Latest / TFTC: A Bitcoin Podcast / #785: You Are Your Own Single Point Of Failure with Alex Bergeron

Transcript

#785: You Are Your Own Single Point Of Failure with Alex Bergeron · TFTC: A Bitcoin Podcast

• 0:00
  You've had a dynamic where money has become freer than free.
• 0:10
  If you talk about a Fed just gone nuts, all the central banks going nuts.
• 0:16
  So it's all acting like safe haven.
• 0:18
  I believe that in a world where central bankers are tripping over themselves to devalue their
• 0:23
  currency, Bitcoin wins. In the world of fiat currencies, Bitcoin is the victor.
• 0:28
  I mean, that's part of the bull case for Bitcoin.
• 0:31
  If you're not paying attention, you probably should be.
• 0:36
  Alex, welcome back to the show.
• 0:39
  Thank you, man.
• 0:40
  It's been a minute, but there's been a whole lot of things happening.
• 0:44
  So we've got a whole lot of things to catch up on.
• 0:48
  I've been hassling you for like six months, I feel like.
• 0:51
  Yeah, I saw you in PubKey, I think was the last time, right?
• 0:55
  during the
• 0:57
  Up Next
• 0:58
  conference.
• 1:02
  It seems like forever ago, but I don't
• 1:03
  know that it was that far.
• 1:05
  It was just like
• 1:06
  it was earlier this summer, I guess, right?
• 1:09
  It was like March.
• 1:11
  The suits are here.
• 1:13
  The suits are here to fix quantum.
• 1:14
  They're stakeholders.
• 1:14
  Oh my god.
• 1:17
  Exactly.
• 1:18
  Speaking of
• 1:19
  that story is completely
• 1:22
  faded into
• 1:24
  nothing really that really felt faded into the background i wonder uh if it's gonna
• 1:31
  rear its ugly head back but with the way the market is uh going at the moment i don't feel
• 1:37
  like anyone's trying to hear some we've had plenty of bad news so um i don't think the
• 1:42
  quantum futsters uh are going to get people's attention with this press action no because i
• 1:50
  remember I'm looking at the chart right now I think you know it was like mid
• 1:54
  February early March but it was mid February remember saying to you
• 1:59
  specifically because they were saying the price won't go up until we have a
• 2:02
  clear path to fix this I was like all right well change your and your tune and
• 2:07
  we are higher than we were in that period but yeah I mean I think we were
• 2:13
  just discussing before we hit record it's been a pretty insane three weeks in
• 2:18
  Bitcoin today were, I mean, the last two days were up over 10%, but this is on the heels
• 2:24
  of a pretty traumatic experience with the cold card.
• 2:29
  Vulnerability being exploited.
• 2:31
  And then on top of that, a number of other projects being probed by AI tools and more
• 2:42
  vulnerabilities being discovered.
• 2:45
  And you were saying this, you think this is like an acceleration point for Bitcoin and everything, really.
• 2:54
  Yeah, well, I mean, it, you know, it certainly triggered a lot of, as you mentioned, reflections.
• 3:01
  And it feels like one of the bigger and most important one is a heavy dose of humility, really, for everyone involved.
• 3:13
  I think kind of all realizing that we were, everyone was swimming naked, basically.
• 3:22
  I was reflecting on this with someone just the other day and thinking about how we have been, you know, on every crypto project's asses, laughing about all of their vulnerabilities and how, you know, shitty their code bases or the DeFi platforms.
• 3:42
  where for getting hacked throughout the last couple of,
• 3:47
  well, little did we know that we had in some way, shape or form
• 3:51
  the same type of exposure in a lot of our fundamental architecture,
• 3:57
  fundamental infrastructure.
• 3:59
  I think the bounties were not as readily available
• 4:05
  as they would have been in the sort of 2021 era DeFi smart contract.
• 4:12
  But it was certainly there.
• 4:13
  So yeah, this certainly takes us off of our higher horse for hopefully what is the next
• 4:25
  step, which is going to be, well, you know, we have to kind of get a little bit more serious.
• 4:31
  And I think, you know, it was time for the adults in the room to kind of...
• 4:40
  Not that it hard to say you know it was kind of like it feels I what I wanted to think that I tweeted really as this was unfolding was just that this was the most cathartic kind of moment for Bitcoin in the sense that you know I think one of the core totem if you will of the OG Max C community And you know certainly forced a lot of reflection on the tribe but I you know
• 5:21
  I, I, I've said, I'm lucky enough that I've, um, unfortunate that I wasn't affected.
• 5:27
  None of my close friends or families were affected and I feel for the people who were.
• 5:31
  But at the same time, this really has made me hugely bullish in terms of seeing the ability for some certain groups, certain individuals to rally together and kind of lead.
• 5:49
  I think, you know, you needed the emergence of leaders, new leaders in Bitcoin for a significant amount of time.
• 5:56
  And I think for the last couple of years, it wasn't really clear who was going to pick up the slack, who was going to who was going to be this adult in the room to say, OK, like, you know, we got to kind of mature and take this to a level where it's not a little project, a little experiment anymore.
• 6:14
  And, you know, I think this process hasn't quite played out fully, but it feels to me like we're kind of like on the right path and surely a number of, again, organization, individual projects as kind of seemingly stepped up to the plate there.
• 6:32
  Yeah. I mean, you said we need to get more serious. What does getting more serious look like in your mind?
• 6:35
  well um i mean you know it's easy to say in hindsight and um but i had some interactions
• 6:48
  with spaces with american hodl you know last week or whatever a couple of weeks ago i guess now
• 6:55
  just in the days ensuing following the events and we were talking about how you know we were kind of
• 7:01
  all enamored with this garage band type of shop that cold card was and how slick it was that you
• 7:10
  know they were true to the authenticity of you know in the spirit of the bitcoin project but at
• 7:18
  the end of the day you know we're realizing i think we all knew in some way shape or form and
• 7:24
  But it was a three or four people shop managing the security infrastructure and providing security products for – I mean, it might have made sense back in 2018, all the way up to 2020 or whatever when the amount of money that was secured by those cold cards might have been significant.
• 7:54
  significantly lower than it was today. But, you know, now that talking about large sums of money
• 8:00
  and people's life savings that have grown considerably throughout the years, I think
• 8:06
  one part of getting a little more serious is having processes in place where it doesn't matter
• 8:11
  if you're an OG and that you've built this from the ground up, you need to kind of institutionalize
• 8:17
  some sort of framework to constantly, especially in the world that we, I mean, again, I also told
• 8:25
  you just earlier that this is not really not only about cold card. It's about, this is a story that's
• 8:32
  larger. That's about the AI security paradigm that we're entering. And it seems very clear that
• 8:38
  it requires constant vigilance.
• 8:45
  But it also kind of, to me, one of the lessons is really that
• 8:50
  we're all most likely to be our own most obvious single point of failure.
• 8:58
  Like in the sense that it doesn't matter how smart you are
• 9:01
  or how many years you've been into Bitcoin,
• 9:03
  you're likely to have blind spots in the way that you operate things and if you depend solely on
• 9:10
  yourself to secure again those life savings you are I think kind of missing out on the opportunity to
• 9:22
  optimize again the security and it's all about just having the ability to distribute the trust
• 9:28
  So that indeed you can follow best practices yourself, but there are too many scenarios now where that's just not going to cut it.
• 9:39
  So for me, yeah, that lesson has been kind of just we can do like, you know, self-custody doesn't mean that you have to do it by yourself.
• 9:49
  And I think this is kind of where I'm hoping we move in this new direction where we're a little more creative, we're a little more open to, because, you know, this dogma I think that we had instituted kind of shunned or kind of probably limited the amount of more creative ideas.
• 10:19
  that could have arrived in terms of potential alternatives to secure your funds right And those are not alternatives where you have to throw away the hardware wallets or anything
• 10:30
  like that, but more so kind of complement where, again, you have the ability to distribute
• 10:36
  the trust.
• 10:37
  And I've always said, you know, and throughout, especially the last couple of years, there's
• 10:42
  always been this misconception about Bitcoiners that trust is to be avoided at all costs almost.
• 10:49
  And it always was kind of, I felt was an unfortunate narrative, really, because ultimately, I mean, Bitcoin was invented because, yes, you cannot trust anyone with the management of a base reserve currency and, you know, the issuance of money.
• 11:11
  but certainly was not created because all forms of trust need to be
• 11:15
  need to be eliminated and you know trust can be usually beneficial
• 11:19
  especially if you can engineer it the way that we're able to do so
• 11:23
  so this is what I'm kind of like looking forward to
• 11:26
  and yeah it's just a great
• 11:30
  it kind of creates this you know it kind of creates a vacuum here now with
• 11:35
  cold card being you know more or less gone of like okay
• 11:39
  Where do we go from there?
• 11:42
  BitKey is one project that is kind of seemingly out of nowhere
• 11:49
  took a center stage, and it's because they deserve it, really.
• 11:52
  It's a fantastic product.
• 11:55
  I use it myself.
• 11:57
  But I think the vacuum is more like, yeah, let's just get back to the design board
• 12:04
  and see how we can push this to really kind of new,
• 12:10
  something that's just more modern.
• 12:12
  You know, we'll get into that, but what I'm building,
• 12:14
  what we're working on at our clubs with Arcade and whatnot,
• 12:17
  one of our mantra internally and kind of externally as well
• 12:20
  is just like modern Bitcoin tools.
• 12:22
  And I feel like we've kind of been living in a very primitive
• 12:25
  sort of development environment for the last decade
• 12:30
  where we've had an evolution of improvements to the protocol
• 12:32
  and improvement in, say, libraries and kind of like new ways to build things on Bitcoin.
• 12:39
  But they were kind of very much iterative and kind of piecemeal attachments.
• 12:47
  But no one really kind of ever took a step back and be like, OK, how do we actually piece this together
• 12:52
  so that someone wants to build a multi-sync solution, include new signers,
• 12:58
  distribute the trust via different methods of time locks and perhaps, you know, additional logic there.
• 13:07
  How can we take all of this, you know, this corpus of protocol enhancement and techniques and arc and batching and all of this?
• 13:16
  How can we bring this together and just provide builders and, you know, users something that is that they deserve?
• 13:26
  after how do they get reliable Lightning payments
• 13:30
  that doesn't require running a node, right?
• 13:33
  It's been 10 years.
• 13:35
  Now people are finally managing to ship that,
• 13:39
  but that's not the only thing that matters,
• 13:41
  and security is one of the big important pain points,
• 13:44
  and so we can do better there,
• 13:46
  and it was just a matter of kind of, yeah,
• 13:49
  just uniting all of these pieces
• 13:51
  and I think creating an interface that people can use.
• 13:54
  Yeah. Looking at my notebook, that's one of the things I wrote down in the wake of this
• 14:03
  cold card stuff. As many lessons we learned from this debacle, mainly in humility and the dangers
• 14:08
  of groupthink enabled by hubris and peer pressure. And I think to your point about the lack of
• 14:15
  innovation, I think a lot of that was driven by hubris and peer pressure, like no trusted third
• 14:19
  parties and to your point, trust exists for a reason in certain situations.
• 14:25
  And not only from that angle, but also the learning from what's going on in DeFi and other
• 14:34
  protocols and incorporating the good ideas in Bitcoin.
• 14:38
  There's been some dogma about even attempting that over the years.
• 14:41
  And I think to your point, it's time to brush all that away, look at it with clear eyes.
• 14:47
  And that's why, again, I've been paying you for for six months because you've been teasing what you guys are doing with Arcade and Arc Labs and Covenants.
• 14:57
  And I think the timing of you guys releasing these products and really leaning into like, hey, we're going for this.
• 15:06
  We can do things a different way is is perfect timing, because when you consider what's happening in the world of stable coins and AI now, particularly with Stripe acquiring Open Router.
• 15:16
  I'm not sure if you saw their investor lender
• 15:20
  about that acquisition,
• 15:24
  but they're really going full bore
• 15:26
  and digital payments using stable coins
• 15:30
  and their permission blockchain I forgot if it called Metronome or something like that but I think Tempo That what it is yeah If Bitcoin wants to The metronome sets the tempo
• 15:44
  If Bitcoin wants to compete, I mean, I think we have to begin exploring these different
• 15:50
  avenues.
• 15:51
  And I think your team at ArcLabs is definitely on the frontier and the tip of the spear of
• 15:55
  doing that.
• 15:57
  Yeah, no, exactly.
• 15:59
  And I feel like, again, just coming back to what we can take away from this saga is also that we need to be less, I think, prescriptive about the models that we had in our head about how things should be built and the level of trust that every solution should carry.
• 16:27
  And, you know, I had one easy example is just like I had someone just a couple of days ago or so that had an interaction about what we were doing with Arcade and asking why our model whereby, you know, we use transaction logic and covenants in a way that's not enforced by Bitcoin consensus.
• 16:47
  You know, how is this useful? Because then you need to, not that you need to explicitly trust someone, but the, you know, the reasoning there is that, well, if it's not secured or validated by Bitcoin, you know, it serves no purpose and has no utility.
• 17:10
  And, you know, for me, this couldn't be further from the truth. And the thing that I quickly mentioned was, well, you know, just like you're not going to have every transaction validated on Bitcoin, which wouldn't make sense, you're not going to have every financial operation or application contract or covenant validated by Bitcoin.
• 17:31
  And there's no room for that.
• 17:33
  And there's no point really kind of doing any of this.
• 17:37
  So, yeah, for us, it's been the perfect opportunity indeed, because if you followed, if you've
• 17:45
  kept track with my account and the way that we've been going to market with Arcade, we've
• 17:53
  been kind of distancing ourselves from quite a bit from the original, I guess, foundation
• 18:00
  of the company, right?
• 18:01
  At the end of the day, our company's name remains ARK Labs.
• 18:05
  And the original vocation, I guess, was to kind of manifest the ARK protocol.
• 18:14
  And it's this idea to scale Bitcoin payments and create the first main net implementation
• 18:20
  of it.
• 18:21
  And we achieved that, but we achieved this quite a while ago.
• 18:25
  But very early on in that process, our mind was set on sort of much more ambitious architecture
• 18:38
  because, yeah, we just realized that, well, it turns out that what we can do with ARC
• 18:43
  appears to fit as only kind of call it a module in the entire sort of toolkit that we're building.
• 18:50
  So this idea that you can do off-chain transactions and batch them on a Bitcoin layer one in some sort of regular fashion, that's just one tool in the arcade system.
• 19:07
  Now, what does that mean?
• 19:10
  You know, practically, I think we got to revisit a lot of the, I mean, it's interesting.
• 19:16
  There's different ways to approach this depending on the audience.
• 19:19
  But there's kind of quite a bit of artifacts from older Bitcoin days where people were exploring some of these programmable directions using cosigner systems.
• 19:36
  like Arcade, right?
• 19:38
  Because when you think about what Arcade is,
• 19:41
  fundamentally, just not even getting into the technology,
• 19:44
  it's really just like an architecture
• 19:46
  where you have a client and you have a server.
• 19:49
  And this idea of a client and a server
• 19:51
  and the server being this entity
• 19:54
  that coordinates Bitcoin transaction with the client,
• 19:59
  the client being a wallet or mobile phone
• 20:02
  or whatever that might be,
• 20:03
  um you know it's um something that from the perspective of 10 years or plus ago
• 20:11
  uh people uh block stream and people at um specifically green the green wallet green
• 20:16
  address originally was um you know i i went i actually went a couple of weeks ago again to look
• 20:24
  back on it in the context of the cold card story but green address in 2014 was advertising a system
• 20:31
  where you had some sort of 2FA type of co-signing service
• 20:38
  using the green address server.
• 20:40
  So you would create a two of two multi-signature system, right?
• 20:44
  And it would require obviously the signature of
• 20:50
  green address the server to be able to just send transactions around from the wallet
• 20:58
  on which the coins were deposited and what this allowed green address to do was
• 21:05
  to tell users well you know what we can do is we can enforce daily spending conditions on your coins
• 21:11
  So as you register your wallet and we create your account effectively, we can set a certain threshold at which point me, myself, as the green address, as the server, I'm going to refuse signing transaction beyond that threshold.
• 21:32
  Say within the daily threshold or monthly threshold, whatever that might be.
• 21:37
  And so, you know, those were very early days. And obviously, I think in 2014, that kind of use case didn't make sense, if only for the fact that, listen, like we've made a lot of progress in terms of Bitcoin commerce and Bitcoin daily payment usage in the last, again, 10 years.
• 21:55
  But still, it hasn't grown that much
• 22:00
  And so imagine putting yourself back in 2014
• 22:05
  Who really had a use for that type of spending policies for their wallet?
• 22:11
  So it's a feature that was ahead of its time
• 22:15
  And on top of that, they had
• 22:17
  So how does the model work?
• 22:20
  Well, obviously, if the server doesn't accept to co-sign your transaction for no reason at all, then you had a backup transaction, which you could actually broadcast.
• 22:35
  So, you know, it was kind of like a de-correnting time lock or some type of lock time where you could just then broadcast your transaction yourself, wait a little bit until a certain amount of blocks, and then you just get your money back.
• 22:47
  So you never had to trust that server with your money.
• 22:51
  That server was never able to spend any of your Bitcoin, but that server provided, you know, very valuable services.
• 22:59
  And, you know, again, that model kind of like fell out of favor, I think, during this last era.
• 23:08
  and part of the reason why that is so and I really believe is that people had started having
• 23:15
  this obsession for like solutions that were completely trustless right or that appeared
• 23:19
  completely trustless right it's like why are you going to have a single server where when you can
• 23:25
  have the lightning network right this super internet like distributed system where nobody's
• 23:32
  in control. Nobody ever holds the funds. And there's not a central node that can actually
• 23:39
  sort of censor you, right? At least the idea is, well, you want to route some payments through
• 23:45
  Lightning. Most likely, you're going to find a node that's willing to do that because there is
• 23:50
  a shit ton of node on Lightning. And so we dedicated all of our resources to this.
• 23:57
  and then I think without
• 24:00
  kind of quite being aware of it
• 24:03
  we've walked this back
• 24:04
  you know over the years
• 24:07
  by
• 24:08
  reshifting the model towards
• 24:11
  client servers right
• 24:12
  what is an LSP
• 24:13
  it's the realization that well actually no
• 24:16
  it's like the notion that nodes
• 24:18
  you know everyone was going to have their lightning node
• 24:20
  didn't really make sense
• 24:22
  so we shift the burden back on the server
• 24:24
  and you know
• 24:26
  I think the, you know, the ultimate kind of full circle moment are systems like ARK inspired,
• 24:34
  find server implementation, and ultimately what Arcade does. And this is why I'm super excited
• 24:40
  about the latest work that we're doing, because it's been hard for people to understand when I
• 24:49
  tell them, listen, Arcade is not about ARK, it's not about sending transactions around. And, you
• 24:54
  I've been telling everyone, yeah, it's programmable money.
• 24:56
  You can build a bunch of script and whatever, and you can do some cool DeFi shit.
• 25:03
  And that's all well and fair, and that's going to play out as the market timing requires it.
• 25:12
  But now with the cold card event, there is kind of the opportunity to create one of the first showcase of why Arcade is more than just fast and cheap payments like Arc.
• 25:28
  um so you know we can talk about what i'm up to with that with the vaults and uh some of this
• 25:34
  stuff but um yeah i mean this is something that i've been spending all of my uh awake hours on
• 25:42
  the last literally in the last week and uh obviously this is all accelerated by by things to ai
• 25:48
  yeah not signed into x on this browser because i only used to come for this but uh uh let's let's
• 25:57
  dive into what this is and how arcade enables this and so it you vibe coded this over the last few weeks yeah it on new vnet it very very rudimentary but just i think it trying to get the idea out
• 26:11
  there of what's possible it's very experimental uh obviously it's like i pointed out it's not code
• 26:17
  that's been reviewed by any actual developer um but i think one of the for me one of the fascinating
• 26:24
  breakthroughs in the last couple of months in terms of AI development and vibe coding and
• 26:31
  whatever is the ability for the codex or cloud or whatever it is you use to nail front ends,
• 26:38
  because it gives myself, people that are not developers, the ability to kind of actually
• 26:44
  materially visualize and have kind of this tangible interface with what I'm vibe coding.
• 26:52
  because, you know, if I'm vibe coding something
• 26:54
  and it's all through GitHub
• 26:55
  and it's all just like a bunch of lines
• 26:57
  and it's just a bunch of documentations,
• 27:00
  then I'm kind of like, you know,
• 27:01
  it really is the blind leading the blind here.
• 27:05
  But in the last couple of months,
• 27:08
  Codex specifically has just been,
• 27:10
  I mean, it's absurd the quality of front end
• 27:13
  that it's able to deliver now.
• 27:16
  And so what I did for this is I just had it
• 27:19
  essentially fork our existing arcade.money application,
• 27:25
  which is something that we've built internally,
• 27:27
  which is something that is, you know,
• 27:29
  programmed by developers
• 27:31
  and has been around for several months now,
• 27:35
  if not a year already.
• 27:36
  And I just told it like, listen, take that,
• 27:38
  use that as a shell.
• 27:39
  And then we're just going to take the internals
• 27:42
  and swap them out.
• 27:43
  And instead of being a spending wallet,
• 27:46
  it's going to be a Bitcoin vault, right?
• 27:48
  So what's the idea of a Bitcoin vault? Well, again, I think the closest approximation that we can give at the moment or at least something that was very inspiring to me was Bitkey and the model that they've kind of pioneered for self-custody, meaning that obviously, you know, they've built a fantastic hardware device, especially the latest version with the screen and the ability to not do blind signing.
• 28:18
  effectively but just the quality of the whole UX speaks for itself I would recommend anyone to just
• 28:24
  give a shot give a try to to BitKey but ultimately the thing with BitKey right is that it relies on
• 28:32
  a server that server acts as a co-signer and it is able to enforce spending policies in the same way
• 28:38
  that the green address wallet from 14 years from 12 years ago was able to do that it was able to do
• 28:47
  And the reason why that's also important is that, again, not only can you rely on the service providers, in this case, BitKey to set spending policies on your funds, you can use them as some sort of intermediary to recover your funds if you have an incident.
• 29:12
  So if you use your phone, so the device with which you've set up your BitKey wallet, then the BitKey server has a key and you also have the hardware wallet that has a key.
• 29:31
  And then therefore you can kind of reinitialize everything and boot up a new phone and get back to fully managing your wallet.
• 29:41
  Same thing happens if you've lost your hardware.
• 29:44
  So there's all kinds of scenarios here where, again, introducing a bounded sort of service providers allows you much more latitude and much more freedom in terms of how you interface with your money, in terms of how you secure your money.
• 30:01
  And so that was the idea for the Arcade Vault is, okay, well, BitKey offers that.
• 30:08
  They have fully open source code.
• 30:11
  to their credit, and you can do something yourself
• 30:16
  with BitKey's code that probably comes close
• 30:19
  to what I'm doing.
• 30:20
  But my idea was, well, what we've done with Arcade
• 30:24
  is fundamentally a two of two signing server.
• 30:29
  Originally it was built to be able to facilitate
• 30:31
  the transition of users from on-chain UTXOs
• 30:35
  onto off-chain virtual UTXOs
• 30:37
  for the purpose of using off-chain payments, right?
• 30:41
  And to be able to use kind of like this ARC batching module, if you will.
• 30:46
  But the pieces are there and the pieces are modular enough
• 30:49
  where for the last couple of months,
• 30:52
  we've spent a lot of time designing something that is versatile enough,
• 30:57
  whereas you're not forced to use Arcade in an off-chain environment.
• 31:04
  You can use Arcade as a sort of,
• 31:09
  I been calling it like a smart signer right And the idea is when you think about wallets and key security and key material and the way that this is being handled nowadays
• 31:27
  regardless of your setup, whether it's collaborative, multisig, multisig that you implement yourself,
• 31:32
  Every key is controlled in a very discretionary way by some sort of entity that'll decide whether they sign off on a transaction or not.
• 31:44
  Whereas what Arcade is becoming is kind of a software signer, right?
• 31:51
  And what Arcade will do is that you will set some sort of template of policies.
• 31:59
  And what you'll kind of negotiate with Arcade is that, well, I'm going to include you in the multi-sync set that I've created here as a sort of coordinating signer.
• 32:12
  And your job will be to only sign.
• 32:14
  And so this is where it becomes really important, right?
• 32:17
  But the distinction is that it's not whether or not you're going to sign a transaction, but like what transactions are you going to sign?
• 32:25
  What is the context?
• 32:26
  What are the policies?
• 32:27
  And this is made available by additional logic.
• 32:30
  So I imagine you've had like ARB doubt on your show talking about SIG Bash.
• 32:36
  SIG Bash, yeah.
• 32:37
  Exactly.
• 32:38
  And some of that.
• 32:39
  So those are kind of very converging ideas.
• 32:43
  We're certainly not the only ones that have had this idea. Obviously, Rob Hamilton with AnchorWatch. I mean, Rob is one of the, I would say, pioneer of commercializing this solution, right? In the sense that you look at AnchorWatch and the reason why they're able to offer this institutional insurance policy is because they insert themselves into the multi-signature process of the client in a way that AnchorWatch can never spend their client on the client.
• 33:13
  money, but in the way that AnchorWatch can act as some sort of fail-safe signer in terms
• 33:20
  of, you know, depending on vast scenarios of outcomes that I'm sure you diligently sort
• 33:27
  of set with them in, you know, when you begin your policy.
• 33:35
  And so, you know, Arcade Vault is simply just an experiment in terms of saying, okay, what
• 33:39
  if we use Arcade as a module to become a generalized cosigner?
• 33:43
  And then we give any entity the ability to offer this type of service to their customers.
• 33:52
  So think if you're a river, think if you're a bull Bitcoin and you have clients that obviously you want to steer them towards self-custodial setup.
• 34:03
  But you know that in the reality that we live in, they might not be able to fully.
• 34:10
  It's just they're not going to be able to do it alone.
• 34:12
  And they should not be doing it alone, right?
• 34:15
  I think that's kind of the lesson again, right?
• 34:17
  It's like you're not sending people.
• 34:20
  And what I mean by doing it alone is like helping them secure with self-custody.
• 34:26
  It's pretty clear that it's not just sitting down with, you know, your cousin and teaching him how to use a cold card and sending him on his merry way.
• 34:36
  That's kind of a recipe for failure.
• 34:39
  One of the most jarring kind of, I think, other observation from the cold car incident was the amount of people that actually, fortunately, still had their funds around, but had to scrap into recovering their money and transferring them to a new wallet.
• 34:59
  And people did not even know how to use, you know, people had presumably hundreds of thousands of dollars on these devices.
• 35:10
  And I was listening to Spaces and I couldn't believe what I was, you know, like literally people were being held through transferring their funds off of cold card.
• 35:20
  And it's very clear that they had no idea what it is that they were doing.
• 35:24
  And so you got to ask yourself, like, how were they put in this situation to begin with?
• 35:30
  Right.
• 35:31
  Like, this is kind of where something really went wrong.
• 35:33
  And so, yeah, Arcade Vault kind of like just is kind of an attempt to open up the design
• 35:40
  space.
• 35:42
  And hopefully someone picks up.
• 35:43
  I mean, we're going to be pushing in this direction and probably, you know, polish a
• 35:50
  lot of what I've been or maybe scrap entirely what I've been doing and start.
• 35:54
  with some sort of at least, you know, legitimate developer foundations on it.
• 36:03
  But the fact of the matter is it still works.
• 36:07
  The application actually, I'm playing around with it at the moment.
• 36:11
  And it really opens up something where, you know, Bitcoin companies that have built a
• 36:18
  certain reputation in the last decade that are still around and that deserve this reputation they should be able to offer financial services to their clients beyond just buying selling Bitcoin in custody And I think the future is going to be that there will be these types of
• 36:37
  collaborative custody model because for any company that has a bit of foresight, especially
• 36:43
  in the context of, you know, potentially changing regulatory environments over the rest of, you
• 36:49
  know, this decade and just liabilities.
• 36:55
  If you're able to offer consumer experience that rival custody in terms of, you know,
• 37:04
  reliability and just the assurance that you can sleep tight and that your money is safe
• 37:09
  because and how does that happen is, yes, you have control of your money, but there's
• 37:15
  also someone that is a professional that has the vigilance of having an architecture in place that
• 37:21
  spends 24-7 making sure that your money is not moving without you making the explicit approval.
• 37:32
  And in the event that it does, you have sort of, you know, vault type. What is the idea of the vault
• 37:38
  is that if some unforeseen transactions hits the network
• 37:44
  because a certain device that's part of a multisig was compromised
• 37:48
  or part of a signing setup,
• 37:50
  ideally, the promise of the vault is that you can kind of interfere
• 37:55
  with this fraudulent withdrawal
• 37:59
  and eventually funnel the funds into a backup location,
• 38:04
  which was kind of previously agreed upon.
• 38:06
  Now, the idea with vaults, or at least the common vault design, obviously involves covenants.
• 38:14
  Or at least, you know, it's a bit of a difficult conversation because, you know, vaults are such a general term.
• 38:21
  And when you ask anyone in crypto, you know, outside of Bitcoin, vaults mean a completely different thing.
• 38:27
  It doesn't have anything to do with security.
• 38:29
  so it's a little hard for me when I when when people are like okay well no vaults are
• 38:35
  consensus enforced covenants setups that are completely trustless it's like well yeah maybe
• 38:42
  that's one design but now that we don't have and we continue to buy shit covenant and maybe that's
• 38:48
  going to be the next thing that continues tripping away at kind of like the social tissue of bitcoin
• 38:53
  well in the meantime covenants are just you know whatever covenants are just contract and
• 39:00
  again the idea as well the idea with arcade as well turns out you can't enforce contracts between
• 39:05
  multi-parties without having to require bitcoin consensus to interject here obviously if you need
• 39:13
  the most trustless type of execution then you're going to need bitcoin but
• 39:18
  like Bitcoin might not apply to, you know, to to to every single financial arrangement and shouldn't.
• 39:29
  So freaks, this was brought to you by our good friends, the Bitcoin team at Block,
• 39:33
  Bitcoin at Block. Bitcoin was originally introduced as a peer to peer electronic cash system. We all
• 39:37
  know that the idea wasn't just a buy and hold to use Bitcoin and many different facets. Holding is
• 39:41
  one facet. Spending receiving is another facet. I've been working towards this for 17 years and
• 39:45
  block has put together a whole ecosystem of companies to really bring about the reality of
• 39:50
  Bitcoin as everyday money. So with Cash App, you can buy, spend, send, receive, earn, get paid in
• 39:55
  Bitcoin. When you're trading over $2,000, you're not going to pay any fees on that. So then there's
• 40:00
  Square, which is helping bring Bitcoin into real world commerce. Eligible businesses can accept
• 40:04
  Bitcoin payments over lightning with settlement in seconds and 0% processing fees. Customers can pay
• 40:09
  with Bitcoin or even dollars over lightning through Cash App and help local businesses keep
• 40:12
  more of every sale. It's making Bitcoin payments feel as easy as tapping a card. And when you're
• 40:16
  ready to hold Bitcoin, there's BitKey. It's a Bitcoin wallet built for real life. No seed phrases,
• 40:21
  it's a two or three multi-sig, so no single points of failure. It has recovery and inheritance
• 40:25
  processes and capabilities built in. And the new BitKey device includes a screen so you can verify
• 40:30
  important actions directly on the device before approving them so you know exactly what you're
• 40:34
  authorizing. So together, Cash App, Square, and BitKey are helping make Bitcoin easier to earn,
• 40:37
  save, spend, and secure. It's a simple idea. Bitcoin works when people can actually use it
• 40:41
  in the way that they want. That's Bitcoin at Block. This episode is sponsored by Bitcoin at Block,
• 40:45
  including Cash App Square and BitKey. To learn more about Bitcoin at Block and access the latest
• 40:48
  offers from Cash App Square and BitKey, check out the links and the promo codes in the episode
• 40:51
  description. Bitcoin services are provided by Block Inc. Bitcoin services are not licensable
• 40:55
  activity in all US states and territories, and not all services are available in all states.
• 40:58
  BitKey is not available in New York. Block Inc. operates in New York as Block of Delaware and is
• 41:02
  licensed to engage in virtual currency business activity by the New York State Department of
• 41:06
  Financial Services. Bitcoin is a non-deposit, non-bank product that is not FDIC insured and
• 41:10
  involves risk, including monetary loss. For additional information, see the Bitcoin disclosures
• 41:14
  in the show notes.
• 41:15
  Freaks look at me, I'm glowing. I've got like an angel's halo going around me. You know why
• 41:19
  that is? I feel good. I feel taken care of. I feel blessed, healthy, happy. And that is
• 41:24
  because I'm a CrowdHealth member. My family and I have been CrowdHealth members for five
• 41:28
  years now, literally this month. Five years ago, we joined CrowdHealth. We've had two
• 41:31
  babies. We've had multiple health events and we're never going back to health insurance.
• 41:36
  CrowdHealth is crowdfunded healthcare. So you sign up for CrowdHealth, you pay
• 41:39
  a monthly fee, you help out with other people's bills, and it's significantly cheaper than health
• 41:46
  insurance. We were on COBRA as a family of three. When I left my last job before I went full-time
• 41:50
  with CFTC, went on the crowd health. Now as a family of five, we pay, I believe, $700 a month.
• 41:58
  It's significantly cheaper. They're going to negotiate prices lower for you. They've consistently
• 42:02
  negotiated healthcare prices as much as 50, 60, 80% in many cases. They help out with babies.
• 42:08
  If you have a pregnancy, you pay the first $3,000 and the crowd covers the rest.
• 42:13
  If you have a regular health event, you pay $500 and the crowd pays the rest.
• 42:18
  Go to joincrowdhealth.com.
• 42:19
  Sign up today.
• 42:20
  Use the code TFTC.
• 42:21
  Opt out of health insurance.
• 42:23
  I'm an insured baby and I love it.
• 42:24
  Use the code TFTC at joincrowdhealth.com.
• 42:27
  You'll get $99 a month for the first three months that you're on the CrowdHealth platform
• 42:31
  in the community.
• 42:33
  Bitcoiners, you found sovereign money.
• 42:34
  Now find sovereign health and sovereign healthcare.
• 42:37
  No. And so you have it. Let's just walk through how you're doing this with the vault demo or prototype.
• 42:44
  Yeah, yeah.
• 42:45
  Because you say concretely, the money sits in two taproot trees. You have a spending tree and a savings tree.
• 42:51
  And so it seems like there's some conditions there.
• 42:53
  Yeah, exactly. Yeah. So there's you have two accounts because what this turns out to be is, you know, is effectively a Bitcoin account.
• 43:00
  You know, in Bitcoin, we don't have accounts. Normally, everything is UTXO and coin selection based.
• 43:06
  But when you start having a setup like this, where you can actually enforce spending policies, it's not looking like accounts.
• 43:13
  And the spending account is your weekly allowance, call it like that, or your daily allowance.
• 43:19
  It's what the budget that you set with the server.
• 43:23
  When you initiate the wallet, it'll tell the server, well, again, anything that is beyond 100,000 sets a day, just don't co-sign that transaction.
• 43:32
  but what you also do is you kind of move that you know you make it so that you move your budget
• 43:38
  to that so you actually you can move more you can move 500 000 sats to your spending account
• 43:45
  and then you can say well i can only spend a hundred thousand uh dollar and that's that's one
• 43:51
  part of the taproot three and that taproot three is effectively co-signed by your mobile device
• 43:59
  by way of pass keys.
• 44:02
  So that's another feature that I haven't discussed.
• 44:05
  But again, the idea was to try to replicate the BitKey model.
• 44:10
  So there's no seeds here.
• 44:11
  Everything is very kind of like native via pass key, face ID, and all of these.
• 44:16
  Again, not for everyone maybe, but from a UX perspective,
• 44:20
  it's kind of like the bar that you need to try to set for modern applications, I think.
• 44:26
  And so your mobile phone has a key.
• 44:29
  the vault. So what you do when you deploy an arcade vault is it's a third party that deploys
• 44:35
  a vault, right? So right now, you know, I have an instance, a server implementation that runs on the
• 44:41
  cloud, you know, that operates the vault. But the idea again, in the future is like River,
• 44:47
  they operate their own vault and then River becomes the cosigner to your wallet transaction,
• 44:55
  your spending wallet transaction.
• 44:58
  And on top of that, well, Arcade is kind of like the global coordinator.
• 45:02
  So Arcade also exists in that multi-sync set.
• 45:05
  So it's a three of three.
• 45:07
  It's an end of end transaction.
• 45:09
  Everyone needs to sign in order for spending, you know,
• 45:14
  checking account transactions to go out.
• 45:17
  And the important thing, again, is if in the event that the third party vault operator doesn't sign or that arcade doesn't sign, then you have another spending condition that exists in this initial vault contract that you set up, which is, well, me and my phone, sorry, my phone, my device and the hardware wallet that I tie up to it.
• 45:47
  we can use our signature to move any amount of money out of that vault that we want right so
• 45:54
  if all of these servers decides to go offline i still have access to money i to my money i don't
• 46:00
  depend on those servers uh to be able to uh to spend my money but now again what happens if the
• 46:09
  vault server is compromised well they still require your mobile uh phone signature it still
• 46:16
  requires the Arcade signature as well.
• 46:20
  So Arcade will actually evaluate
• 46:22
  and enforce the policies itself,
• 46:26
  meaning that even if the vault operator gets compromised,
• 46:29
  tries to broadcast a transaction
• 46:31
  that bypasses the limit,
• 46:33
  Arcade will catch it.
• 46:35
  Obviously, yourself as a,
• 46:37
  well, perhaps even your phone
• 46:38
  might be compromised as well.
• 46:43
  But the whole thing is set up so, right,
• 46:44
  that like there's no single point of failures.
• 46:47
  There are like the worst case scenario again in that kind of setup is that your spending wallet gets drained because effectively you you know your phone been stolen
• 47:00
  And listen, fucking the arcade cosigner is compromised and the vault cosigner is compromised.
• 47:09
  And if that happens, well, you're spending, you know, your checking account gets drained and it sucks, but your life savings are still safe.
• 47:15
  Because you're in the taproot path of your life savings, there is never a signature from these other third parties, right?
• 47:25
  It's only hardware and mobile and your phone.
• 47:31
  And then there are, you know, for example, there's a policy, which is your hardware wallet alone, right?
• 47:38
  so if you lose your phone if the arcade server is offline and whatever you can broadcast a
• 47:44
  transaction that'll spend from your hardware wallet but it'll just take more time right it's
• 47:48
  going to have a long delay and what is the purpose of that well the purpose is that if your hardware
• 47:53
  wallet gets compromised and just in the same way that cold card did and someone tries to drain the
• 47:59
  wallet well you um you'll have the opportunity to kind of claw back that that that that transaction
• 48:07
  You'll have the opportunity to stop the attacker from being able to drain your entire wallet.
• 48:14
  So, and, you know, the parameters that I've set here in the Tread example, they're not prescriptive.
• 48:24
  In fact, they use kind of mutiny net settings.
• 48:28
  So, you know, talking about like six blocks, which in mutiny net is literally like a couple of minutes.
• 48:34
  so they're not to be used in production
• 48:37
  but again, it's a blank canvas
• 48:39
  for people to be able to deploy
• 48:41
  and I'm very, very interested
• 48:44
  to see people
• 48:46
  because I think honestly
• 48:48
  it's kind of a
• 48:49
  it'll become a no-brainer
• 48:51
  it'll require people to harden their infrastructure
• 48:54
  and this is kind of what people
• 48:56
  are doing at the moment, right?
• 48:58
  So it's hard to be pushing
• 49:00
  new features right now
• 49:02
  because everyone's busy stopping fires.
• 49:06
  But eventually, we'll move past that phase.
• 49:10
  And I think hopefully customers start kind of like,
• 49:13
  users start demanding better solutions.
• 49:17
  And I think that's just a very promising one.
• 49:20
  And it's one that in the spirit of what we're doing with Arcade
• 49:23
  is completely open source, completely modular.
• 49:28
  It doesn't really...
• 49:29
  If you've heard about ARK and you don't like this idea of virtual UTXOs and expiring funds and all that jazz, you don't have to.
• 49:41
  It's perfectly voluntary.
• 49:43
  And that's the beauty of it.
• 49:44
  I think the direction we're going into with Arcade is that we want to have voluntary systems where we know that working within consensus is extremely hard.
• 49:54
  We've had like a great example of the difficulty of working within consensus in the last couple of weeks.
• 50:02
  And it'll become increasingly hard.
• 50:05
  So but people still need to have the optionality to be able to kind of deploy, you know, the sort of commercial policies and sort of just using Bitcoin in the way that they see fit without having to ask permission is ultimately what it's all about.
• 50:21
  Right.
• 50:23
  Yeah, I think
• 50:25
  I definitely want to touch on the AI stuff because it wasn't an announcement
• 50:28
  this morning.
• 50:28
  I know you guys are doing stuff with Breeze in that capacity.
• 50:32
  But before we go there, I think just like really leaning into the
• 50:36
  sort of breadth of the aperture of potential use cases here with the arcade.
• 50:42
  Set up.
• 50:43
  I mean, you guys launched
• 50:45
  Intense to Arcade Intense in the aftermath of bolts going down.
• 50:50
  And I think it highlights just another sort of mechanism that the ARK transaction batching protocols can enable.
• 51:00
  Because Boltz was one of the most dependable and relied upon swapping services between Lightning and other aspects of Bitcoin, whether it's on-chain, liquid, whatever it may be.
• 51:13
  And they succumbed to the attacks of AI black hats disrupting their systems.
• 51:21
  And I think we learned a lesson that that was another sort of central point of failure or maybe not a point of failure.
• 51:28
  I don't believe any funds were lost on customers behalf, but it was a central location point in that part of the industry with the swapping services.
• 51:38
  And I think what you've done with Intense is really creative.
• 51:43
  And so you sort of open the marketplace for potential providers that could execute those swaps by using ARC to create a, um, a swap message board for lack of a better term or order book.
• 51:57
  The order book, exactly.
• 51:58
  It effectively sort of like an order book And you know the idea the idea behind this and the idea behind a lot of what we doing with Arcade is simply to standardize practices right So again with Arcade
• 52:15
  you have a toolkit where you can build the most basic multi-sigs up to the most complex
• 52:21
  contract you can imagine. But the reason why we believe it has a lot of value is ultimately that
• 52:28
  If you create standards that other people can build around, well, everything becomes interoperable, right?
• 52:36
  And therefore, any market participants, provided they agree, it's kind of like Bitcoin's consensus rules, right?
• 52:41
  It's like, why was Bitcoin valuable?
• 52:43
  In part, it's because everyone agrees on this set of rules and everyone coordinates around those rules.
• 52:49
  So what we've done here is, well, we've had these ideas of smart contracts and everything for so many years around Bitcoin.
• 52:58
  you know, it was never not formalized or specified, but kind of like we needed to create a common interface for market actors to be able to use these contracts on Bitcoin, as limited as they may be.
• 53:15
  So the idea with Intense indeed is, well, you know, Bolts was the gold standard for Lightning swaps or for just not Lightning swaps, but kind of like being this bridge, this non-custorial bridge between every sort of layer of Bitcoin applications and services.
• 53:33
  but unfortunately you know their their dominance for lack of a better word and I think kind of like
• 53:40
  just like I said the the immaturity of the technology stack made it so that it was hard
• 53:49
  to keep up with them and they were kind of like pushing the boundaries and they were like doing
• 53:53
  things where people at the time up until now maybe that AI tools are picking up you know it was like
• 54:01
  You needed to have significant expertise with Lightning to be able to operate a node of that importance.
• 54:09
  And a lot of people were willing to or just had the time to dedicate into deploying that kind of architecture.
• 54:16
  And for that reason, and probably just a combination of smaller market, just generally,
• 54:25
  you know, most is doing great, but ultimately, you know, it's still a relatively small market and
• 54:31
  perhaps new entrants didn't see enough incentive to enter it. But it's a chicken and egg problem,
• 54:39
  right? So at least what we feel like we've done with Arcade Intense is we've solved one part of
• 54:43
  that issue, which is that you don't have to be a professional sort of swap operators to be able to
• 54:51
  serve Lightning swaps.
• 54:56
  And one of the reasons why that's super important is because
• 54:59
  unfortunately, central to the vulnerability of Bolts and the reason why they took this
• 55:09
  sort of blunt force attack from AI attackers the moment that these new models were out
• 55:15
  is that the nature of the model made it so that they advertised themselves as a central
• 55:20
  API that were processing all of these swaps. And obviously a lot of their code was all open source.
• 55:27
  And so it immediately puts a massive target on their back. Right. And it tells every, and you
• 55:33
  know, listen, Bolts, obviously we know it was not the only one, but there's a certain class of
• 55:40
  services in the Bitcoin and wider crypto ecosystem that were specifically targeted by those systems.
• 55:48
  And those are all bolts type service, meaning services that were able to basically that we're just providing swap services or trading or services between coins, chains or whatever that is, because those were the ones.
• 56:03
  LMP2P, they had to shut down as well.
• 56:05
  LMP2P.
• 56:06
  You know, I know that the people at Garden Finance have had some had some issues.
• 56:12
  I think fortunately no funds was lost there.
• 56:16
  Zeus.
• 56:17
  Zeus.
• 56:17
  Zeus.
• 56:18
  Zeus as well. So, you know, even, you know, even beyond in the wider crypto world.
• 56:25
  And so, you know, it's very clear, it's becoming therefore very clear that the architecture that's required for these types of service to operate nowadays needs to be very different from what it was before.
• 56:40
  And so what we've done with Intense is kind of leverage the architecture of Arcade and create a RFQ model.
• 56:48
  So a request for quote model where you have a marketplace of users and service providers that are effectively advertising.
• 56:58
  The service providers are advertising different services.
• 57:02
  One of them, the leading one at the moment is Lightning Swap Services, right?
• 57:08
  And they advertising via different rates They might be you know right now it very primitive again you know we just launched it It kind of a developer preview So the full vision hasn materialized yet But the idea is
• 57:23
  you'll have, you know, potentially other companies, you could have a voltage, you could have a Zeus,
• 57:29
  you could have an Nambos directly running. What does it mean for these providers is they're
• 57:37
  running a small little server that's attached to their lightning node it's a very lightweight
• 57:44
  sort of interface and it's not something that they need to advertise publicly to everyone
• 57:51
  they don't have to go out on the open internet and say hey this is the end point by which i'm
• 57:56
  serving lightning swap and if you can figure out how to drain it you know it's all yours
• 58:01
  Rather, then what happens is we have a service by which we coordinate the users that are requesting those swaps.
• 58:12
  So the wallets, the wallets are able to select from a provision of service providers and say, OK, this is the list of service providers that I want to process the Lightning transactions for my application.
• 58:25
  and what I want to be able to do is whenever the user requests an action,
• 58:32
  this is where the intent word comes from.
• 58:36
  The intent is an action.
• 58:37
  The intent is the user saying, I have 100,000 SATs.
• 58:41
  I want to pay this 100,000 SAT invoice.
• 58:46
  Find me the best Lightning provider.
• 58:50
  and by way of the intent protocol,
• 58:54
  the wallet will ping all of these providers,
• 58:57
  check what are their current prices
• 58:58
  and return that to the wallet.
• 59:01
  The wallet will select one,
• 59:02
  will use that route.
• 59:05
  Again, everything is atomic.
• 59:06
  Everything happens via HTLC swaps.
• 59:09
  So there's no chain of custody here.
• 59:12
  But the important thing is
• 59:13
  you have a market, right?
• 59:15
  You have something where
• 59:17
  for the applications,
• 59:19
  it's extremely important because every single sort of financial service right now,
• 59:25
  especially crypto financial services, are sort of orchestrated in a way where,
• 59:33
  you know, you want to build an application, you want to build a wallet
• 59:35
  where users are interfacing with your product.
• 59:39
  You're going to have a single provider for Lightning payment.
• 59:41
  You're going to have a single provider for, you know,
• 59:45
  those that are interested in swaps to stable coins.
• 59:49
  You're going to have a single provider for that.
• 59:51
  And how does it work, right?
• 59:53
  It's all APIs and you're programming an integration from your wallet to these APIs.
• 1:00:00
  And then every time you want to add a provider, you need to add another API integration.
• 1:00:06
  And then, you know, so there's a bit of a limit there, right?
• 1:00:09
  You're kind of duplicating a lot of work.
• 1:00:12
  Whereas with Arcade Intents, the idea is you have a single interface and you create a market for all of these swaps, whether they're Lightning swaps, whether they're stablecoin swaps, and eventually, you know, tokenized asset swaps, whatever people fancy, that'll be available.
• 1:00:31
  but the important thing is providers can directly tap into these wallets.
• 1:00:38
  So from the provider perspective, if you're a Lightning service provider that wants distribution,
• 1:00:43
  you can either go to every single wallet in the ecosystem and say,
• 1:00:46
  hey, I have extra liquidity. I don't know what to do with it.
• 1:00:49
  Maybe you guys want to integrate, you know, maybe you guys want to use my routes
• 1:00:54
  to be able to send payments via Lightning.
• 1:00:56
  Well, that was impractical and that remains impractical, right?
• 1:01:00
  It's kind of for many different reasons, from a technology perspective to a business development perspective, that type of thing is not sustainable.
• 1:01:11
  But now you have the opportunity where if users are using the arcade, you know, if wallet applications and operators of these services are using Arcade Intense, the provider can only only has to advertise their service via Intense.
• 1:01:28
  And then they can be matched with users depending on the application need.
• 1:01:33
  So it's very early, but we think it's a massive, it has a lot of opportunity because again, it just comes down to we're going to a direction where you cannot rely on a single service provider.
• 1:01:50
  The single points of failures are obvious.
• 1:01:54
  And if you would have told us a year ago that we should have made contingency plans for
• 1:02:02
  bolts because maybe someday they were going to go out of service, we would tell you, man,
• 1:02:07
  what are you talking about?
• 1:02:08
  They've been running for I don't know how many years.
• 1:02:11
  They have a perfect track record.
• 1:02:14
  We're not going to lose sleep over this.
• 1:02:16
  Well, you know, we're living in a world of black swans and where everything is accelerating.
• 1:02:21
  And so it's become very clear that whatever is the architecture, whatever is the infrastructure you're building, you want to make it resilient to.
• 1:02:30
  the failure of a single party.
• 1:02:33
  And so you do that by creating markets.
• 1:02:35
  And so that's kind of like what Arcade Intense does
• 1:02:38
  is it allows applications and financial services
• 1:02:44
  to plug into Bitcoin-native financial markets.
• 1:02:49
  Yeah.
• 1:02:50
  I mean, what's beautiful about this too,
• 1:02:52
  anybody listening is like,
• 1:02:53
  oh my gosh, am I going to have to do all this?
• 1:02:55
  No, the service providers are.
• 1:02:56
  This will be all abstracted away for a net user.
• 1:02:59
  They'll just see.
• 1:02:59
  Well, exactly.
• 1:03:00
  And they'll pay and they'll get paid.
• 1:03:03
  Exactly.
• 1:03:05
  It's actually much more intuitive.
• 1:03:07
  It's actually just something that is going to significantly improve the user experience for the end user.
• 1:03:18
  And specifically, prices.
• 1:03:20
  I think one of the things we didn't discuss is that intents also potentially apply to credit markets.
• 1:03:26
  So what if you get to a point where you want to get a Bitcoin back loan and rather than negotiating the terms with a single service provider and going to shopping around to the LIGOs, the Latin and every single one of them?
• 1:03:43
  Well, if every one of them uses a standardized protocol, a standardized market interface, well, the wallet can integrate all of them and it'll give you the best rate depending on your loan request, right?
• 1:03:56
  And I think this is kind of the natural way for things to evolve is you want to remove those silos because, yeah, they're just not competitive.
• 1:04:13
  First off, the idea is also you want to create a competitive market.
• 1:04:17
  It's going to give better prices to users, better experience.
• 1:04:21
  But you also want to avoid censorship as well.
• 1:04:24
  You want to avoid a single, what if a certain service provider or certain lightning swap provider decides that they need to censor payments going to an address or whatever that is.
• 1:04:39
  You want to have it built into your application so that you can route around this.
• 1:04:43
  So freaks, when you take Bitcoin seriously, you start with custody.
• 1:04:46
  You want to control your keys, avoid single points of failure, and make sure your savings cannot disappear because you or someone else screwed up.
• 1:04:52
  That is what Unchained has been focused on since 2016.
• 1:04:54
  Unchained is the leader in collaborative multi-sig custody and Bitcoin financial services that keep you in control.
• 1:04:59
  They secure over $12 billion in Bitcoin for more than 12,000 clients.
• 1:05:03
  That means about one out of every 200 Bitcoin sits inside an Unchained vault.
• 1:05:07
  Their model is simple.
• 1:05:08
  You hold two keys, they hold one key.
• 1:05:09
  It always takes two keys to move Bitcoin, meaning their single key can't access your Bitcoin on its own.
• 1:05:14
  Just resilient, shared custody that gives you institutional-grade security while keeping you sovereign.
• 1:05:18
  Unchained also lets you trade straight from your vault.
• 1:05:20
  Access Bitcoin-backed commercial loans, open a Bitcoin IRA where you hold your own keys,
• 1:05:23
  and set up personal, business, trust, or retirement vaults.
• 1:05:26
  They even offer inheritance solutions
• 1:05:28
  built for long-term hodlers.
• 1:05:29
  Or offer the highest level private client service
• 1:05:32
  with Unchained Signature
• 1:05:33
  and get a dedicated account manager,
• 1:05:35
  discounted trading fees,
• 1:05:36
  exclusive access to events and features,
• 1:05:38
  and much, much more.
• 1:05:39
  If you want a partner that helps you secure
• 1:05:40
  and grow your Bitcoin without giving up control,
• 1:05:42
  go to unchained.com and use the code TFTC10
• 1:05:45
  at checkout to get 10% off
• 1:05:47
  your new Bitcoin multisig vault.
• 1:05:48
  That's TFTC10 at unchained.com.
• 1:05:51
  All right, freaks.
• 1:05:52
  You know, I don't take sponsor money from products.
• 1:05:54
  I wouldn't use myself.
• 1:05:54
  So listen up.
• 1:05:55
  The Avon Bitcoin Visa card is one of the most interesting things I've seen in the Bitcoin
• 1:05:58
  lending space in a long time.
• 1:06:00
  Here's the deal.
• 1:06:00
  You can get a line of credit up to a million dollars backed by your Bitcoin without selling
• 1:06:04
  a single sat.
• 1:06:06
  No games, no annual fees, no minimum draws.
• 1:06:09
  And your Bitcoin is custodied by Bitco, one of the most trusted names in digital asset
• 1:06:12
  security.
• 1:06:13
  Avon never lends it out.
• 1:06:14
  There's no rehypothecation.
• 1:06:15
  You stay in control.
• 1:06:16
  You can lock in a fixed rate for up to 10 years.
• 1:06:19
  10 years.
• 1:06:20
  That's 10 times longer than most lenders out there.
• 1:06:23
  We'll go interest only for up to five years.
• 1:06:25
  Rates start at 8.99% APR.
• 1:06:28
  For a product that lets you keep your stack and still access liquidity, it's hard to beat.
• 1:06:32
  On top of this, that's what?
• 1:06:33
  You also get 2% unlimited cash back every time you use the card.
• 1:06:37
  Spend fiat, keep your Bitcoin.
• 1:06:39
  That's the whole game.
• 1:06:40
  If you've been stacking for years and need liquidity without triggering a taxable event,
• 1:06:44
  this is worth a serious look.
• 1:06:45
  Go to aven.com slash bitcoin.
• 1:06:48
  that's aven.com slash bitcoin check it out yeah that's what i think i'm i'm looking up right now
• 1:06:55
  because i'm trying to this is stoking there's questions in my mind really like how does this
• 1:07:01
  compare to morpho just like thinking about like how this marketplace compares to like the the defi
• 1:07:08
  lending protocols and there seems to be a distinct difference um
• 1:07:12
  There is, yeah, there's a bit of, it's interesting, there's quite a bit of difference.
• 1:07:21
  There's also some similarities.
• 1:07:23
  You'll notice if anyone that pays attention to these crypto, Ethereum type of protocols.
• 1:07:30
  I'm thinking of like Morpho specifically.
• 1:07:32
  Yeah, Morpho.
• 1:07:33
  So well so you know Morpho specifically has been going into a direction where they increasingly building off of a model that is more let say order book than these pool models right
• 1:07:49
  Because we found out that a lot of these pool models are inefficient.
• 1:07:54
  They might be risky.
• 1:07:56
  they might sort of
• 1:08:00
  incur bad debt by
• 1:08:03
  you know you have these pools that are lending
• 1:08:06
  against a collection of different collateral
• 1:08:08
  and if the quality of the collateral
• 1:08:10
  isn't up to par
• 1:08:13
  which is the case for a lot of the shitcoins
• 1:08:15
  well you know you can be lending
• 1:08:18
  against your bitcoin in a certain pool
• 1:08:20
  but someone else is lending
• 1:08:22
  against their other shitcoin in the same pool
• 1:08:24
  And because that position of the shitcoin, the collateral has gone to zero, let's say, speculatively, while the pool cannot liquidate that collateral, what happens is this creates bad debt for every participant in the pool.
• 1:08:42
  So everyone kind of ends up being on the hook for that hole.
• 1:08:50
  But, you know, those are things that we're exploring.
• 1:08:54
  I mean, you know, we're talking a lot about Arcade and we're not necessarily going to go too much.
• 1:09:00
  I'm not going to go too much into details, but one of the exercises that we've also done is that R-Collabs is kind of evolving beyond just a research technology company, but more so towards what is going to start looking like a sort of financial product company.
• 1:09:16
  So our idea with R-Clabs is that we're not spending anymore our entire days building the infrastructure and building the Arcade Protocol, but rather we want to be first to market with a lot of the applications that we think are best implemented on the Arcade Protocol.
• 1:09:34
  Right. And so when you think about the standardization of credit markets or Bitcoin back loans, well, this is something that we're looking at is, OK, well, how can we build something that is going to be versatile enough that it can fit any type of custody infrastructure?
• 1:09:53
  So again, we're not prescriptive anymore in the sense of, okay, well, if you use Arcade, everything is going to be self-custodial. Actually, that's not the case. You can actually use Arcade as part of your sort of custodial service.
• 1:10:09
  But the idea is, well, anyone that operates, even a custodian needs to be managing keys, needs to be managing, you know, lending origination flows.
• 1:10:21
  There's some sort of logic, OK, tracking the collateral.
• 1:10:24
  Where does this go?
• 1:10:26
  Who are the creditors and all of that stuff?
• 1:10:28
  Well, it turns out that, you know, you can do that on a very simple database.
• 1:10:33
  But there's also the opportunity to do that by just using modern Bitcoin software that is going to be much closer to the Bitcoin and kind of just provide something that, again, lends itself more to interoperability between providers.
• 1:10:52
  So what does that mean?
• 1:10:54
  Well, it means potentially open credit markets, right?
• 1:10:57
  Where you have a custodian originator, you have a self-custodial LIGOS type originator,
• 1:11:04
  and everyone is starting to have the ability to open their books to each other, right?
• 1:11:10
  In the sense that if someone comes to you and says, I need a $100 million loan against
• 1:11:16
  my Bitcoin, maybe you don't have the lender available.
• 1:11:19
  Maybe you don't have the liquidity to serve that loan.
• 1:11:21
  but maybe someone else does.
• 1:11:24
  Now, to what extent do you want to open those books?
• 1:11:26
  And a lot of the game in lending is origination.
• 1:11:30
  So you're not going to send a potential client
• 1:11:34
  to one of your competitor,
• 1:11:36
  but there's an opportunity here to just,
• 1:11:39
  again, standardize a lot of that.
• 1:11:42
  And the other things that we're looking at
• 1:11:43
  is very sort of grounded financial primitives.
• 1:11:49
  What does it look like if you want to,
• 1:11:51
  exercise options,
• 1:11:54
  hash rate derivatives
• 1:11:55
  on top of your Bitcoins,
• 1:11:58
  if you're a mining operation,
• 1:12:00
  if you're a treasury
• 1:12:02
  that wants to hedge
• 1:12:04
  against market volatility
• 1:12:05
  or take directional position
• 1:12:07
  against market volatility.
• 1:12:10
  Again, these are not necessarily
• 1:12:11
  DeFi protocols, right?
• 1:12:13
  We don't think,
• 1:12:14
  or at least I don't think
• 1:12:15
  that what we are doing
• 1:12:18
  and what we've been needing to do
• 1:12:19
  is bring DeFi to Bitcoin.
• 1:12:21
  Rather, we need to just improve what Bitcoin is able to do.
• 1:12:28
  And just, you know, smart contracts were never about making everything trustless.
• 1:12:34
  If you look at the work of Nick Sabo, and if you go back and read a lot of these papers,
• 1:12:39
  they were simply about elevating business logic into software, right?
• 1:12:43
  And how do you create checks and balances around the handling of treasury,
• 1:12:48
  handling of budgeting of you know all these movements and flows of monies within organizations and institutions there an ability to be able to elevate that at the software layer where you have
• 1:13:02
  something that is more compatible with a native digital currency, right? But that doesn't mean
• 1:13:07
  that every single third party, you know, needs to be removed. No, I mean, actually, I like to,
• 1:13:18
  I've been thinking a lot about how to frame this.
• 1:13:20
  And I like to say that what we're doing is not decentralized finance.
• 1:13:23
  It's more like actually it's collaborative finance, right?
• 1:13:27
  It's like, hey, listen, if we get a bunch of parties together and we get them to agree on a set of rules,
• 1:13:32
  as long as everyone is collaborating with each other, the outcome is running great.
• 1:13:36
  And it's going to be very efficient.
• 1:13:39
  And we're able to do a lot more with the liquidity that we have available.
• 1:13:44
  We can open it up to new markets and find new business lines and opportunities.
• 1:13:49
  And worst case scenarios, if for whatever reason, the agreement between the parties fall apart, well, everyone has kind of like these routes back to get their money back on chain.
• 1:14:02
  No one gets to censor anyone.
• 1:14:05
  Again, it's all just voluntary, right?
• 1:14:07
  So it's kind of like this voluntary, collaborative finance.
• 1:14:11
  and I think it's a more productive way to think about it, right?
• 1:14:15
  Ultimately, like this sort of decentralized finance
• 1:14:17
  where we almost kind of disenfranchise the individual, right?
• 1:14:22
  And be like, okay, well, you can't be trusted.
• 1:14:25
  We don't want to have to trust you in any ways.
• 1:14:27
  And kind of like, actually, no,
• 1:14:29
  if you can build solid, bounded trust,
• 1:14:34
  then that is going to unlock a lot of things.
• 1:14:38
  And I think this is what we're trying to do
• 1:14:40
  on the commercial side by kind of creating commercial interface to the arcade protocol
• 1:14:47
  so that businesses that don't want to implement everything from scratch themselves can actually
• 1:14:54
  rely on our product as standards to be able to do so.
• 1:14:58
  Yeah, it ties nicely with the conversation I had earlier this week too.
• 1:15:03
  And if you can successfully implement that collaborative lending protocol, whatever you want to protocol or market, you can see a scenario where liquidity begets liquidity.
• 1:15:17
  Customers are happy.
• 1:15:18
  They're choosing it more as an option.
• 1:15:20
  And yes, competition exists and it should exist.
• 1:15:23
  And the individual actors within that marketplace should refine their systems based off of the sort of competitive feedback that they're getting and try to outcompete.
• 1:15:32
  But you can see a scenario in the long run where if you collaborate instead of trying to dominate the market, you could actually see yourself doing more origination because people just like the UX of that collaborative marketplace much better than having to go to individual operators.
• 1:15:53
  yeah i do i look at it very much kind of like we need to find a substitute for
• 1:16:02
  um the rule of law right in the digital in the digital world right where why do
• 1:16:10
  you know you can you know everyone has their own opinion but like why are u.s markets
• 1:16:16
  you know the most you know act again the most successful markets in the history of the world
• 1:16:24
  is that they're ultimately you know you can you can you can debate it nowadays but ultimately
• 1:16:30
  they're kind of held by a rule of law framework that has proven itself over um you know uh like
• 1:16:38
  more than a century now um and the underpinning of the value of a lot of these you know the value
• 1:16:44
  of private property and things like that is held by the implementation of the rule of law.
• 1:16:50
  But the rule of law only exists to a certain extent in the digital world with digitally
• 1:16:57
  native currencies.
• 1:16:59
  And what is the rule of law?
• 1:17:01
  I mean, you know, on one hand, it's the ability, yes, to throw someone in jail if it doesn't
• 1:17:07
  do what society expects them to do.
• 1:17:09
  But it's also kind of just a bit of a common framework, which we agree to operate under, which just facilitates the attraction.
• 1:17:24
  Don't tell me you're going to say it.
• 1:17:28
  Code is law.
• 1:17:30
  Code is law.
• 1:17:31
  No, I mean, no far from it.
• 1:17:33
  No, no, I mean, code is never going to be law, of course.
• 1:17:36
  So that's why, you know, you got to be very careful.
• 1:17:37
  of um but there there are there is room whereby you know it's like it's like in normal law you
• 1:17:45
  don't need to adjudicate everything in court right we people were making that uh analogy with
• 1:17:50
  lightning back in the days but it's kind of like if you have a system in place that align everyone's
• 1:17:55
  incentives in a way where the game theory is that um it a preferable for everyone right it kind like Nash equilibrium the game theory is that everyone is better incentivized to follow the rules and to break apart from it Then you start building a system that you know it relies on
• 1:18:16
  trust as much as it relies on incentives. And I think we haven't done a lot of that, right?
• 1:18:22
  This is kind of when you were asking, how do we mature out of this last era of Bitcoin is
• 1:18:29
  stop being so pedantic about trying to disenfranchise the individual, the institution,
• 1:18:40
  and their ability to kind of have agency in the market by collaborating with others, right?
• 1:18:46
  It is true that, you know, the ultimate form of agency is the sovereignty of the individual.
• 1:18:52
  But the reason why, you know, free market capitalism works is that, you know, listen,
• 1:18:58
  And we operate in a market that we're able to collaborate with one another to achieve the outcomes that we need.
• 1:19:06
  And this needs some sort of rules, right?
• 1:19:09
  This needs like the ultimate law.
• 1:19:12
  The only law in Bitcoin is the laws of consensus, right?
• 1:19:16
  And those are the ones that everyone needs to abide by.
• 1:19:20
  No one can really debate or discuss.
• 1:19:24
  well certainly they can debate and discuss them but in their individual agency there
• 1:19:30
  very little recourse to be able to change them but on top of that there's a there's a room there's
• 1:19:36
  room for systems of rules and this is what protocols are right you know smart contracts
• 1:19:43
  are protocols function nick stable used to refer to them as functionary code right functionary
• 1:19:50
  protocols, which is how do you, these kind of flow charts that you think about, maybe if you've
• 1:19:56
  taken like an accounting class back in college and you think about these flow charts where it's like,
• 1:20:02
  okay, the auditor signs off on this transaction, this debit, and then it goes somewhere and then
• 1:20:08
  the treasurer, whatever, and so on and so forth, right? I mean, there's books and books and books
• 1:20:15
  and books of these processes that are embedded in the fiat monetary system.
• 1:20:23
  And it's part of the reason why it's as big as it is.
• 1:20:27
  So the way that I see it is we have a lot of work to do to kind of pick up the slack
• 1:20:33
  and kind of be able to create these types of frameworks so that regardless of where you come from,
• 1:20:39
  if you're an ardent, sovereign maximalist that's never going to trust anyone with your coin,
• 1:20:45
  or you're a sort of middle-aged person
• 1:20:51
  that prefers dealing with Coinbase or whatever,
• 1:20:55
  there needs to be some sort of ways
• 1:20:56
  by which we can connect all of these people together.
• 1:20:59
  And these are the market interface
• 1:21:01
  that we're hoping to create with Arcade and with our clubs.
• 1:21:06
  Well, I'd love to see it.
• 1:21:08
  I've been telling you, I'm going to see you in person and in DMs.
• 1:21:12
  I think what you guys have been building,
• 1:21:15
  what you've been teasing is very exciting.
• 1:21:18
  And I think you are finding success
• 1:21:23
  in achieving your goal of sort of expanding the conversation
• 1:21:27
  and pushing the boundaries of how we think about
• 1:21:32
  individuals interacting with Bitcoin,
• 1:21:35
  businesses interacting with Bitcoin,
• 1:21:37
  and interacting with each other
• 1:21:39
  as interacting with the protocol at different layers.
• 1:21:42
  And I think Arc Labs Arcade has been a breath of fresh air that's been underappreciated in the market for quite a while now.
• 1:21:54
  But I think per our discussions and DMs, it seems like you guys are really ready to begin leaning in and going after all this stuff that we just discussed.
• 1:22:04
  So I'm very excited to see that.
• 1:22:06
  But before we wrap up, too, I mean, AI has been a sub theme threaded throughout this conversation.
• 1:22:13
  And I think it would be remiss if we didn't sort of tie the knot on that theme and what it means for the state of Bitcoin security today, what it means for the state of Bitcoin security maybe a year or two years from now.
• 1:22:27
  Are we going through hell right now to harden the system, to make it more robust than we could have ever imagined just a year ago, a year from now?
• 1:22:39
  And then what does it look like hardening the system?
• 1:22:42
  I saw that Prem AI announced this morning, the launch of Cyber Scam, which is Prem's proprietary security agent for vulnerability detection built partnership with Arc Labs and Breeze Tech.
• 1:22:54
  obviously we've seen the bitcoin red team um uh supported by individual donators uh donors excuse
• 1:23:03
  me and and open sets and what what this is the security landscape within bitcoin look like
• 1:23:10
  today and a year from now um you know it's interesting i i think i i i
• 1:23:19
  I mean, those are maybe words that I'm going to eat, but I think it's as bad as it's going to get.
• 1:23:32
  Meaning it's going to get better from here on out, I believe.
• 1:23:37
  And that is because as bad as we got hit, and probably in certain ways we're continuing to get hit by the advance in these frontier models.
• 1:23:49
  At the very same time when the attack surface became obvious and just fucking manifested itself by the various attacks we've seen play out during this month.
• 1:24:10
  the AI
• 1:24:14
  again
• 1:24:15
  the scope of AI development
• 1:24:18
  has fundamentally shifted as well
• 1:24:20
  and so now you're seeing the open source
• 1:24:22
  models catching up
• 1:24:24
  the open source models are the entire
• 1:24:26
  story here
• 1:24:27
  they were
• 1:24:28
  the attackers
• 1:24:30
  they were used by the attackers
• 1:24:34
  or most likely we suspect that they were
• 1:24:36
  and they were also
• 1:24:38
  the ones that were being used to fend off the attackers.
• 1:24:43
  And I think that's a massive story, if only for the fact that, you know, I think that's
• 1:24:49
  a big debate within the technical community now as to whether or not, I guess, whether
• 1:24:58
  or not it's possible at all to write secure software in this day and age.
• 1:25:02
  and you know as someone who knows nothing about software or at least hasn't written any software
• 1:25:09
  other than vibe coded something my intuition and always someone that operate and someone that
• 1:25:16
  operates always off of my intuition is that it's actually going to make writing software
• 1:25:26
  and safe and secure software much easier than it's ever been.
• 1:25:31
  And I think you can get a lot of that software
• 1:25:33
  to certain security domains
• 1:25:39
  and sort of standard of deployments
• 1:25:43
  where when you look at the spectrum of security of software,
• 1:25:50
  maybe on the further most secure end of the spectrum
• 1:25:54
  is all of these, what has been referred to as kind of formally verified systems, right?
• 1:26:01
  Where you can kind of mathematically prove the integrity of the system, right?
• 1:26:07
  Whereby literally, you know, unless you get to a point where, you know, it's a different
• 1:26:13
  conversations if AI starts breaking, you know, elliptic curve cryptography, I think we're
• 1:26:20
  not really there yet.
• 1:26:21
  But if we assume that cryptography is sound and is going to remain secure for the foreseeable future, I think we can push the development of Bitcoin software into a direction where you're going to need to harden kind of like the dependencies.
• 1:26:43
  A lot of the security failures, a lot of the time, they come from this seemingly small dependency in your code that calls another service.
• 1:26:58
  And perhaps that service gets compromised and it opens up an avenue for compromise of your own system.
• 1:27:06
  So it's going to be teamwork.
• 1:27:11
  work, right? It's like you can build the most secure system that you want for your own organization.
• 1:27:18
  But ultimately, if everyone again, kind of like doesn't do the same thing, it only takes one.
• 1:27:27
  In a lot of cases, it only takes one domino to make the whole thing fall apart.
• 1:27:32
  You're only as strong as your weakest link.
• 1:27:35
  Yeah, exactly. I mean, yeah, it's very true in some ways. But so I mean, so this is why
• 1:27:41
  Part of that then becomes how do you assess those threats?
• 1:27:47
  And this is why Prem,
• 1:27:49
  Prem being kind of a non-official sister company of Arc Labs,
• 1:27:56
  actually a project that is led by one of the co-founder of Arc Lab,
• 1:28:02
  a completely separate project in its own right.
• 1:28:05
  It's been around for a couple of years already.
• 1:28:08
  And they started off with the thesis of sovereign AI and prem being in part meaning on prem,
• 1:28:19
  so on premise meaning local AI, locally deployed AI.
• 1:28:23
  And they been doubling down on that for again the last couple of years And what we seen over the last couple of months is just such a validation of the direction that they were going into
• 1:28:38
  And so it really has been, you know, as far as they're concerned,
• 1:28:44
  such an opportunity to kind of put themselves on the scene
• 1:28:48
  and demonstrate the expertise that they've acquired.
• 1:28:51
  And so one of the things that we did with them,
• 1:28:55
  the moment that some of those dominoes started falling is get together in the same way that the
• 1:29:01
  red team did with Rob and Callie and all of that. I think it was, again, it's not just a single group
• 1:29:09
  of person or contributors or individuals. It was a collective effort across all different
• 1:29:14
  kinds of teams. And what we did is in a lot of ways, very similar to red team or project loop
• 1:29:24
  from
• 1:29:25
  block, from
• 1:29:28
  spiral, sorry,
• 1:29:30
  which is just create a
• 1:29:31
  secure security harness.
• 1:29:34
  What is our harness? It's just
• 1:29:36
  sort of a
• 1:29:38
  scope AI.
• 1:29:43
  It's a method
• 1:29:44
  by which you're able to run
• 1:29:46
  scans of your
• 1:29:48
  software, of your projects
• 1:29:50
  via an AI model
• 1:29:52
  and instruct the model to work in very specific ways
• 1:29:56
  whereby it's going to look at very specific part of the codes
• 1:29:59
  and follow a certain mechanism to really audit.
• 1:30:05
  Because you can go on KimiK3 and give it the GitHub link of the organization
• 1:30:12
  and it's going to do a security audit.
• 1:30:15
  But we know that these processes of these AI and these LLMs,
• 1:30:20
  it's not deterministic. It's very random. You never really know how it's going to go through.
• 1:30:25
  And in a lot of cases, for whatever reason, it's going to skip one part of the code that might be
• 1:30:31
  more critical than the other. And you never know. It just does not tell you. So you kind of have to
• 1:30:37
  be very stringent in terms of what you want the audit to actually be able to do. So Prem,
• 1:30:47
  the last couple of weeks, working with Breeze, working with us internally, Andrew.
• 1:30:52
  I mean, we didn't talk about Andrew Cooks, but obviously on our end of our clubs,
• 1:30:57
  we were certainly targets of attacks by people poking around.
• 1:31:02
  Fortunately, we dodged the bullets and were able to harden our infrastructure
• 1:31:10
  before sort of any vulnerability was exposed.
• 1:31:14
  but
• 1:31:16
  you know
• 1:31:18
  Cook's
• 1:31:19
  our CTO
• 1:31:19
  Andrew is also
• 1:31:21
  the maintainer
• 1:31:22
  one of the maintainer
• 1:31:22
  of BTC pay server
• 1:31:23
  and unfortunately
• 1:31:25
  they were not
• 1:31:26
  as lucky
• 1:31:28
  as we were
• 1:31:28
  and so Andrew
• 1:31:29
  decided to put
• 1:31:30
  a lot of work
• 1:31:31
  you know
• 1:31:31
  I think just
• 1:31:35
  you know
• 1:31:37
  I think everyone
• 1:31:38
  is kind of
• 1:31:38
  scrapping together
• 1:31:39
  and learning
• 1:31:40
  as we all go
• 1:31:42
  in terms of
• 1:31:43
  okay how do we adapt
• 1:31:44
  to this new threat? What can we put in place? But the pace, man, the pace, again, this is all
• 1:31:50
  comes back to the acceleration is like, dude, I'm seeing what's playing out and the ability now with
• 1:31:55
  pram to do constant monitoring. So, you know, it's like, no, you're not doing like a security audit
• 1:32:01
  any every quarter. You have a machine and an agent that's running a permanent security audit
• 1:32:08
  on your system, if there's a new pull request that comes in, a new commitment to the code base,
• 1:32:16
  that gets audited, that gets cross-checked. And so you have something that's a lot more robust.
• 1:32:23
  And so generally, I'm super positive. I mean, I think it's just like,
• 1:32:30
  I think AIs are just getting so good as well that, you know, they're going to just write secure code.
• 1:32:42
  I think it's possible to do that.
• 1:32:44
  I mean, I've been on the, I don't know if I'm even just a reliable opinion at this point, at least in the last couple of weeks, because man, I'm on this, I'm on the edge of the one-shotted sort of meme.
• 1:32:59
  where it's you know for the last year there's been kind of like these cycles and i remember
• 1:33:07
  very distinctly like two or three occurrence of me chatting with a friend and being like yeah you
• 1:33:16
  know what like the ai models man like they're they kind of all suck now like they're like you
• 1:33:22
  know i would tell them like you know i feel like they've all gone to shit like it's like they're
• 1:33:25
  I would tell them, I feel like they're not getting any better, and perhaps they're getting worse.
• 1:33:31
  And then inevitably, dude, a couple of weeks after, somehow, dude, everything would change.
• 1:33:38
  And somehow I would I feel like I in that position right now where like I was telling you what I developing with the vault And obviously I can audit any of what happening but the quality I mean I building an app which would have taken two years for someone
• 1:34:00
  I built that in four or five days, dude.
• 1:34:04
  Yeah.
• 1:34:05
  It's insane.
• 1:34:05
  And it's not like a, it's not a fucking, you know, it's a very complex code base that interacts with Bitcoin.
• 1:34:15
  And, you know, again, is it full of holes?
• 1:34:17
  Maybe.
• 1:34:18
  But the reason why I'm saying I'm on the edge of being one-shotted is I was actually having a conversation with Andrew just yesterday about this.
• 1:34:25
  And I was like, man, I feel like I'm getting close to a point where I have to ask you guys, like the developers to be like, hey, like, can you honestly look at this?
• 1:34:35
  because it works.
• 1:34:37
  And I wonder if the code is not actually quite good.
• 1:34:42
  I have no ability to judge it, right?
• 1:34:44
  And maybe it's not,
• 1:34:46
  but eventually it's going to be good,
• 1:34:49
  maybe even better than the developers can do it.
• 1:34:52
  And I was telling also Andrew,
• 1:34:55
  how are the developers going to react to that?
• 1:34:58
  Are the developers going to start pushing?
• 1:35:00
  I feel like we're on the verge of a sort of crisis
• 1:35:04
  in the technical community where the developers are going to realize that,
• 1:35:10
  holy shit, like, you know, the guy that's white coded something by himself,
• 1:35:17
  like he did a better job than I could have possibly done.
• 1:35:20
  And there's nothing I can like this.
• 1:35:22
  And like, then how do you know, how do you yourself kind of accept this outcome?
• 1:35:28
  Right.
• 1:35:28
  Obviously you can pick up on the tools and start working on with them yourself.
• 1:35:34
  but um it's going to be uh yeah the acceleration is completely crazy so i i don't even know
• 1:35:41
  what it's going to look like three months from now let alone a year from now uh what it means
• 1:35:47
  for us what we're doing um it has so many implications and i think that's why it's very
• 1:35:55
  hard you know from the perspective of someone that operates in technology at the moment to
• 1:36:00
  make a plan for that make a plan that's beyond three months the next quarter like focus on what
• 1:36:07
  you can do in the next quarter because you don't know what what's next no it's insane and to your
• 1:36:14
  point about using models but like they're getting worse i mean i i agree like you'll use it happened
• 1:36:20
  with 4.6 and 4.8 opus for me um but it's become obvious like yeah they'll get you addicted to the
• 1:36:26
  crack of the frontier model and then they'll get a ton of demand for it and they'll need to train
• 1:36:31
  the next model so you like divert gpus they just like degradate things and the uh you know the lack
• 1:36:37
  of computing power that's out there is like the bottleneck right now but to your point i too have
• 1:36:43
  the intuition that it's going to make it um it's going to make it so you can vibe code very secure
• 1:36:50
  systems um and just when you think about this having competing frontier models able to cross
• 1:36:55
  audit each other um as you're building uh and if the rate of intelligence continues to increase at
• 1:37:03
  the pace that it has i mean just logically you're going to get and if that's the case you know like
• 1:37:07
  i i think you know maybe what if what if we're just a couple of models away from you know something
• 1:37:12
  that can build the like perfect code call it like like a perfectly secure system right so if that's
• 1:37:18
  true, then we're three models.
• 1:37:23
  We're actually one more model away from, you know, open source, because, you know, you've
• 1:37:28
  seen what's happening with like Quan and all of those where you're actually running it
• 1:37:32
  on your laptop, right?
• 1:37:33
  You're running it locally on your laptop.
• 1:37:37
  You don't need a fucking GPU next to you.
• 1:37:40
  And it performs in the same way that, you know, Opus 4.6 did like six months ago.
• 1:37:47
  Right.
• 1:37:48
  So I think there's a certain limit where it's like everyone assumes that,
• 1:37:53
  I think a lot of technologists assume that everyone is always going to demand frontier intelligence.
• 1:37:59
  But it's like, at what point is this true?
• 1:38:02
  Right?
• 1:38:02
  Eventually you're going to...
• 1:38:04
  Do you need a thousand IQ?
• 1:38:05
  Or is 200 good enough?
• 1:38:07
  No, no, exactly.
• 1:38:08
  200 should be able to operate, again, write perfect code.
• 1:38:13
  So by the time you're able to write perfect code, you know, unless, again, it's hard.
• 1:38:19
  It's hard because it's kind of like all just like I look at it from the perspective of like the singularity and entropy and everything, just like the randomness of everything.
• 1:38:26
  It's just like, what is even secure code, you know, like in this chaotic environment?
• 1:38:32
  Like you never know.
• 1:38:34
  Maybe fucking, maybe we're all worried about, I said that earlier, but maybe we're all worried about quantum.
• 1:38:39
  and actually you know ai is just gonna crack uh elliptic curve and then we're just all fucked
• 1:38:45
  anyways well hopefully we not fucked but um yeah yeah it uh i mean we go far down this rabbit hole it never ending but I going to give you a time I very I very positive myself I super I find myself to be very optimistic about all this stuff
• 1:39:08
  I mean, I feel, I feel like really it would Bitcoin with, with open source hardware, with
• 1:39:17
  local models.
• 1:39:20
  It feels to me like we have, like in a certain way, like Bitcoin was, you know, Bitcoin is
• 1:39:26
  what we all believe to be the ultimate kind of bearer of sovereignty.
• 1:39:34
  But in the world that we live in, it was probably not enough.
• 1:39:37
  And I think, you know, local AI models are going to, you know, open source local AI models
• 1:39:44
  are going to be kind of like this companion
• 1:39:47
  that everyone's going to have
• 1:39:48
  to be able to navigate this world.
• 1:39:52
  And so the convergence is impressive.
• 1:39:56
  Yeah, and if you're out there,
• 1:39:58
  you got to touch it.
• 1:40:00
  Can you got to use it?
• 1:40:02
  I had a couple of conversations yesterday.
• 1:40:04
  It was like a ton of people out there like,
• 1:40:05
  yeah, I use AI,
• 1:40:06
  but they're just using ChatGBT and Opus
• 1:40:12
  or Claude, excuse me,
• 1:40:13
  as like another Google search.
• 1:40:15
  It's like, no, you've got to integrate this in the systems
• 1:40:17
  and create agents and see how these things interact with each other.
• 1:40:24
  Once you see it, it's hard to become pessimistic.
• 1:40:28
  I mean, you can have the lingering thought of ASI leading to the paperclip problem
• 1:40:33
  and us bringing Skynet to the world.
• 1:40:36
  But I don't know.
• 1:40:38
  To your point, these things aren't deterministic,
• 1:40:40
  and I don't think they ever will be.
• 1:40:43
  Yeah, no, no, exactly. No, I mean, I'm a huge bear on AGI. I don't think that this class of intelligence, if you can call it that, is ever going to get close to.
• 1:40:57
  You can tell, you know, like it's kind of ridiculous how good they are at doing a million things.
• 1:41:03
  But you can still, you know, you can still tell that they're kind of dumb as rock, right?
• 1:41:09
  It's kind of like, it's just like, as our friend Rob always says, it's just like pulling the slot machine and just getting a random result.
• 1:41:17
  And maybe it's going to be, it's going to be great.
• 1:41:20
  Maybe it's going to be completely retarded.
• 1:41:22
  it um and and so from that perspective there's there's very little in that tells me that but
• 1:41:30
  you know maybe they'll fucking figure you know maybe they'll maybe he'll pull the slut machine
• 1:41:34
  one day and they're gonna be like oh yeah i found a way to get smart you know to actually be smart
• 1:41:39
  seventh across the board every time
• 1:41:41
  yeah well um it's been great catching up dude thank you for uh for coming on i'm pumped to
• 1:41:51
  see what you guys are pushing out at Arcade and Arc Labs and love to see you guys collaborating
• 1:41:57
  with Breeze and Prem to secure these systems. And like you, it's been a, I'm very optimistic
• 1:42:04
  on the future of Bitcoin. I think we've learned a lot of hard lessons in the last month. A lot of
• 1:42:11
  introspection and reflection has certainly been part of my day-to-day process over the last three
• 1:42:19
  weeks and um i i think uh it's hard to say to your point about like this cathartic release that
• 1:42:30
  that the the cold card thing probably represented um it uh it's it's just like i'm not gonna come
• 1:42:40
  out and say it's hard to say but um lessons that for me personally taken very seriously and um
• 1:42:48
  hopefully will lead to better outcomes, better conversations, better risk assessments in the future.
• 1:42:59
  Yeah, and hopefully better or kind of more willingness to contribute also in whatever ways like is available to the people, you know, just circling back to this AI conversation.
• 1:43:17
  I mean, one of the reasons why I am so optimistic and one of the reasons why I feel like at our clubs and with Arcade, we've been able to keep ourselves away from the distraction and kind of survive through this very difficult bear market and whatnot is that we, you know, controlled our narrative.
• 1:43:41
  We shaped our own future.
• 1:43:43
  We were not out there trying to entertain identity politics or whatever that might have been.
• 1:43:51
  Or just like, you know, it's just like, man, there's never been a better time to build in Bitcoin.
• 1:43:57
  Like every, you know, we've literally hit a knock on wood, like peak bottom from here.
• 1:44:06
  And if you think that...
• 1:44:09
  This is the case every single bear market, right?
• 1:44:12
  But through the deepest part of the bear market, we always underestimate how crazy things can get during the next bull phase.
• 1:44:25
  And I think people are in this state right now where people are dejected.
• 1:44:29
  There's a lot of apathy about the future of Bitcoin and are suits going to control everything and whatever that might be and whatnot.
• 1:44:35
  And well, listen, like if you care about it that much, one of the things that you can do is not just, you know, yap about it on Twitter.
• 1:44:43
  But listen, man, just get a couple of agents going and let's just fucking churn out some product ideas and try to contribute in some ways.
• 1:44:53
  There's really no excuse anymore.
• 1:44:55
  And if you have an idea and if you think things should be better, act on it.
• 1:45:02
  Because otherwise, you're not going to have an excuse.
• 1:45:06
  You don't have to build a full-fledged product either.
• 1:45:09
  Like a simple work.
• 1:45:09
  No, exactly.
• 1:45:10
  Exactly.
• 1:45:11
  A developer can look at and be like, oh yeah, that is actually pretty cool.
• 1:45:14
  Here's how you would actually make it viable.
• 1:45:17
  That's the most, I mean, everyone now, again, I am not going to be able to write
• 1:45:23
  a secure code base myself ever,
• 1:45:26
  but I'm at a point now where I can own an entire sort of product
• 1:45:32
  design scope research phase and execute on it in a matter of a week.
• 1:45:38
  And I have a proof of concept and a prototype that actually works on MuniNet And I can go tell Andrew like Hey listen Andrew like I know you been busy kind of you know patching every code base and making sure that everything is secure In the meantime you know here what I because I don you know I not going to touch
• 1:45:56
  that. You don't want me touching the critical parts and fixing everything. But while you're
• 1:46:01
  busy doing this, here's what about, here's the direction, here's a direction that I thought
• 1:46:06
  we might want to take. And, you know, when you get back to things being a little more normal,
• 1:46:12
  Like this gets us ahead of steam in terms of being like, okay, do we discard this idea or do we start iterating on top of it? And, and yeah, there's just like, there's just so, so much opportunity for people to participate. And it's, it's the better remedy to be able to get through.
• 1:46:30
  I can tell you 100% from my perspective that if I hadn't been at ArcLabs and I hadn't had this opportunity to work on this project in the last two years, I probably would have been completely checked out of Bitcoin.
• 1:46:43
  you know i would probably have my own little stash and i'd probably you know be uh shit posting
• 1:46:48
  here and there on twitter but actually i'm afraid that i would have probably fell into one of these
• 1:46:54
  you know just aimless identity sort of like wars because i i'm a sucker for a lot of that shit but
• 1:47:02
  at least i found that you know i have an outlet now um so i would encourage people to just to just
• 1:47:09
  put their energy there.
• 1:47:12
  And with, you know, hopefully the market rallying back, you know, the opportunities will manifest
• 1:47:19
  themselves.
• 1:47:19
  It might not look like there a lot of market there a lot of demand there a lot of attention for some of the ideas that people have at the moment But man this industry is going to be 10
• 1:47:30
  50 times bigger in the span of a couple of years,
• 1:47:34
  the next decade.
• 1:47:35
  If you think that the people
• 1:47:37
  that own the application layer right now
• 1:47:41
  or own the service layer right now
• 1:47:43
  are going to be the same 10 years from now,
• 1:47:46
  I mean, you're kidding yourself, right?
• 1:47:47
  I mean, there's just like,
• 1:47:48
  there's going to be so much market.
• 1:47:50
  There's never going to be one light spark.
• 1:47:52
  And listen, I love the guys at LightSpark.
• 1:47:54
  I've always been very, very explicit
• 1:47:58
  that we don't see them as competitors
• 1:48:00
  with what we're doing at our clubs and Arcane.
• 1:48:02
  And if there's room for collaboration there,
• 1:48:04
  we'd be happy to do so.
• 1:48:06
  But the point is, again,
• 1:48:07
  that there's going to be room for 10 light sparks
• 1:48:10
  in 10 years from now.
• 1:48:13
  So if you think that they have built something good
• 1:48:17
  that works and, you know, look into perhaps using that model and serving eventually a different
• 1:48:24
  part of the market with perhaps that has different demand. But there's really no excuse for anyone
• 1:48:30
  anymore. So yeah, this is what makes me optimistic. And this is why I kind of keep my circle tight
• 1:48:37
  with hopefully people that are of the same idea and that have never kind of relent. I mean, dude,
• 1:48:45
  Yeah, let's just say we've lost a lot of good warriors, if you can call it that, in this bear market, it seems like.
• 1:48:51
  I don't know if we're going to get them back.
• 1:48:54
  Hopefully we do but hopefully it in a context again where they can contribute in healthy ways and not just kind of like in uh the social area of uh x yeah the roi on engaging that is very low if positive at all
• 1:49:16
  yeah but time is a flat circle this happens every bear market um
• 1:49:19
  Um, I appreciate you again, coming on, doing what you do and, uh, we'll do this again at
• 1:49:28
  some point.
• 1:49:29
  Yeah, man.
• 1:49:30
  Thanks for hearing me out.
• 1:49:32
  And, uh, yeah, let's hopefully we bump into each other soon.
• 1:49:35
  Yeah.
• 1:49:36
  All right.
• 1:49:36
  Peace and love freaks.
• 1:49:37
  Peace man.
• 1:49:39
  Thank you for listening to this episode of TFTC.
• 1:49:41
  If you've made it this far, I imagine you got some value out of the episode.
• 1:49:45
  if so please share it far and wide with your friends and family we're looking to get the word
• 1:49:50
  out there also wherever you're listening whether that's youtube apple spotify make sure you like
• 1:49:57
  and subscribe to the show and if you can leave a rating on the podcasting platforms that goes a long
• 1:50:03
  way last but not least if you want to get these episodes a day early and ad free make sure you
• 1:50:09
  download the fountain podcasting app and go to fountain.fm to find that $5 a month, get you every
• 1:50:17
  episode a day early ad free helps the show gives you incredible value. So please consider
• 1:50:24
  subscribing via fountain as well. Thank you for your time. And until next time.
• 1:50:39
  Thank you.