking about that, but I think there's so many self custody topics
that we definitely need to dig into that. I kind of want to bring it back to that and
maybe bring it back to Bitcoin and ask you just like a very general big picture question,
which is, you know, why should people take self custody at all? You know, why not just
trust Coinbase like BlackRock is doing and Fidelity is going to be doing, right? Why
not just put your money in Coinbase and avoid all the hustle of self custody?
Well, really, it's because you're giving up some of the most valuable properties of this
ecosystem. If everybody is using a trusted third party to manage their money, they no
longer have permissionless money. They're just going back to a banking model and the
bank may arbitrarily decide that they don't want you to use that money anymore. So there's
no shortage of examples of people posting on social media showing that they've gotten
locked out of their exchange account or whatever other custodian they're using. And sometimes
it's because the custodian didn't like some aspect of a transaction they were trying to
send or receive. Sometimes it's because of some other, you know, AML KYC issue that got
flagged. Other times it could just be, you know, some suspicious activity of, you know,
trying to log into your account, you know, through Tor or VPN or who knows what. So there's
innumerable things that can go wrong that may not even be your fault that can get you
locked out of these third party accounts. And the whole point of, you know, Bitcoin
is that as long as you abide by the rules of the protocol, you know, there is no authority
that can say no to you. And so, you know, the only way that you can get into a sovereign
position where you're not asking permission to use your own money is to actually have
control of those private keys. And this is why Casa exists is because we believe that
your private keys are an empowering technology and it's going to be more than just Bitcoin
and more than just assets. You know, we believe that private keys are going to give people
more power and control over their own identity and how they interact with the world in a
variety of different ways over the long run. This is, there's been a promise of cryptography
helping people, of course, for decades now. And this is what the cypherpunk movement was
all about. It's taken us a long time to get here. And I think it's still going to be a
long time before we really get to the point where this is a mainstream thing. But, you
know, that goal is certainly in sight. But once again, coming back to, you know, convenience
versus security, part of the problem is that the default for most people coming into this
ecosystem is that they're basically getting onboarded through trusted third parties. And
you know, defaults are very sticky. So when someone comes in through a trusted third party,
you know, they're probably using a web app or mobile app or something. And it probably
seems like almost any other banking experience that they're used to. And they're probably
okay with that. But in order to prevent the system as a whole from becoming more centralized
as we onboard billions of people, we need to get more people into self custody. So,
how do we do that? I really think that the most obvious way and not necessarily the easiest,
but we basically have to route around these third parties. Like I'm not particularly bullish
on being able to convince people to withdraw from trusted third parties. Rather, I think
we need to be setting people up so that the default is self custody. And an example of
that is basically when people are receiving Bitcoin directly as a result of some good
or service that they're selling. Like they're working and they're receiving that Bitcoin
directly. I think that when that is how people are getting onboarded to the ecosystem, it's
much more likely that they will then immediately be in a position of self custody. But you
know, there's obviously a lot of questions around exactly how we optimize that.
Yeah. And shout out to Bull Bitcoin in Canada, who's doing great work on this front. You
know, you send them money, they send you Bitcoin, and there's very little time in which you're
trusting them with your money. And they, by default, kind of force you to give them a
Bitcoin address. And thus, you have to set up your wallet. They educate you on how to
set up your wallet and they have a good amount of work on that front. So I think that that
kind of model or the model that Swan was following to some degree where you could actually do
your DCA out into your wallet. I do also believe that that's the future of exchanges. And it's
you know, as we see more of these custodians fall, you know, and as we see companies like
FTX try to do a 2.0 scam, you know, make sure they can try and get more of your Bitcoin.
I think we're going to see, hopefully, we should see more of these kind of companies
show up and take the main stage.
Now, OK, so people, you know, but today, you know, on average, people will join something
like Coinbase or get on Kraken or go to Binance and they'll buy their Bitcoin and then they'll
there. Maybe somebody will convince them to get their money out and they realize they
should. And so the first thing they'll do is usually they'll download a mobile app,
right, and they'll get 12 words, right. And they'll, you know, hopefully the Apple warn
them not to take a screenshot of the 12 words, to write it on paper, to, you know, stash
it underneath their mattress. Is there any other, let's say, pro tips for first, like,
let's say, new users or early users that are not maybe ready for multi-sig, but they do
want to take self-custody other than not taking a picture or did not digitizing their 12 words.
What else might you recommend?
Well, the most important thing to do when you're setting up, especially your first wallet,
is to actually test the recovery aspect of it. So I think what goes wrong in a number
of cases, and I've helped a number of people with recovery situations, though not always
successfully, is that, you know, they may set up that wallet and they may write down
those words, but they don't double check and triple check and make sure that the words
are actually correct or that the, you know, the wallet can be reconstituted. So, you know,
when you're, especially dealing with any non-trivial amount of value, if you think you're going
to be putting more than pocket money into a wallet, when you create that wallet, you
should make a small deposit into it and then just literally delete it. Delete everything
about that wallet and then recreate it and, you know, import that seed phrase and make
sure that the money shows up. Like that step is something that I think, I don't think
I've ever seen a wallet that actually tells you to do that. And that is, you know, that's
a foot gun that a lot of people have run into because, you know, if you write down the words
wrong or, you know, some people try to get cute and creative with how they actually write
down the words and do the backup and whatever and end up screwing something up. You know,
it's just important to have that sanity check that your data is actually going to work when
you need it.
Yeah. You know, there's a lot of theoretical ways in which, let's say when you're more
paranoid, like I think you and I are, there's a lot of ways in which you can think, like
you can come up with ways things could fail. But I think empirically speaking, you know,
it sounds like this is one where people actually fall, where they maybe, you know, again, they
get cute, they write the words, but they don't define the order in which the words are placed.
And so then they have 12 words, but no idea what order they should go in. That's one that
I've seen. Or where, yeah, maybe they just misspell a word and now you have 12 words,
one of which could be wrong. So you got to go into some brute forcing and like check
them against the dictionary and so on. So yeah, that sounds like a good tip. But I assume
that you're referring to testing the seed against a mobile phone or do you suggest this
be done on desktop wallets as well? Because there's some different thread models there,
right? Like I believe, let's say key loggers, which is one of the big scary, you know, cybersecurity
threads, like a key logger is more likely to infect a desktop computer than a mobile
phone, right?
Absolutely. I mean, the optimal scenario would be, of course, that you're using a dedicated
hardware device and you're wiping that device and then inputting the seed phrase into it.
But you know, if you haven't invested that 50 or $100 or whatever for a dedicated hardware
device, then there is a question of attack surface. And you know, mobile phone tends
to be safer. The mobile operating systems are a bit more clamped down and are less likely
to have key loggers and malware and so on and so forth. It's not impossible, but you
know, if your phone isn't rooted, it's a lot less likely. So it's certainly, I would say,
safer to be tapping or typing your seed phrase into a mobile app than it is to be doing it
on a laptop or desktop where it's much larger attack surface, much easier for there to be
some sort of malicious software running.
Absolutely. And when it comes to mobile wallets, you know, before we get into hardware wallets,
which is also a very, very important topic, there's this question of open source versus
closed source and or even just source availability. There's a company that I'm a big fan of called
WalletScrutiny.com. It's a project that ranks mobile wallets, Bitcoin wallets in particular,
and checks whether their source is available and deterministically reproducible. That means
that you can download the software from, let's say, GitHub, compile the APK file that you
would install on your phone, and that APK file would be the same file that you would
get from Google, which would be, it seems like the minimum standard that the industry
needs to reach such that they can actually benefit from open source. What are your thoughts
on these? I know that CAS is still closed source, though I believe that it's not obfuscated.
So it's like if you open up the wrapper of the APK, you can see the code and it's not
like intentionally hidden or anything like that. But what are your thoughts on this aspect
of the industry?
Right. I mean, there are very few companies that go to that level. I don't know. I'm sure
it's single digit percentage of wallets out there that have reproducible builds. And it's
better than nothing. It's not panacea. I think an issue with open source is, once again,
it's a network based problem of how many eyes are actually on that project. So one thing
that I've found really interesting in the space more recently, especially, is that there's
a number of wallets out there that are literally one developer. Even if it's open source, if
there is only one developer behind the project, there's no peer review. That one person is
deciding all of the code that goes into it. So once again, there's a question of, are
there any other eyes that are actually on that project checking to make sure that there's
no malicious code going into it? Because if you have a wallet project that's only run
by one developer, sure, they could be doing the reproducible builds and so on and so forth.
But if nobody's looking at the source code, it doesn't really matter. So it's a complicated
issue. And I think very, very few people bother to dig into the GitHub repositories and actually
look at the different attributes of these projects. So Kasa, our mobile apps are not
open source. Our kind of philosophy behind this is that we build on open source and we
make it very easy for people to, instead of try to reproduce and verify our software,
what they can do is actually use a number of different other open source software projects,
recreate their wallet on there, and in a much more user-friendly fashion, be able to verify
that the Kasa software is doing exactly the same thing that all of the other open source
multi-sig software is doing. We encourage people to do that, set up a watch-only wallet
essentially, because you get a similar level of integrity assurances. Like if we were trying
to screw around, for example, with any of the address generation stuff, then that would
become very immediately apparent because you'd get different addresses in the different wallets.
Also it's a tenet of our design that we don't have control over aspects of the transactions
that are created. So when a Kasa user is actually signing transactions, they are doing that
on software and hardware that Kasa has no control over. This is kind of a pain for us
in terms of managing the user experience and the compatibility and making sure that everything
works, especially over long periods of time because it's a system with many different
moving parts, but it takes power out away from Kasa where we don't have the ability
to screw around with the user's verification of their transaction details. Since we don't
have control over cold card or trezor or ledger or whatever, that's another important aspect
of verification that I think it's far more user friendly for someone to be verifying
transaction details on a hardware device screen than it is to be asking them to look at code.
Right. Yeah. And just to be clear, with the multi-seat side of Kasa, you're plugging
in your hardware wallet, you're getting an email with a link that lets you basically
connect through let's say trezor connect or through a kind of interface for cold card,
and then you're compiling these transactions with the hardware wallet, which means the
seed words or the private key is not leaving the hardware device. It's just like a signed
message that's being transmitted to a computer. So yeah, you guys, Kasa in a sense, never
has any kind of choke point or control point over that transaction construction.
But okay, so just to kind of finish up on that question of open source, because yeah,
you're right, like a lot of open source projects are basically abandoned. In fact, I've been
hearing that Coinomi is essentially abandoned. And that's alarming because Coinomi is a very
popular mobile wallet in the crypto space. It was supposed to be open source for a while
and then it went closed source. So anyway, just watch out for that. But do you, at what
point, like it seems like open source is an essential aspect of self custody, nevertheless.
And again, you've sort of recognized that, because a lot of these wallets, like even
if they give you the 12 words, you don't know if they're just keeping a copy, right? And
I understand that that can't happen with the multi-seat offering of Kasa, but it certainly
could with the single SIG. And it's probably happening with a lot of these wallets that
are single SIG and closed source that are fairly popular. I mean, do you think that
this is a direction that the industry should move into? It seems like it's a very tricky
thing, you know, like Samurai Wallet finally implemented deterministic bills after years
of people pestering them about it. You're right, only about five wallets in this space
actually managed to meet this bar. But like without that bar, you know, you are essentially
trusting the single SIG wallet provider, right?
Yeah. Even with the open source and verifiability aspects, like it's still possible for stuff
to go wrong. I mean, you know, Bitcoin Core is the gold standard, I think, you know, when
it comes to, you know, open source, high security projects. And even they have had vulnerabilities
from time to time, even having dozens, if not hundreds of eyes on the project, things
can go wrong. So, you know, part of my problem, I would say with using a single signature
wallet, whether it's on the mobile phone or desktop or whatever, is that, you know, even
if you're maintaining great security 99.9% of the time, it's still possible that some
sort of bug or vulnerability can get introduced that can then cause, you know, keys to leak
or can cause transactions to be malformed or, you know, created maliciously or so on
and so forth. So, you know, it is a single point of failure by default, you know, regardless
of whether or not it's open source or closed source or whatever. It's just as I've kind
of been trying to allude to, it's very, very difficult to quantify what the like level
of security that something being open source actually provides. And that's because, you
know, the security is basically a function of the quality and quantity of developers
that are actively working on that project.
Okay but I mean, it is nevertheless preferable to closed source. I think because closed source
you're just sort of trusting the third party. And again, this doesn't apply necessarily
to the multi-sig side of things with Casa because of the variety of tools you guys support
in Harbor wallets.
But I guess, and this will probably be controversial, but I would actually say that Ledger, for
example, closed source project is probably far more secure than many of the single developer
open source projects. And the reason why I say that is because Ledger has a ton of employees.
They have a lot of engineers and a lot of eyes on their own code. Now it's true the
rest of the world cannot audit and verify that code, but compare that to a single developer
project that nobody's even looking at is what I'm saying. That's why it can be difficult
to quantify exactly what the level of security of these things is in terms of the sort of
integrity and quality of the code itself.
Yeah, no, that makes sense. I think that's a good argument. I've never really heard that
argument, but yeah, you're right. So single sig mobile wallets are, they're arguably better
than a sketchy exchange, but especially done right. And again, of the ones like, I think
people should check out this website, walletscrudeny.com, which actually goes through that, but you're
right. Like if there's one developer on that wallet, then how secure is it? Is anybody
looking at it? So let's go into the next step of security or custody. Maybe somebody has
had their coin in this one of these wallets and they're listening to this and they're
like, okay, I guess I got to move to a hardware wallet. And we're looking at this story of
Ledger, right? Which just had this, the most ridiculous sort of blowing up of their brand
that I've seen in recent years, right? Where they, first of all, they started doing this
marketing where it was like, hey, you just put your Ledger on your chain, a literal chain
on your neck. And that meme with the Indian guy with a huge flavor flake clock of Bitcoin
gold, like 50 ounces of gold, right? And so that was ridiculous. But then also they implemented
this thing where basically they could withdraw key material from the Ledger through an internet
connection and set up like this sort of charted backup thing with three other companies. What
are your thoughts on Ledger specifically? And then maybe how to choose hardware wallets
more broadly? Like how do we make this decision?
Yeah, I mean, I think it's pretty clear that Ledger's various faux pas have been a function
of them trying to push more mainstream. And I actually, I understand why they've gone
down the path of this particular seed recovery solution, because, you know, they're not trying
to create a security model for someone who has, you know, millions of dollars that they
need to secure in a sovereign way. They're specifically, you know, this is targeting
the sort of end of the bell curve of the sort of the laggard mainstream adopters who may
have 50 or maybe a few hundred dollars, maybe even a few thousand dollars worth of assets.
And people who aren't dedicating a lot of time and effort into their security, because
really, if you have dedicated the 50 or $100 to buy a hardware device to manage your keys,
you're already probably better off than 99% of people in the space, having taken self
custody and taken those keys off of internet connected devices. At that point, the biggest
threat to your keys is actually yourself. And it's basically just screwing up and losing
access to those private keys. So from that perspective, I think it makes sense that Ledger
has prioritized creating a user friendly recovery solution, because, you know, quite honestly,
people suck at backing up data. It doesn't matter what data it is. It doesn't even matter
how technical you are. Like a lot of technical people lose data all the time, because it's
just a boring IT administrative task to do data backups, and especially to check on them
regularly to make sure that they still actually work. So, you know, while there's certainly
issues, I think, with how they did it, like it's not nation state resistant by any means.
I think it is better than not having a recovery option for a lot of those people for those
those target audience, those people who they're not necessarily securing their life savings.
So I think that it will come in handy, it probably will save a number of people from
losing access to their private keys. It's not something that I would use myself. But
we'll see. I think their biggest screw up was really around the marketing of it. And then,
you know, you can argue the minutia of whether or not they should have put more effort into
or done a better job of selecting the actual key custodians for that stuff. But, you know,
they published a white paper, I think, in the past week, and I looked at it. And, you know,
from a technical perspective, it looks pretty solid to me, it is like the quality and caliber
of work that I've come to expect from ledger. Okay, and do you believe that? Does this open
up a door for hackers to maybe like, you know, like, is this a kind of backdoor? Or because,
you know, if you can actually remotely with the remove or make a copy of data, even if it's well
sharded, even if it's like, does this not provide a path for, let's say, hackers to, you know,
backwards engineer, reverse engineer ledgers, you know, door, you know, remote door in?
I don't know, I guess that's my concern. Like, I feel like I can't, again, like, and I'm probably
on that side of the bell curve of like, you know, way too paranoid to give that kind of power to a
hardware device. But is that a concern that you think they've addressed sufficiently?
I think the well, the main problem in this case is the fact that we can't verify, you know, the
code that is actually running on the ledger device. So like I said, I read their white paper,
and I guess the caveat is, if the code running on the ledger is doing what their white paper
details as the the architecture being, then, you know, it's a pretty secure setup. But
kind of what you're getting at is, if there are any code issues, that introduce some sort
of vulnerability where that security architecture can be bypassed to extract the key material,
then of course, yes, that's a problematic backdoor. So it is possible that, you know,
these changes could introduce a backdoor, we don't really have a way of verifying that this is,
this is the trust model that you are getting into, you know, when you buy a ledger, and it has always
been the trust model that you're getting into when you buy a ledger, even before they have
offered this backup functionality. The only thing that I can tell you, like from a verifiability
perspective, is that there are there are no API calls, at least that we know of that we can find
that, you know, that's possible for us as developers to make to the ledger device to request
the extraction of a seed phrase. That doesn't mean that, you know, there's, there's not any
way that you could try to glitch the API or the device to try to extract that. The only thing
that we know for sure is that no one has succeeded in doing it so far. And, you know, ledger has been
around for a number of years, and they have, you know, they've withstood the test of time, they,
I would say, have one of if not the most proficient team is when it comes to hardware
security, you know, ledger, Don, john breaks a lot of other companies devices in this space.
And as far as I'm aware, nobody has broken theirs.
Right? Yeah, I mean, that's, that's very fair. You know, they do, they do very, very deep work on,
on trying to break other hardware wallets. You know, I think that they also announced after this
controversy that they were going to be open sourcing or like fast tracking their path towards
open source. And they're, they're a big enough company, they could probably actually draw enough
eyeballs for that to be, you know, properly peer reviewed. So I do hope that they take that path,
you know, like, I can't recommend ledger going forward until the ICC significant changes on that
front. But you're right, they do seem to have a very strong team. So hopefully, they'll turn
the ship around. And so on that topic of hardware wallets, how do you suggest, let's say the average
person that's ready, like to move to the next level of security, let's say they're not the
they're not, you know, let's say that the boomer crypto friendly class, that is okay, that will
buy a ledger so that they can play with crypto, but also, they will KYC themselves into ledger to
make sure their backup is secure, right? Like, for people that are not that user, right? How do you
go about choosing a hardware wallet that let's say, takes you to the next level of security,
you know, because the options that we have left are, you know, obviously,
Tresor, I think Tresor, Tresor Model T, and then you have the cold card
wallets. And then there's like six other ones that are kind of like, you know, not that popular. I
think BitByPit is one of them. I mean, no, it's not by me. That's an exchange. But anyway, there's
a there's like a 10 more hardware wallets in the market. How do you think about those? And how
should people think about those hardware wallets? Yeah, well, I think there's way more than 10.
It just depends on how far out on the long tail you're willing to go. I think there's definitely
a lot of like really low quality, crappy Chinese products out there that have not been very well
tested. And in many cases, they're basically just old Android phones, I think, probably not even
running up to date Android software. So you're going back to that recurring theme that one of
the important aspects of security is safety and numbers, you don't want to go too far out on the
curve of adoption, it is definitely safer to stick with one of these devices that has been sold at a
high volume of level and therefore has been subject to more scrutiny than other devices. So
I wouldn't, you know, go outside of the big name brands. Then after that, there's kind of the
question of usability and, you know, how, how much effort or how technical are you as a user? And
what do you need from a UX perspective, I do lean towards preferring people using devices that don't
require USB cables. So for example, you could use cold card with as NFC, that is pretty convenient,
though there's not a ton of wallet software that supports it yet. It's on our to do list. Certainly
to add NFC support, we already support cold card. Just from a like raw, you know, PSBT file transfer
perspective. I'm also a fan of really any of the devices that support animated QR code standard for
transferring. So I know Blockstream Jade, Passport, Spectre DIY, though almost nobody has that.
And really, you know, that's the reason why I say that is because what we've seen from a support
perspective is that when you're when you're having to plug devices into like laptops and desktops,
it's just a much more complicated software stack where there's a lot more things that can go wrong.
And basically, cause the user experience to have more friction. Whereas if you can literally just
tap your device or point your device at your screen and do that data transfer, there's
far fewer things that can go wrong just from a user experience perspective.
Very cool. Yeah, I haven't played with these, these Passport special QR codes,
but I'm going to definitely check that out. And I do have a Jade that I bought
ages ago that I'm going to I'm going to plug in and start playing with Jade is one of these
companies or one of these products that has been kind of rising in the ranks, right? I mean,
for a long time, Tresor had the lead and then cold car really kind of I think took the lead
as far as like the most paranoid Bitcoin Harbor wallet. And then Ledger has always been in the
game, more crypto friendly sort of wallet. But now Jade is in the game in the kind of like Bitcoin
only, but plus liquid side of the market. What do you what do you think about Jade as a hard
wallet specifically? I don't I haven't heard too much conversation about it, but it does seem to be
firmly in the top tier at this point. Yeah, I mean, it's user friendly. I would say similar
to Tresor though they do have the camera, which is a plus. I'll say I was I was definitely
disappointed with Ledger stacks, which I think they have they aren't even shipping yet. But
I was sad that Ledger stacks didn't even they didn't bother to put a camera into that device.
I was like, you know, you really you're losing out on better user experience with that cold card.
Their Q1, I believe, which is supposed to start shipping by the end of the year is also going to
have a camera, though I know that they're not a fan of the QR code standard that all of these
other devices are following. So we'll see how that goes. They're going to want to implement
their own could be tricky, I think, to get a lot of other wallet software to add support for
whatever they come up with, unless it's significantly better. I kind of yeah, I kind
of see I think from a security perspective, Jade is probably similar to Tresor in the sense that
I think there was actually a physical extraction vulnerability reported recently on it,
because I think, you know, neither of them actually have secure elements. So that is something that
people might want to think about, you know, whether or not you are worried about a sophisticated
physical attacker being able to extract key material from the device. You know, I think that
is a pretty edge case thing to worry about these days. It's not something that we've ever actually
seen happen in real life. These tend to be more, you know, theoretical, laboratory based attacks,
but you know, it is something that is possible.
Absolutely. Quick side question. How are you doing for time?
Oh, I'm good. I think I have several hours. Awesome. Awesome. Because we could definitely
I could definitely ask you. We could definitely go for several hours. There's so many things I
want to I want to talk about. And this is a topic that's so it's so complex and so deep that and so
interesting to people, you know, that I think it almost merits it. But all right, so let's go a
little further. So, yeah, the physical threat model really changes your setup as well, because
if you don't consider physical attacks, let's say home invasion from a sophisticated thief that,
you know, knows about crypto or let's say you're you're unless you're worried about like the Feds
raiding you or something like that, which to be fair, there are some people who really do need
to worry about that and that are not necessarily bad people. Right. I mean, I'll bring up the
example of Tate, which is, of course, you know, being now accused of significant crimes and maybe
he is guilty. I don't know. But, you know, how would somebody like Tate how do you think somebody
like Tate should actually secure their Bitcoin? You know, I don't know if he had good enough
counsel. You know, we're talking last week that he had like four coins for Bitcoin, some Binance.
And was it his pocket money or his huddle? And he just had really bad advisors. But he probably is
concerned about, you know, some sketchy raid from whatever Eastern European countries in and
and they would probably get this hardware wallets and then have to do something about it. So maybe
maybe go on to that, talk a little bit about that story or somebody of that level of prominence in
aggro. How should somebody at that level of, let's say, popular controversy and government enemies
handle their Bitcoin security, given that they do have, let's say, physical security as a part
of their threat model? Yeah, basically nation state resistance. There's, of course, a number
of ways that you could go and, you know, you can get really creative with any of this stuff.
But, you know, the short version is you need to have your keys distributed. So
this is something that we help our clients kind of grapple with because everybody has
different threat models and different thresholds of convenience versus security that they're willing
to manage and deal with. And at the extreme, which we certainly have some clients do,
the extreme of sovereignty and security is that you have a multi-key setup where
no threshold of keys exists within one jurisdiction or one set of jurisdictions that
are cooperative with each other. So essentially putting your keys in different countries
that do not cooperate with each other. So, you know, we have clients who literally have to
get on a plane if they want to access their extreme cold storage. Now
you can create, of course, any number of different wallets that you want. And so
most people who have that level of extreme cold storage, that's not their only wallet. They
probably have three, four, five, six different wallets that have different key setups that have,
you know, higher levels of convenience for spending. You basically have your
checking account and your savings account and your pocket money. For example, that would be like
pocket money is your single signature hot wallet checking account. Maybe that's like a two of three
wallet where you can access both keys within a matter of minutes or less than an hour. And then
your savings account is your super cold storage where it might take hours, if not even days,
to be able to access a sufficient threshold of key material to sign a transaction for.
But then, you know, you can of course go even deeper and try to decide exactly
what level of coercion you're even willing to try to protect against. So, you know, there are ways,
for example, where you could have this, call it like jurisdictional arbitrage or distribution,
and not have an extreme level of inconvenience if you're at the level where you
either have semi-trusted friends or family, or perhaps you have employees who are scattered
around the world. And in that case, you can have the key material in countries that don't
cooperate with each other, but you can be able to request signatures from those people. Of course,
then you have to develop, you know, various secure communications protocols and authentication
procedures, and that can get pretty tricky, especially these days as, you know, deep fake
stuff is becoming more problematic.
Yeah, I was going to ask you about AI. I know you've been playing with AI as well,
but I fear if we get into AI, we'll steer too far from Bitcoin self-custody. But yeah, I think,
you know, deep fakes and KYC authentication is increasingly tricky in the age of artificial
intelligence. So that's going to be fantastically entertaining to watch and also quite risky. You
know, you really have to have that ship under control. So, okay, so yeah, I mean, some,
you know, highly creative James Bond level, multi, you know, jurisdictional, multi-sig sort of
almost institutional grade security for that level of aggro. Now, let's roll that back into
something a little bit more, let's say, likely for a lot of people, you know, people that are
perhaps concerned with physical attacks, you know, there may be, they have a hardware wallet and a
12-word seed and their seed is stashed, you know, buried in some sort of metal device engraved
deep underneath their dog house or something. But then they're afraid that somebody gets to it or
maybe that they get their treasurer and then, you know, they'll go into a lab and try to extract
the seed and then they'll get their Bitcoin. So they'll do this 25th word thing, right? And the
25th word is like a feature that I believe a treasurer, you know, let the charge with,
which lets you add an extra word on your seed. So it adds an extra kind of piece of cryptographic
material to the seed, to the private key. And so your 12 words are now no longer sufficient
to regenerate the public addresses of that wallet, which in theory protects you from a physical
attack. But in practice, it also makes your backups fragile to loss because now you have to
securely store this 25th word and where are you going to store it? I mean, obviously, you're not
going to store it necessarily with your hardware wallet because otherwise, why create it in the
first place? Or you're not going to store it with your other 12 words because if you're afraid of
fiscal theft and they get to it, they're going to get the 25th word. So you got to store it somewhere
else. You know, what do you think of this particular security feature? Because I think it's
actually very problematic. I think it's more, I've seen more people lose their money to it than I've
seen people successfully defend their Bitcoin with it. And to do it right, you know, you kind of have
to spread that 25th word around enough so you don't lose it. What do you make of it?
Yeah, I mean, you basically just summed it up is that I've seen a lot of people lose money as a
result of problems with the 25th word as well. And, you know, I track the other side of this where
all of the physical attacks that do happen, at least that we hear about, I track. And
the vast majority of the time those attacks are successful. This is kind of related and touches on
another issue of, you know, decoy wallets, because you can use this feature, for example,
to have a decoy wallet. And the short version is we just don't have enough data. Like we,
there are zero cases that anyone is aware of where someone has been attacked and has successfully
either used a decoy wallet to pay off the attacker and save the rest of their stash,
or has been attacked and was saved because they just, they had a, you know, a 25th word somewhere
else that was inaccessible. So, you know, part of it is just sort of an issue of the amount of
incidents that we have to go off of. And, you know, comparing that to the number of people
that we have seen anecdotally lose access to their money because this is a more fragile setup.
So I certainly don't recommend using 25th word option for people. Now, like we said,
Trezor and I think Jade as well have potential physical extraction vulnerabilities, and this
is how you mitigate that. But also, as I said earlier, those physical extraction vulnerabilities,
as far as we are aware, have never actually been used in real life. So this is why security is so
tricky is because there's a million things that can go wrong, but unless you're talking about
what the actual frequency is, you know, when you're creating a sort of a threat model,
when you're looking at all the different things that can go wrong, you just, you have to prioritize
based on the likelihood. And so I think this is why a lot of people kind of get frozen up is we
talk about security, and I can only imagine like if a sort of normal person or a non-technical
person comes in, they listen to the security experts talking about all of these edge cases.
It can be incredibly scary because often we don't talk about, you know, what's the actual
likelihood of the edge case happening? And at the end of the day, nothing is perfectly secure,
right? So there's always a potential for something to go wrong. And I think that can cause people to
freeze up and say, well, I don't want to bother with this self-custody stuff because I'm never
going to be able to have a perfect setup. But the way that I try to describe it very simply,
like the different threat models, is that if you are using a trusted third party to custody your
funds, the number of things that can go wrong, like all of the different vulnerabilities and
threats to your money, are the same and actually greater than the number of things that can go
wrong if you self-custody. So basically the sort of total scope or map of possible vulnerabilities
and self-custody is a subset of the total number of vulnerabilities of third party custody. Because
basically a third party that is doing custody, you know, they are doing quote unquote self-custody
on behalf of other people. So all the things that can go wrong with the keys can go wrong on their
end. But also you have a whole new class of things that can go wrong with basically insider attacks
and basically the people who you are trusting to run the infrastructure that is essentially a black
box. You have no idea what security practices they are actually doing behind the scenes. And
even if they told you, you probably can't verify it. So when you take self-custody, you can at least
know what the security practices are that are being put in place around those keys and you
can verify them because you are probably doing it yourself. That's kind of the best way that
I have been able to quantify the scope and the threats of like why is it worth the effort
of trying to put yourself into a self-custody setup. >> You know, one of the ways that I framed it
when I was teaching this stuff to consulting clients was that there's one of the big threats
is that you build yourself a maze around your money that is so complex that you get lost in it
and you effectively screw yourself. So that's one aspect. That's like the extreme paranoia side of
it. But on the other hand, you know, trusting your money to a third party is basically the worst
case scenario on the other end from an overall security perspective.
Yeah, so okay, let's get into hardware wallet supply chain attacks because this is an area
where let's say the critics of hardware wallets are, they've kind of established somewhat of a
beachhead, so to speak. So what can you tell us about supply chain attacks in terms of,
you know, let's say Tresor or cold card and how can people have some certainty? Like what
protocols have been implemented to give people certainty that their hardware hasn't been
manipulated? >> Well, that's certainly tricky, but I will also preface it by saying that I'm not aware
of any incidents in which a supply chain between the actual manufacturer of a hardware device and
the end user ended up being compromised. More frequently what we see happening is someone buys
a bunch of Trezors or ledgers and initializes them and then resells them on like eBay or Amazon or
whatever and basically hopes that whoever receives them is naive enough that they don't understand
that the seed phrase that's on that device is already in possession by someone else and they
just start using it and deposit money on it. We've also seen counterfeit devices,
I think mainly Trezor counterfeit devices, but that's always, once again, like people
buying not directly from the manufacturer, either buying off a secondhand website
or I think there may have even been some like conferences where people were giving or selling
fake devices. So as long as you're buying directly from the manufacturer website,
you're in a pretty good position. And the devices themselves,
they will at least validate when you're updating the firmware on them, that firmware has to be
signed with a private key that ostensibly is only in the control of the manufacturer.
And if it's not, the device is going to throw up some major warnings and usually won't even
let you install the firmware unless you put it into a very special developer mode.
Other than that though, I'm not a hardware security guy, but once again, at the end of the day,
nothing is perfectly secure and it's practically impossible for even security experts to verify the
integrity of hardware. And this is something that has come up a few times over the years,
especially just like Silicon chips are so amazingly complicated that it's possible to hide
malicious pathways inside of the transistors, for example, that no one will notice for a long time,
if ever. It's kind of similar problems with really any hardware. And that is why, for example,
we talk about simplicity of if you are putting a private key on a desktop or a laptop,
just the overall complexity of that entire system of hardware and software is so huge
that there's just a ton of potential vulnerabilities. If you're putting your key
onto a mobile phone, it's still huge, but it's probably an order of magnitude less
than laptops and desktops. And then finally you get to these dedicated key manager devices,
which are extremely minimal in terms of their complexity. But even then, none of us,
except maybe Ledger Dungeon, but almost nobody is actually inspecting these devices at a chip level.
Yeah, that's amazing. And so the hardware wallets are the simplest of the devices.
This brings us to one of the, simplicity is good for security in general, right?
So this brings us to one of the aspects of Casa that I actually am a big fan of.
So let's say somebody has had these 12 words in a hardware wallet, they're good, but they want
to level up. Maybe their Bitcoin stack just grew 10x after a bull market and they're ready to level
up. One of the cool features with Casa is that you can have the two of three multi-sig setup,
which is the gold account, I believe it's like $120 a year. And you can have that set up in a
kind of custom setup where you can have two hardware wallets plus the mobile key. And that's
really nice because the UX of Casa is very good. And then you have these two hardware wallets,
and then you can actually have a hardware wallet diversification, right? So you have,
maybe cold card and then the tracer. And then so if one of them fails and you find out about it,
you'll have a little bit of time to react, right? And switch it up.
So I guess I just wanted to compliment you guys on that because I think that's a big piece of
the innovation that you guys have put forward. It lets you not only benefit from the open source
security of these hardware devices, but also lets you hedge against them by using multiple of them.
So maybe you can talk a little bit about that and tell us, I don't know, if you have any comments
on that particular issue. - Yeah, so multi-vendor, multi-sig
is I really think the only way to go. If you're going to the extremes of setting up multi-sig
and you're trying to eliminate single points of failure, then it's pretty dumb to put all
of the keys on devices that are controlled by a single entity. And it's not just about
supply chain attacks. Like I said, I'm not aware of any actual supply chain attacks
happening thus far. It is still a theoretical vulnerability. But what we've seen over the years
is that there are a lot of things that can go wrong other than just a supply chain attack
when it comes to using these hardware devices with your Bitcoin wallet. Sometimes there's
just sort of compatibility issues. Sometimes software or firmware updates can go out that
create a compatibility issue and it can take a while to get fixed. And so you may temporarily
not be able to use that device. Sometimes there's edge cases we've run into where a device may not
be able to handle a certain transaction because it's overly complex, especially when you're doing
multi-sig and you're trying to spend 100 plus UTXOs. Some of these devices kind of choke on
larger data payloads, especially with PSBTs. Now, you can easily end up with PSBTs that are
hundreds of kilobytes and can cause problems with these dedicated devices because they are
very simple. They are designed to be minimalistic. I did a bit of research back in 2020 where I kind
of stress tested a bunch of devices with complicated multi-sigs and complicated transactions.
I'm actually going to be redoing that hopefully in the next few months here with some of the newer
devices and using some of the newer Bitcoin scripting and just trying to see how well is the
hardware device ecosystem doing for that. But like I said, it's single points of failure.
And I don't want to focus on any one given thing that can go wrong because as we've basically
covered, there's a ton of things that can go wrong. And since you don't know what thing is
going to go wrong, having more diversity amongst your keys and how they're managed just gives you
more flexibility and often more escape hatches so that if something goes wrong, hopefully it's only
going to affect one of your keys. And that one key, if you're in a multi-sig, won't be a single
point of failure. So this sort of resiliency, it protects people from the complexity of the entire
ecosystem that you're dealing with, both at a software and hardware level. And it protects
people from themselves and making mistakes. That's what I think you and I have both seen over the
years, especially people who are using single signature setups. You can very easily screw
something up. And that one mistake that you made just because you were naive or you weren't paying
attention when you did it can result in catastrophic loss. And that is the type of
problem that I think is a major inhibitor to mainstream adoption, or at least mainstream
adoption of self-custody, which is what I think we should all be going for, is that simply the
ability for someone to be their own bank is not enough. It's that we need for people to be
confident that they can be their own bank and that they don't have to expend a ton of their time and
resources managing that aspect of their lives. - Yeah, there's a question of sovereignty,
which is, let's say, power or control over your jurisdiction, right? So let's say in this case,
we're talking about your private keys. You have power and control over them in theory,
but then there's also that aspect of responsibility. Are you actually able to
respond to threats? Are you competent enough to handle it? And do you feel confident in that?
And that, I think, is the area where sovereignty starts to get interestingly tricky, because you
do have to get to a certain level of competence to actually achieve any kind of sovereignty.
So somebody mentioned recently to me something about security, a concept in security, which I
thought was really interesting. He said, with security, the best that a security system can do
is alert you to a threat, some sort of breach, some sort of attack, somebody trying to jump over
your lawn or the wall around your house, whatever. And that alerts you to give you time to respond
to that threat and respond appropriately. So when it comes to Bitcoin security, especially now that
we're entering, we're talking about multi-sig setups with Casa or multi-sig in general as well.
How do you think people should respond? Let's say, maybe describe the two main offerings that
you guys have, two of three and three or five, I believe. And then talk to us about maybe the ways
in which people should respond to any kind of breach of a key or compromise and what process
you guys have in place for that. Right. So that's, I would say one of the main deficiencies of,
well, Bitcoin in general, like really any of these crypto assets, but especially single
signature setups, is that there is no responding to a threat. By the time the threat has compromised
your setup, your money's gone and there's no recourse, right? That's one of the reasons why
people are afraid of taking on the responsibility is because you basically have to front load
all of the security. There's no reactive security, at least for single signature setups.
Now, I will say that the OpVault proposal would change that. And that's why I'm a big fan of
OpVault because it would actually complete this whole other half of the security model that we
currently don't have in Bitcoin at the protocol layer. Essentially such that you would be able to
see that your funds had been compromised and you would be able to claw them back and put them into
a new, fresh, secure setup. But since we don't have OpVault, the next best thing that we can do
is put you into a setup where there are multiple keys and you are on a regular basis checking
that those keys have not been compromised. So this is where we have some concepts in CASA.
One of them is called health checks. Basically, every six months, if you haven't used a key,
we just send you a reminder, "Hey, go sign this cryptographic message and prove that you still
have control over that key." And so that kind of is a forcing function to get people to go and
check on perhaps some of the keys that are stored in locations that are harder to access,
that they don't go to very frequently, just making sure that they haven't suffered from
physical attack, physical theft, environmental damage, bit rot, really any number of things
that could go wrong where that key is no longer usable for really absolutely any reason.
And if you're checking in on a key and you find that it's gone or just not usable anymore,
then we have this other concept called the key compromise and key rotation, which is built in.
And basically, that allows you to just tap on the key in the app and say, "Hey, this key is
compromised. I need to replace it." You then go get a hardware device of your choosing, anything
that we support, and you just set up that new hardware device and we walk you through essentially
rotating your funds over from the old key set to the new key set. And the only difference will be
that one key that was compromised and the rest of the keys remain the same.
- Right. Okay. That's fantastic. What has been... What kind of data do you maybe give us a little
bit of an insight into whether people actually do rotate their keys? Or at least not rotate them,
but do they actually do the health checks? How often do they do it? Do people get lazy around
this issue? Or are they fairly responsible with it? - Yeah, that's actually something I would be
interested in knowing more. I do know anecdotally that definitely people are lazy about it. We have
clients who ask us, "Hey, I'm getting this notification. What should I do about it? How
worried should I be?" type of thing. This is some ongoing discussions. I think that we probably could
be pushier about it because we have certainly had cases where people ignored those notifications for
several years. And then when they finally got around to it, they discovered that maybe multiple
of their devices have stopped being reliable. Usually, it's just a result of the firmware being
really, really out of date. But of course, any number of things can go wrong. So we try to find
this happy medium, of course. Once again, it's trade-off between convenience and security.
And when you're in a multi-signature setup, there is more complexity. And the way that I see it is
that you should not think of it as a set it and forget it type of architecture. It's kind of like
a living organism, and you need to make sure that the heartbeat is still there every once in a while.
You need to make sure that it's still alive. Otherwise, you can still get into a situation
where you go too long, and then you try to spend your money, and you discover that too many things
have gone wrong with your keys. So I think incentivizing people more on that side of things
is going to be a path, a room for future improvements of getting people to actually
put in the work. You would think that the incentives are there because in many cases,
people basically have their life savings stored in these things. But it's kind of like what I
said earlier is data backups, for example. It's very hard to get people to do data backups,
even technical people, because it's just excruciatingly boring, tedious, monotonous task.
And the same is true if you need to go travel to one of your keys just so that you can tap on a
button a few times to make sure that the key material is still there and that the device is
still functioning and so on and so forth. So it's definitely room for improvement.
- Yeah, that's definitely interesting. And I definitely recognize that feeling of like,
"Ah, I'll do it later." Twitter is more interesting right now than going and digging up this device or
whatever. But yeah, and so, okay, so that's on the health checks. And then on the key compromise
issue, I mean, that's I think over time, the more I think about it, the more it seems like one of
the core principles of self-custody is like if you have any reason to think that your key has
been compromised, even to the degree like maybe you lose your phone or you lose your 12 words,
on a single SIG issue, immediately you got to move it to a new wallet. Because one of the ones that
I've seen a lot is people will like, they'll get like a 12-word seed from a crypto wallet,
right? Like something like Exodus, for example, which is a great wallet, I think, in general.
And then they'll grab those 12 words and then import them into their laptop and then their
desktop and then they'll import them into a bunch of computers. And then, of course, one of them
gets compromised and then that seed has been compromised. And they're like, well, but I have
my backup or whatever. They don't realize that that seed is no longer functional. They actually
literally have to move all their assets to a new set of public keys. And so, even if like maybe
their phone gets, they lose their laptop and then it takes three days for the hacker to get to their
seed. Like if you have any reason to think you're compromised, you got to basically move your coins.
And with single SIG, it literally means moving all your coins. Whereas with multi-SIG, you can
actually, well, I guess you will have to sign a transaction and rotate your keys to a new multi-SIG,
right? So you still have to. - Yeah, this is kind of an interesting
related point is there's single signature wallets already irk me enough, but I think the extreme
end of sort of concentration of single points of failure is the people who use
one of the wallets, like Exodus, for example, or you mentioned Coinomi earlier. And essentially,
they have all of their different tokens and assets all in one place, all secured by the same
set of keys. Like that is like the ultimate single point of failure. And as you mentioned,
of course, you start copying and pasting that data around into other pieces of software.
It's actually kind of analogous to credit card fraud of the more merchants that you deal with
and give your credit card info to, the more likely it is that one of them will be compromised or
will have a skimmer or something. And it doesn't matter which one of them, it's just like every
time you interact with a new one, you're handing your private keys to them essentially, and you're
trusting them with it. So it's the same type of thing with reusing private keys across multiple
different wallet software. - Yeah, like I imagine they don't,
probably something like this, they'll hear about some exciting new ICO or whatever the flavor of
crap coin it is this time around. And then they'll go in like download trust wallet, right? And
they'll get the 12 words and then they'll realize that it's on Ethereum on layer five, and then
they're like, "Okay, well, I guess I got to use MetaMask." And so MetaMask will be like, "Wait,
do you want a new wallet?" It's like, "Well, a new wallet doesn't have my coin." So they'll
import their seed. So now their seed is on trust wallet and on MetaMask and on two devices.
And this is probably how it just snowballs slowly out of control. And this is why Harvard wallets
are so interesting and so powerful 'cause it at least isolates you from that kind of shenanigans.
We saw the trust wallet extension hack recently, that was bad. And who knows what happens there?
Those extensions are pretty sketchy, but fun. Let's get into inheritance, right? So, okay,
you've got your multi-sig, your deep huddle, your Bitcoin OG at this point. You're going to be rich,
multi-generational wealth if you're not rich already, but you want to make sure your kids
can get access if and only if you get hit by a truck in 30 years. How do you guys
manage inheritance? Because inheritance, I mean, the Bible of crypto inheritance was Pamela Morgan's
crypto asset inheritance book, which I think is still very strong and people should go ahead and
buy that off Amazon if you're interested in this topic. But it's still like when she wrote that
book, you know, like 2018, at least the version that I have, and it was still, you know, the
wallets were not where we're at today. I don't think Casa existed yet. And her processes were
very manual and sort of paper-based, right? Where are we today with inheritance? Hopefully,
we've advanced a little more. How do you guys deal with it? - Yeah. And so, you know, the really cool
thing about multi-sig is that we finally have the ability to walk this very fine line. And
that line being that, you know, when you're doing self-custody, you want your setup to be such that
you and only you have access to the money. But then if you want inheritance, you want this
sort of switch to flip when you die that changes the security model of your money.
So that now, you know, some predetermined set of beneficiaries and executors have access to it.
And, you know, often a lot of people have gone down the path of saying, "Oh, you know, we need
to do some sort of like smart contract, or we need to have some sort of dead man switch." Or,
you know, they're trying to architect some fancy technical thing that usually ends up running into
the Oracle problem, which is essentially, you know, garbage in, garbage out on any of these networks.
There's no way to like programmatically know whether or not a human being is alive or deceased.
So the short version is that, you know, the authentication procedures that should be in place
around inheritance, they do need to be manual. They do need to require humans to verify the
death of, you know, the owner of a given wallet, and then, you know, institute whatever processes
have been put in place. Now, you know, the reason why I say multi-SIG is innovative here,
and it lets us walk this fine line, is that the way that we achieve that is by putting
a threshold of, you know, spending threshold of keys in positions such that they can be
accessed through an inheritance process. And, you know, more practically, some examples of that
would be that, you know, CASA holds a key, for example, so, you know, we can verify if someone
has passed and we're, you know, we're given the death certificate that we can verify with the
local jurisdiction that is, that's valid, and we can then process a signature request for our key.
You can place keys in other custodians. Usually, we recommend people put one key in a safety
deposit box. Now, that could be a bank, that could be a private service, but basically,
some other physical custodian that allows you to list beneficiaries on, you know, the access
to that box. So, once again, you know, that key would be accessible to you while you're alive,
but if someone other than you showed up, they would have to hop through the authentication
procedures to go through the sort of beneficiary process to access the contents of that box.
And then, of course, you can also, you know, you can potentially onboard other friends, family,
attorneys, so on and so forth, but, you know, the short version is it is, it needs to be a human
based process for, you know, ensuring the security of those keys, but the, you know, the reason why
this was maybe not completely impossible, but a lot harder to do before multi-sig is that if you
had a single signature setup, yeah, you could put that in a safety deposit box and it has,
you know, beneficiaries listed on it, but all of a sudden, that becomes a single point of failure
because a safety deposit box in and of itself is actually not completely secure. There is precedent
for safety deposit boxes being opened by malicious employees, by law enforcement, or even by like
the company itself, just like shutting down or going out of business and you're not telling
people. So, you know, safety deposit box has never really been foolproof, at least for, you know,
the security of a single signature setup. But, you know, now, you know, if you have your keys
distributed around, you don't have to worry about something going wrong with the safety deposit box
because as we already covered, if that did happen, no sweat, you just mark that key as compromised
and you go set up a new key and put it somewhere else that has not been compromised.
So, would this work with the 305 diamond package where, let's say, I don't know, you have one key
in your phone, you know, a couple of hardware wallets here and there, and then you guys have
one of the keys and then the last one would be in some sort of legally protected vault, you know,
that people would have to, you know, jump through a few hoops to get to? Or what, how does the
distribution of the keys look like? What are the numbers? What's the multi-sig range?
Right. So, generally, the people who are doing this, they're usually using our 305 setup. We
also have a 306 setup, though that's less frequently used. That's usually for people
who want to either onboard people like attorneys or possibly employees.
Now, it's possible to do sort of a do-it-yourself setup with the 203, but it's not something that
we really officially support as inheritance. But even within the 305, there's no single
cookie cutter template for this. And this is part of the reason why our premium tier offerings are
so much more expensive. It's not because, you know, the software is more expensive for us to
develop and deliver to people. It's because you're literally spending a lot more time
on the phone with us, essentially doing consulting to, you know, decide exactly how you want to do
your inheritance setup. And, you know, for example, some people only really care about
inheritance setups in the case that they die. Other people need to take it a step further,
and they need to, you know, maybe they have, you know, they have a spouse and they have children,
and they also want to account for the edge case where, say, maybe they and their spouse are,
you know, in a fatal car accident and both of their children survive. You know, that can get
a little bit trickier. Because, for example, one thing that we do see happen more commonly,
we have the ability for, you know, you and your spouse can essentially share and, like, clone
the mobile key. You can both have access to that. So, you know, in that case, if you pass and your
spouse is still around, they can access the mobile key. They can, you know, go to the safety deposit
box. Maybe they even have access to one of the other keys in, like, a safe at your house. You
know, it's a multivariate problem, and there is no one single solution. But the short version,
you know, comes down to is that you ensure that you have a threshold of keys that will be accessible
by your executors of your estate. And you can get pretty creative with exactly how you do that.
- Right on. Right on. Yeah, that's really interesting. I think,
I can't think of another company that's doing, that's providing this service.
And if they are, I don't think they're doing it kind of at the level that you guys are doing it.
Because, you know, multisig is a big promise, right? It's a big promise of Bitcoin. I think
it does solve a variety of problems. And, you know, I wonder, this kind of takes us to the next,
I guess, the cutting edge of Bitcoin security. There's a variety of wallets that are kind of
being developed, you know, very much in the cutting edge that are kind of playing with
this concept of kind of like a community huddle or let's say something like Fetimint where
custody is technically centralized, but maybe we can enforce some rules around what kind of spending
conditions those funds can move on, right? I'm not too familiar with the details of these kind
of cutting edge technologies. Maybe you are, I know you mentioned OpVault. Where do you see like
five years out, you know, where could we go, right? Let's say we're in a world of high fees
where Bitcoin transactions are a hundred bucks on average, right? Or something like that. And
everybody's on, you know, on lightning, right? So now we're dealing with lightning level security.
How do you see that side of things? You know, I know I just asked two very big questions,
so feel free to like navigate or however you like, but like, you know, community enforcement,
right? Like maybe I trust my brother enough with a certain amount of power over my huddle,
or maybe my family trusts me as the local familial Bitcoin nerd. And then they decide
that they want to give me a key to their multi-sig or to their FedE, you know, vault or whatever.
Do you have any thoughts on how that might develop?
Right. This could go a number of different ways. But one thing that I'm interested in,
aside from OpVault, because we don't know when or if that will become available,
but there is a very large design space that has not been explored yet with Bitcoin scripting. And,
you know, Taproot and, you know, Tapscript itself have not really been utilized that much yet.
And there's a number of interesting things that you can do of, you know, creating different
escape hatches, if you will, you know, ways of getting out of situations where you could
essentially start collaborating with other people. I do hope that we'll start seeing more kind of
like multi-institution custody, if that's even a word. But this is where I think some of the lines
are going to get blurred between self-custody and third-party custody. And it's definitely
going to be interesting to see how it falls out on the regulatory side of things. But,
you know, there are some interesting questions to be answered around, you know, how does the law
view ownership of an asset where no entity has the unilateral ability to spend or to block
spending of the asset? And by that, I mean, you know, what if you're in a multi-SIG setup where
you have one or two keys, say it's a three or five, you have one or two keys,
and then say, you know, the rest of the keys in your key set are each held by completely different
people or institutions. That could get interesting. You know, you're really, you're distributing trust
even further, and you can get to the point where, you know, in a sense, you're distributing more of
the trust away from yourself. You know, this could work well for the people who don't trust themselves
to do self-custody. Maybe, you know, right now it sucks that a lot of people just threw up their
hands and are like, "I'll just let Coinbase handle everything," or "I'll just let Binance handle
everything." You know, that's a single point of failure. That makes me cringe. I would,
while it's, you know, I would still push for pure self-custody, I would still feel better
if someone was in a setup where, you know, they had a key and Coinbase had a key and Binance had
a key, for example, which is just distributing the trust more so that, you know, collusion is
less likely, but, you know, catastrophic failure is far less likely just because the keys are spread
out. You know, this starts going down a sort of deeper rabbit hole around reputation, but the
point being that, you know, you can spread your keys out amongst a number of reputable providers
and they can't rug you. That's really what we're going for. It can get even more complicated as
we're seeing, you know, many script technology and the tooling around that being developed out
so that, you know, for example, I've talked about degrading multi-sig for a number of years.
What if you have your two of three or your three of five setup and, you know, that's kind of your
happy path for spending, but now thanks to the magic of Tapscript, you can start adding in a
number of alternative spending paths that are time-locked. So, you know, you could say, you
know, this is my normal multi-sig setup, but if a year passes or if several years pass and the
funds haven't moved, then, you know, maybe that is because I've lost a spending threshold of keys
and I've essentially gotten locked out of my own multi-sig. Well, if you have been smart enough to,
you know, plan ahead of time and you've created alternative spending paths, you can say, okay,
if that really long period of time has gone by, now allow the funds to be spent with some
completely different set of keys. Now, of course, we start going down a completely different rabbit
hole here. You know, maybe that's a different set of keys that you yourself are holding and
have distributed around in sort of emergency backup situations, or maybe they are what I
mentioned earlier, a kind of distributed multi-institutional set of keys. So,
this is what I've been thinking about more recently because the tooling is starting to
get there for us to actually, you know, implement this stuff and make it more practical. Unfortunately,
we're not yet really at the point where we have sort of system-wide compatibility,
so that this is user-friendly for people to do. And it will also involve
a fair amount of experimentation. So, what I do expect to happen, much like with what has
happened with Ethereum and smart contracts, you know, for the first several years of that,
it was a complete shit show. There was no guidelines and best practices. And so,
the vast majority of people would write smart contracts that had vulnerabilities.
And over time, you develop those best practices, you develop more tools and sort of verifiability
and sanity checks and stuff. So, I would expect that, you know, what will happen is we'll see a
lot of experimentation and people creating all types of weird conditions. And then as the years
go by, we'll figure out, oh, that was actually not a great idea because of X, Y, or Z. And we'll
probably settle on, you know, a handful of fairly well-understood kind of templates, if you will,
for self-custody models and ways of making them more resilient.
Yeah, yeah. That sounds like that's probably how it's going to play out, you know. And in that
sense, the wild experimentation of Ethereum does kind of inform us about how things might play out
in the future in Bitcoin. I mean, as a libertarian, I have this fantasy of, let's say, a kind of civil
society, you know, like a societal huddle, right, where people choose all these different kind of
civil contracts of self-custody. And we build this sort of, you know, this mesh of security
where, yeah, you don't have any single point of failure or any obvious single point of failure.
And where we're also not standing exactly on our own island. Like maybe, you know, I think
that the multi-sig templates that you guys have created are very strong and they're probably going
to be one of those templates that stand throughout time. I certainly am a big fan of it. But yeah,
like I do wonder where we're going to go, especially in a lightning world where custody
itself, like it's not obvious to me what self-custody is in a lightning world, right? Like
you're in a two of two multi-sig with a time lock. Who controls that? Who owns that? How do you
define that as, you know, how do you define the custody of that? It really does blur the lines.
So yeah, maybe you can talk to us a little bit about that, like the two of two multi-sig and
the lightning side of things. And I think after that we can open it to the audience. We've had a
great conversation and I think we kind of covered a wide range of self-custody, which is quite a
feat. So thank you for that. Yeah, you know, it becomes less of a sort of static question of who
has custody. Whenever you're creating a dynamic system, and by dynamic I really mean a system
that can change over time due to any variables, then as you said, the sort of the lines of custody
get blurred. So, you know, lightning works primarily due to game theory and incentives.
And, you know, not only are there various timeout issues of, you know, if you're not paying
attention for several weeks, then your counter counterparty could screw you. There are other
issues of, you know, if you're sending extremely tiny payments, for example, and the on-chain
fee environment is fairly high, then the sort of game theory and incentives can make it such that
you could get screwed out of some of those micro payments as well. And, you know, this is just
another rabbit hole that you can start to go down. Yeah, I have a question that's a little bit off,
let's say off topic. Where are you at with artificial intelligence and in particular,
the identity issue that's kind of coming up, right? All these bots, very conversational and
chatty bots that are difficult to distinguish from conversational chatty humans. Do you,
you know, is Casa going to be developing an identity offering, some sort of cryptographic
authentication to identity? Like, where do you see that side of things? Because it does seem like
there's a great, they call it the deep, the great flood or the digital flood, the deep fake flood,
you know, where it seems likely that we're going to get these impersonator bots becoming very sharp
and very convincing. And it's a whole new era of phishing schemes. And the only solution might be
cryptographic authentication of identity. So yeah, where are you at with that?
Yeah, so, you know, there's three major types of authentication, you know, something you have,
something you know, something you are, and the AI stuff definitely weakens the something you are
aspect. Usually something you are comes down to sort of biometrics and your biometrics in general
is a pretty sub optimal form of authentication, because you can't rotate those credentials. So
if they get compromised, once you're screwed for the rest of your life, the biometrics can be
helpful as a sort of additional factor of authentication. But like we've been seeing,
especially with the deep fakes as well. We're, we're at a point in time now, where what's
happening is that the ability for you to believe what you see and hear is being degraded, you know,
it used to be for the first several decades of the internet, even though, you know, there'd be a lot
of, you know, misinformation and disinformation. If, at least if you were able to like see and,
and talk to people through the internet, you could still rely upon that as being, you know,
a real person and real information that isn't being made up by a machine. But now,
you know, we can't even really believe what we're seeing and hearing anymore, because the machines
are getting to the point where they're going to be able to construct, you know, entire realities
that are indistinguishable for humans. This is going to get even crazier, I think, you know,
as virtual reality and augmented reality. And, you know, especially if, if technologies like
Neuralink come along, and people start literally, you know, jacking their, their brainstem into the
internet, it's going to be some pretty crazy experiences, and you're kind of like, ready player
one slash snow crash, style stuff where like, the hackers may literally be compromising your brain
and what you see and hear and believe. So you know, this is something it's, it's a dynamic
environment that we're staying on top of, as usual, you know, CASA does not want to implement
any novel solutions ourselves. Rather, we want to continue to remain apprised of what all of
the tooling is that is being developed and is being made available. And really, like our bread
and butter is to, to look out and survey the landscape of technology and find the technologies
that exist that are highly promising, but are still way too difficult for the average person
to actually utilize. So my, my take on this really, ever since I started working on Bitcoin stuff,
in 2015, my, my mission was never to, you know, focus on Bitcoin, my mission since then,
has been to use my skills as a technologist to help individuals empower themselves. And it just
so happens that, you know, working on Bitcoin was one of the most straightforward ways to do that.
But, you know, there's going to be plenty of other, you know, cryptographically enabled
technologies that also let people empower themselves in other ways. So identity, of course,
is going to be one of those. And, you know, we have yet to really see, I think, a leader emerge
from the pack. There's a number of different projects that are trying to solve that. If we
see one that we think is far superior and is gaining traction, then that's what we're going
to go with. But, you know, for example, one of the biggest identity technologies that's out there
right now is WorldCoin. And I'm sure that if we integrated WorldCoin, we would get a lot of flack.
And, you know, I'm not even saying that it's something that I would want to implement.
Obviously, there's a lot of really terrible aspects to WorldCoin and the sort of biometrics
aspect of it. But, you know, I have talked to some of the people at WorldCoin about,
you know, their perspective on things. And they are, you know,
they are trying to solve some real problems. And they recognize that AI and identity is going to
continue to become a big thing. I may not completely agree with their solution to it,
but I certainly agree that the problems that they've outlined are real and need to be addressed.
>> Yeah, it's such a thorny problem because, you know, again, like that conspiracy
theorizing part of my brain goes like, well, this is clearly a problem reaction solution thing. And
so the problem is AI, the reaction is, oh, my God, everything's fake on the internet.
The solution is KYC all the humans, right? And I really don't like this KYC all the humans feature
because, you know, if we've seen any big revelations against power and their dark
agendas, which do happen at times, they've come from anonymous sources that they needed to remain
anonymous for long enough for the data to get through and out into the world. So again, Snowden
for the two weeks that were like two days that he was anonymous. And then, of course, Julian
Assange and all the work that he did, you know, it's a miracle that Satoshi remains anonymous.
And that's a whole other rabbit hole. But, you know, is there a path to survive the deepfake
flood without KYC all the humans? Like, do you see a path to that?
Well, I mean, part of the problem is also the sort of underlying question of what is identity
and reputation. And so, you know, I reject the premise that identity needs to be a sort of one
to one, you know, human body mapping, sort of government issued identity. I really think that
people should have the flexibility to create different identities for different things and
to, you know, not necessarily have to associate all of their different activities and all of
the different aspects of their lives sort of together as one single entity. And so,
the, you know, it may, the question may not even be correct. Like, worrying about whether someone
that you're interacting with and whether or not they're human or AI, in many cases, that
may be irrelevant. What's more important is that you know that you are interacting with the
identity that you believe that you're interacting with. So, I don't see any reason why, you know,
AIs couldn't have and manage their own identities and reputations and so on and so forth.
All, I think what we should really be focusing on more is just preventing AIs from being able to
spoof themselves as other entities. And so, there's certainly ways around that. It just means
we can no longer, we can no longer just rely on kind of unauthenticated data streams that we're
receiving, aka, you know, we can no longer rely upon simple text or video or audio and consider
that to be a form of authentication. And of course, this is where private keys and cryptography will
come in and play a big role. Yeah, it seems like the whatever PGP brought into the world
might be really the only solution here. And hopefully, web of trust can actually make a
comeback. Obviously, technology that might be a little bit easier to use than PGP or just
integrated into things like Twitter and so on, which to some degree does seem to be like that's
Elon Musk's vision to make Twitter the everything app or whatever that means. It probably will mean
identity and authentication. And yeah, that's going to be fun. Are we going to see a Jamieson
or a lopp.ai bot in the near future? I don't know. I think I saw a project recently that was trying
to do something like that, but for Bitcoin in general, and I would certainly love to have an
AI bot that is basically ingested every educational resource that I have linked from my website,
there's like 1500 different websites that it links to. If we can get something to
ingest all of that content and sort of, you know, be as you say, the sort of the lopp.ai expert
where people can basically talk to me in digital form, but I can scale myself out. That's really
what I've been trying to do for the past several years, you know, with the website, it was an
entirely self serving effort, because I got tired of having to go find resources, you know, getting
the same questions over and over again, having to give people the same answers over and over again.
And so I compiled all that information so that I could just send people, you know, one link,
say, you know, go, you know, read, read all of the things that I have in this particular section.
But yeah, having an AI of me, or of, you know, of all of those resources that you could easily query,
I think would really take that to the next level.
Yeah, I'm definitely bullish on that, that side of the AI world. So cool, very cool. So if somebody
made a third party project of Lopp.net or lopp.ai, would you be cool with that? Or what?
Should we just DM you? Because I may or may not be plotting to create such a thing anyway.
Yeah, well, you know, read the footer of every page on my website. I claim no intellectual property.
Wonderful. Fantastic. Yeah, I think I think that would be a really interesting product.
One question that I had gotten from a friend of mine was, what do you think about
Green Wallet and Jade's two of two time lock multi-sig scheme? He was complaining that
once he realized what it was, he said that they could potentially block you from spending your
money for a while and that it was kind of sketchy and that he turned it off immediately. Do you
think there's any value to these kind of like on-chain security two of two time lock multi-sig
schemes that have nothing to do with lighting specifically? Right. You know, I mean, they're
trying to eliminate single points of failure. I just I do fundamentally have a problem with
two of two when it comes to anything. You know, as we've said a few times today, 25th word pass
phrase also ends up creating a two of two setup. So you end up finding two of two setups in a
number of different places in the space. And it's not always obvious that you're putting yourself
into a more fragile position. So, you know, that's why they have to offer that that time lock and
sort of escape recovery ability because it does it actually creates a single point of failure.
The whole issue of doing time locks, that's something I've given a few talks about and
it's something that's also on my mind a lot recently. You're thinking about many script
and tap script and creating all of these alternative spending paths that will have
time locks associated with them. That's going to create a number of sort of compatibility issues
and sort of ongoing maintenance issues for wallets. We've already talked about how,
you know, with CASAS multi-sig, we view it as more of a living type of setup where you need to be
checking on it every once in a while, make sure that everything is healthy. And that's going to
become even more true as we implement these more complex, you know, time locking spending conditions.
Because basically, in order for you to be able to implement any sort of time locking
for your funds, like for alternative spending paths, then that means that you're adding more
dynamicism to your architecture. And it basically means that if you don't change your UTXOs
and those time locks expire, then the security model of your funds changes. So you actually
need to get to a point where you are more frequently staying on top of the exact state
of your wallet. And, you know, as these time locks get passed, you need to rotate those funds
to new UTXOs that have, you know, updated spending conditions around them.
So obviously, this is adding more complexity. And we're currently in the sort of research
and development phase of like, how do we allow people to leverage this functionality,
but do it in a way so that, you know, they're not going to be shooting themselves in the foot,
they're not going to be accidentally degrading their security due to laziness, which we have
certainly seen there is an abundance of. - Yep, the laziest path that works tends
to be the one that people take. All right. Cool. So anybody have any more questions?
It's been a great show. We've gone pretty deep, I think, really kind of scratching some of the
edges of self-custody. I think, again, like we're going to be publishing both of these
conversations with love today, the Q&A and the earlier conversation at juangal.com. So go sign
up to the emails, that way you'll make sure to get it on your inbox. And it'll also show up on
your podcasting app. And so you can listen to it on podcast and check out BitcoinNews.com. They have
a new part, part two of their interview with Michael Saylor in Europe out now. And it's a
really solid show. Michael Saylor is a fascinating cat. So do check that out. And we also have,
I think there's an announcement with Bitcoin Racing. There's doing some sort of early bird
tickets for their same racing event in London. I believe that's London, UK, not London, Canada.
And yeah, so check that out. That's pinned on the nest. And I mean,
unless anybody else has any more questions, I think that we can call the show a wrap. It's
been really good to have you, Jameson. Hopefully you enjoyed the show and can join us some other
time when there's something new to cover. Awesome. Thanks for having me.
Hey, everybody. Thank you for listening to the show. If you made it all the way here,
you are a true fan and a true OG. This is not the kind of conversation that's going to get mass
attention, but it is the kind of conversation that the ones that will be here in a long time,
that'll be here in 10 years, will probably have listened to because this conversation,
like many others that we have on the Juan Galt Show and in this Bitcoin News,
Twitter Spaces collaboration, are trying to scratch the edges of the industry while also
covering the most important news. So thank you for being a part of it. And I hope you'll join
us on the next one. We are doing these shows every week on Twitter Spaces. You can go to
the Bitcoin News Twitter handle or look for me and my tweets, as well as sign up at juangalt.com
to make sure you get an email notification when the shows come and ahead of time so you can
hit a reminder. Finally, I have been building an archive of my work at juangalt.com, as well as
publishing new articles touching the most important topics, I think, of our era. This is
bigger than Bitcoin. It's macro. It is romance and love and relationships. It is spirituality.
And these are long-form articles, not the kind of short-term content that most people are used to.
And honestly, you're probably going to want to read them in a couple of sit-downs.
Maybe I'll split them up into pieces. But there's a lot more to say. And I am very proud of them. I
think we have some very, very good work there. So I hope you'll check that out and follow me
in this journey to try and influence the future of humanity into a positive, optimistic, and
harmonious future, which integrates not just, let's say, the wisdom of the past, but also
the technologies of the future. So thank you again for listening in. And I will see you in the next
one.