Research
Timechain
Search Donate
Sponsors
Cake Wallet, private Bitcoin and Monero walletSideShift.ai, no-signup crypto exchangeOrangeRock, non-custodial bitcoin spot and perps tradingBraiins Hashpower, buy and sell Bitcoin hashrateLayer Two Labs, drivechains for Bitcoin
Home / Podcast / S16 E1 / Transcript
Transcript · S16 E1 · 1:59:56 · 3 Jan 2025

S16 E1: Mike Belshe on BitGo, Self Custody & The Future of Bitcoin

With Mike Belshe. Hosted by Vlad Costea.

← Episode page Plain text
Machine transcript, corrected by hand. It can still contain mistakes, so the recording is what counts. Tap any timestamp to play from that moment. Use Cite to copy a timestamped link and a short attribution. Dotted words were hard to make out in the audio.

Search this conversation

0:44Vlad CosteaHello there and welcome to the Bitcoin Takeover Podcast. It's a new year, it's Bitcoin's 16th year of existence and we have started season 16 of the show. This is episode one and we are live across Twitter and YouTube and Twitch and probably something else with Mike Belshe, who is the co-founder and CEO of BitGo. BitGo is probably the biggest and most trustworthy custodian of Bitcoin. I mean, they were some of the first people to believe in Multisig as a standard for security. They basically gambled on this because it was not even a standard when they started using it for custody funds, but they reaped the rewards because today they're a super successful company and Mike is still going.

1:30And I did see you, Mike, on the stage in Nashville last year already. And it's interesting because prior to doing this interview, I was doing some research and I was looking at some interviews with you that are more recent. And I read the comments for that YouTube video with you on a panel with Nick from Casa and the guy from Unchained. And someone in the comments said, you know, this Mike Belshe guy, he's new, but he's very interesting. And that's when I realized you don't really do interviews and most people don't even know about what you're doing.

2:05Mike BelsheWell, look, first, thanks for having me, Vlad. Happy New Year. I'm glad to finally get on this podcast. I feel like I do a lot of these things, but for the most part, I guess I'm a little bit more behind the scenes. And I do remember the Nashville thing. I think it's interesting in particular, the topic there, if you recall, you know, centered around kind of self-custody and custody and like, you know, a lot of people have very interesting views, what's right for them, et cetera.

2:34There's no wrong answer, by the way, everything's right. But it was it was an interesting discussion. I remember some were like, no, you should never have never have any institutional custody. And so we discussed that point. On the other hand, you know, BitGo over the years has become more known for the institutional custody side of things. And people actually forget that we pioneered self-custody. So you mentioned multisig. I mean, we started completely as a self-custodial concept. And back in 2013, started pioneering that kind of wrote up this mechanism like, hey, well, what if we use multisig in this two out of three way where it protects against both theft and loss?

3:16People have forgotten why that is differentiated. It's very important. There's a lot of people out there using, you know, two out of two or three out of three multi-party or multisig wallets, and they protect against theft, but not against loss. And, you know, you can forget about these details. But anyway, I believe self-custody is a huge part of today, the future success. Absolutely. We want to have it. BitGo does a huge amount of business. We are the on self-custody. We are the only provider.

3:48I don't think there's anybody else in the world that has a full complement of self-custody services, as well as, you know, qualified regulated custody is where we're most known. We have to spend a lot of time, money with regulators and auditors, et cetera. Fine. But you can actually go between self-custody and full custody wallets seamlessly, all in a single platform. And of course, you've always got the two out of three protections. We started out with multisig, as you mentioned, these days, a lot of chains.

4:17They don't support multisig on chain. So we have to use MPC. Unlike other MPC vendors, we think the two out of three threshold signature is really, really important. So we waited until we could get full two out of three MPC, which is what we have today. We deploy a combination of actually in some chains you can choose. You can choose multisig. You can choose MPC, depending on what you're trying to do. There's advantages and disadvantages to each. Anyway, a little intro there. Good to see you. And thank you for having me here.

4:46Vlad CosteaYeah. So what does the C in MPC stand for? Computation?

4:51Mike BelsheYeah. Multi-party computation. Right. So multisig is a much simpler technology. So it needs to be recognized on chain. In the case of Bitcoin, we've got this thing called P2SH. And underneath that, there's a little script that's native to the chain where you can specify a set of public keys and that you need two out of these set of three public keys in order to enact a transaction. That's not available on Ethereum. Ethereum does not have that down at its core. MPC is a fascinating arena of computer science.

5:27It's been around since the 1980s. Highly computationally expensive, as you're probably aware. It's super easy to mess up. Cryptography is one of the hardest things that people do. And you can have implementers that can make small mistakes that have big impacts. Anyway, with MPC, in the last 10 years, there's been a lot of progress and innovation. By the way, one of the guys, he's not a formal advisor to BitGo, he advises a lot of companies: Dan Boneh. He's a cryptographer at Stanford, which is right around the corner from here.

5:58It's fascinating. He's been a professor over there for, I don't know, 10 or 12 years. Maybe longer, actually. But 12 years ago, he was the least exciting professor at Stanford. And now he's like the most popular professor in the computer science division because he deals with cryptography. Anyway, he's the author, I'm sorry, he's one of the authors of an algorithm called BLS signatures, which is a signature aggregation scheme similar to Schnorr, different. Some blockchains use it. So he's quite prominent and pretty much one of the top guys.

6:33Anyway, yeah, today we support both multisig on some chains where we can. On Ripple, for instance, XRP, I should say. XRP Ledger supports native multisig. Others like Ethereum don't, and so we have to use MPC. MPC challenges, unlike multisig, where I can have a key and you can have a key, I can sign using my kind of standard ECDSA signature software. You can sign using your standard ECDSA software. Very simple, not complicated. When you're doing MPC, actually you and I are going to do math together.

7:12I'm going to start doing a little bit of math. I'm going to send you the output of that. You're going to do a little bit of math. You're going to send it back to me, and it's going to iterate. It has to go back and forth about eight or nine times. So because of that nature of having to do all those iterations, it pretty much has to be done online. You can't do cold storage very easily. Imagine if each time you got the piece of math from the other party, you had to go take it offline, do some math offline, bring it back online, send it back to the other party.

7:40It's almost untenable. There are some folks out there that say that they've got cold storage MPC. Look, it's relatively new, last four years maybe, and they've self-certified it. Typically, the way you make crypto great is you put it up for peer review, and you let it be reviewed for like 10-plus years. And some of those newer cold storage algorithms, they do a bunch of pre-compute mathematics, which, you know, we will find out if hacking that would allow for a single-side hack to lead to something bad. Anyway, BitGo doesn't do any of those shenanigans. And today, we just do a full complement of both multisig and multi-party computation.

8:27Vlad CosteaYou mentioned two things which I think are super interesting for this interview. On one hand, there's Ethereum. And I know that BitGo is responsible for creating the standard to tokenize Bitcoin on Ethereum. And you're responsible for enabling DeFi using BTC with WBTC. And that's sort of a big deal. I think right now, WBTC, which is Wrapped Bitcoin, and you have a white paper about wrapping assets, and that got published like 10 years ago or something.

8:58Mike BelsheI'm not sure it was that long ago, but maybe, yeah.

9:01Vlad CosteaBut it's bigger than Lightning. It's bigger than Liquid. It's bigger than pretty much every other layer two out there for Bitcoin.

9:09Mike BelsheYeah, look, I mean, different people have different views as to what's the most exciting elements of digital assets, crypto. People ask about use cases. There's a few really strong use cases that emerged. So we've got Bitcoin, of course, which is doing the store of value component. There's payments, which I think is still being cracked a little bit for the digital asset side. All those stablecoins are looking to fill that gap pretty quickly. And then the third one is DeFi itself. And especially if you're on the regulated side of things and you understand a little bit about how market structure works and what broker dealers do and how do exchanges differ from that?

9:43How does clearing work? You start to realize, holy cow, DeFi is the thing that can start to completely replace broker dealers altogether in a way that's less risky, more transparent, et cetera. All right. Anyway, obviously, with Bitcoin not having smart contract capabilities, DeFi emerged on Ethereum first. Now there's some other chains that are trying to win there as well. But the DeFi applications, which includes borrowing lending services, trading liquidity services, other types of swap services. I think these have tremendous interest. But back in around 2017, they were having a hard time getting off the ground.

10:24And part of it was the fact that the price of Ethereum was relatively light. How do you use that as collateral to build these services, et cetera? You couldn't. So some of the team here at BitGo, I wish I could take credit, but actually no, it's the great engineering team here, came up with the concept of like, look, let's create basically the world's first digital asset stablecoin. Tether predates, I think, as a stablecoin for dollars, but this was the first kind of Bitcoin stablecoin. And could that be then used in DeFi applications?

10:57So they put that together. We did a prototype version of it that was out for a little while, and then we put out the final version. I think for the first year and a half, it kind of sat in, I don't know, maybe there's a million bucks on it or something like that. And then all of a sudden, DeFi hit. I think it was around 2017. And maybe it was 2018, and then it really took off. So we're proud to have played a small part of that.

11:21I do think it was a critical element in the history of helping DeFi come to life. And at least in my book, I don't know about yours, but I think DeFi is clearly the third strong use case for digital assets, and it's only going to get better from here.

11:37Vlad CosteaIt is interesting. And for perspective, it should be mentioned that you have been involved with internet projects for a long time. You worked for the Netscape browser, you developed some encryption HTTP protocol that was used in the Chrome browser for a while. So it's important for those listening to understand that you come from this Silicon Valley technology background. And the way that you look at these technologies is not dogmatic. It's more of a, you know, technologist perspective that you have.

12:12Mike BelsheWell, I do come from the technology side. I'm a computer science major. I've been doing technology my whole career. I hit browsers and internet a lot. I was at Netscape in the early days. It was the most fun ever. I don't know if you know Brendan Eich. He's the CEO of the Brave browser, which does a little bit of crypto and a lot of privacy. He was the creator of JavaScript over there at Netscape, same time I was there. But then I was lucky enough to get into Google just at the time they were creating the Chrome browser, and, you know, the web protocols had stagnated for 15 years.

12:48We had had HTTP 1.1, which came out in the late 90s, and literally nothing had changed. And so we were trying to make the web faster with Chrome. By the way, Chrome exceeded all expectations for what the development team had for it. We really just wanted to help drive standards a little bit better. The idea that it would become the 90-plus percent market share browser was absolutely never expected. But part of it was, okay, if you really want to pull out all the stops and make web pages load quickly, how do you do that?

13:18And the networking protocols that we had were outdated, both at the transport layer and also at the application layer. So I pioneered an internal protocol. It was originally called FLIP way back when. Then we changed the name to SPDY. Eventually it became HTTP/2. And that didn't finish up until after I had left Google. So by the way, interestingly, I think BitGo, so I was at BitGo at the time that we finally finished it. And I think BitGo is the only crypto company that has its name on an RFC.

13:51So if you're not familiar with the way internet protocols have worked, there's this thing called the IETF, the Internet Engineering Task Force. It's been around since the early 70s. It's responsible for guiding standards from everything from TCP/IP to Ethernet protocols to routing protocols to the web protocols. So anyway, we did standardize HTTP 2.0 through that. I'm the lead author of that. And BitGo was lucky enough that it gets its name on an RFC simply because we finished it when I was here doing BitGo at the time. That was 2013, I guess, maybe 2014.

14:33Vlad CosteaAnd what's really exciting about it is that I think I've heard you say in another interview that you got inspired when you discovered Bitcoin by a quote by Tim Berners-Lee, who is the creator of the World Wide Web. And he mentioned at one point in the 90s that there's going to be a moment in the future of the Internet when computers are going to send money to each other automatically.

14:57Mike BelsheYeah, that's sort of right. It wasn't a quote from him, but actually it's in the protocols themselves. There's this little known error code. I think it's 402, which is payment required. And it's never really been used. And I was in Netscape back in those early days. And at the time, everyone thought the reason we needed secure protocols was for e-commerce. We wanted to be able to transact on the web, which is true. And unfortunately, we didn't have Bitcoin at the time. And it's difficult to put cash in your computer and have it go through a wire and show up at somebody else's computer.

15:35So we instead built it on top of credit cards, which is a whole morass of payment technology that all of us pay for through the nose. So even though the protocol had anticipated an idea that your computer could potentially negotiate for you know, with a newspaper, whatever, you could say, actually, here's a question, Vlad, you ever go to unsubscribe from a service and then you get this wait, but one more thing, we'll cut your price in half, will you stay? And you're like, well, damn it. So I've just been paying you like twice as much as I should for the last whatever.

16:09Imagine your computer, your AI bot being able to just kind of negotiate this for you and you're at the New York Times site and it says you want to read this article and they want 10 cents and your computer's like, nah, I'm going to negotiate five cents for them. And that can happen. So I don't I don't think we're that far away from having it. But yes, the idea that we can do payments directly inside of our browsers is obviously going to happen. Tim Berners-Lee anticipated that, you know, he was the author of the early the early web specs.

16:38He anticipated way in advance. So he deserves way more credit than certainly I do. I just I just kind of read about it. And then actually, also, we hired a guy, Ben Chan here at BitGo. He was here for a long time. He's gone on to do a bunch of other great stuff. Part of his resume when he came in, he was CTO of BitGo for a while here. But part of his resume was he had worked on this project called Zero Click. You know, and, you know, if you look at like web applications, you know, you used to minimize the number of clips, clicks that you had to do in order to do a checkout.

17:11I think Amazon patented, you know, one click checkout. What's better than one click? Zero clicks, right? And so once you actually have something like Bitcoin, like digital money that you can then set up inside of your browser, it can just pay for you automatically under the rules and guidance that you've given it, and zero clicks. Like, why do we have to deal with this stuff anymore?

17:36Vlad CosteaYeah, it's really beautiful in the way that it sounds, but it's a bit ironic also, because nowadays you have Bitcoin that's being transferred across custodians. You have Bitcoin on PayPal, you have Bitcoin on Cash App, and none of these transactions are actual Bitcoin transactions. They're just settlements that are made on some sort of banking layer. And then people, when they hear this, they appeal to a post on BitcoinTalk by Hal Finney. And they say, you know, this was envisioned from the early days that Bitcoin would not scale for this.

18:07But then again, why do we have all of these scaling layers and why do we have blockchains in the first place when you're going to have banks settling these transactions and adding trust and removing basically all the advantages of using the Bitcoin blockchain?

18:25Mike BelsheYeah, look, the big thing which got in the way of anybody trying to do Bitcoin related payments or small payments, and imagine paying people paying for this podcast, you know, by the minute where like they're just getting a little small micropayment happening, it's hard to do. And on the blockchain, on the Bitcoin blockchain, the fees are too high, even though right now they're relatively modest from time to time, Bitcoin blockchain fees spike. We've seen a few of these in history, and it causes massive, massive disruption. So the answer that's been picked by the Bitcoin team today, of course, is lightning.

19:01Lightning has been a long time in development. I think it started around 2017, so we're seven years in on it. The idea that it should take 10 years to build, it's probably not unreasonable, but it certainly isn't here yet. And we've known that it was going to be long. I know some of the early guys said, oh, we'll have lightning within 18 months or something. They were clearly not right about that. But in the meantime, there's a lot of other technologies that have come along which can handle micropayments and small payments at scale.

19:32And so, look, I think Bitcoin's got a bit of a challenge. I would love to see Bitcoin win here. Bitcoin is the early roots of digital money. It should be the winner, but it's got to figure out ways to scale. And it turns out that when it comes to small amounts of money, the need for it being as perfect and uncorruptible as Bitcoin is, becomes less. And you know this already, right? You've got your net worth, whatever it is, and some of that you have invested in investments.

20:03Maybe it's real estate, maybe it's stocks, maybe it's funds, who knows? You've got some amount that's in the bank, which is your larger amounts of money. You certainly don't put it in your wallet, right? And then you've got a small amount of money that's in your wallet. And for whatever's in your wallet, maybe you've got a couple hundred dollars or 200 euros. But if you lose it's not the end of the world. So how you secure it doesn't really matter. Which blockchain you're on probably doesn't really matter, right?

20:31So if other blockchains are able to build better, faster kind of retail small payment stuff than Bitcoin and Lightning, then of course, those are going to win. And you'll still be able to use Bitcoin for your big storage. But it doesn't have to be a single blockchain world. I know this is the Bitcoin Takeover Podcast, so I hope I'm not speaking sacrilege here. I'm just trying to be pragmatic. I mean, everything's got trade-offs. And I do think that the Bitcoin fees issue is significant, not really solved. Lightning is really complicated, hasn't been made easy enough to solve yet. So we'll see what happens. There's going to continue to be payment competition across many, many blockchains.

21:14Vlad CosteaI guess the utopia is that we're going to have everything divided by the 21 million supply, which is unrealistic, because if you want to have that, then you need every transaction in the world to be settled on the Bitcoin network or some layer. And I just don't see it happening. One sponsor of the show is LayerTwo Labs, and they're working on this idea that's called drivechains. They're two-way peg sidechains. I think it's really interesting. I do like what they're doing. I do understand the trade-offs.

21:44But I think for myself, as someone who likes Bitcoin so much, it's preferable for users to actually be on Bitcoin with some trade-offs as opposed to go to Ethereum. But it seems like at the same time, the Bitcoin Core team seems to be fine with Ethereum existing, and they're going to say, yeah, it's a shitcoin, it's going to implode, it's centralized, it's going to get shut down or whatever. But it hasn't, and it's competing with Bitcoin at a level that's actually becoming interesting because the dominance of Bitcoin has lowered since 2017, or the block size wars.

22:23And I don't know where this is heading, but I do wish there was more happening to Bitcoin. And I'm curious what you think about drivechains, by the way.

22:30Mike BelsheSure. Well, if you go back to the scaling discussions that were all the way back, 2013, 2014, none of these conversations are new. These are long-standing conversations. And people are asking, well, how do we manage the safety of Bitcoin? That is, we don't want to change the protocol and screw it up and cause people to not be able to trust it. But how do we also be able to handle these new use cases? And remember, one of the great things about Bitcoin is that people can trust it for a billion dollars today.

23:04So if you're trying to invest a billion dollars into a blockchain, regardless of which one it is, how do you know that you're not going to have that disappear two seconds after you put it in? Only Bitcoin has really achieved the status where it's ready for billion dollar payments. And that's thanks to being pretty static, right? If you've got a protocol that can be easily changed, then of course you can't get that level of trust. So that was necessary. So back 2014, 2015, the discussions about how do we blend the need for innovation and change with the need for being static and immutable?

23:40And the answer that came back initially was, well, let's do sidechains, right? So you remember Blockstream, they created Blockstream Liquid. It's out there still today. And basically, you can peg your Bitcoin into it. And they've got like a quorum of fiduciaries that help you get it back out when you're de-pegging.

23:58Vlad CosteaBut that's not really what they described in the white paper. And that's what the problem is. They gave up on the idea of creating the two way peg. And they just made this federation of exchanges and big Bitcoin businesses. And they decided that it's good enough because they leverage this geopolitical arbitrage, which I think is a way of giving up on the idea of having real sidechains. It's like true communism. No, you know, it's never been tried.

24:29Mike BelsheWell, I think each of these technical solutions has pros and cons. And is there a second evolution that would be even stronger? Sure, I agree with that statement. However, look at the use on, you know, Liquid and sidechains. It's been slow to adopt, takes a long time to get things going. And if people aren't interested in it's hard to get to kind of the next iteration of it. I think where protocols get better is actually with these iterations. So again, it's kind of back to this problem of how do we keep Bitcoin solid, completely stable, immune to arbitrary changes, and yet also accommodate innovation.

25:06So you mentioned drivechains. So, you know, after the sidechains, then along came the drivechain concept. I think the Rootstock guys may have been the first to drive that. And again, it was maybe a little bit ahead of its time, they created a full EVM, you know, that can be merge-mined down to Bitcoin, and had a lot of interesting benefits, but didn't quite take off. So look, I don't know, I think this is the same battle that exists to this day. Although I think it's more time, which is, you know, innovation versus immutability.

25:44So it's the same battle that we have to this day. But I think one thing has changed, which is, I think, proof that you've actually figured out the innovation only comes when you actually have real apps deployed that are doing real work. So, look, I think it's hard for people to deny the success of Aave, it's been great, done very, very well. And then at this point, it's deployed, right, and they could switch layer ones, and they will, they'll, they'll make themselves present on many chains.

26:14But it's unlikely that it's really going to come back to Bitcoin at this point. I think it uses a set of features and capabilities on smart contract chains, like Ethereum, that are difficult to bring back. And it's also unclear, there's a really strong reason to bring it back to Bitcoin at this point.

26:33Vlad CosteaI guess it's the same as Ethereum, which started from this concept of colored coins and counterparty and Omni, which were already deployed on Bitcoin in early 2014. And they took that to the next level.

26:46Mike BelsheThat was terrible. Right. So I, there's so many people, and in fact, you still see it, you know, with some of the, I guess, the ordinals and other, you know, on Bitcoin chain stuff, like, you know, stuffing things into transactions that the miners are not familiar with, is simply not decentralized. And so many people get confused on this. Like, yes, it's true. You can use the on-chain Bitcoin signing mechanism to ensure that like the transaction can only occur if a certain signature happened. But if the miners aren't aware of the state that's being maintained, well, then that state is being maintained in some software at an application level above.

27:34So Omni, previously known as Mastercoin, clearly had this, colored coins, you know, like the state is maintained in some other software. And so all you've done is you've kicked the dispute out of the Bitcoin consensus layer and into this new layer. Okay, well, how did you make that decentralized? And, you know, people like to hand wave over that and say, well, I get my decentralization from Bitcoin. No, if the miners are not aware of whatever state it is you're trying to maintain, it does not leverage Bitcoin's decentralization.

28:03This lesson keeps coming up. I think every five to 10 years, there's another initiative like, oh, we'll just stick it into the back of a transaction or we'll put it, we'll hide it in the signature field inside of the Bitcoin network and it'll be great. All right, whatever. No, those never work, will never work. And people that understand the protocols are aware of this.

28:25Vlad CosteaI guess a better example is Zerocoin, which was presented in 2013. It was and still is pretty exciting. At the time, the conversation was that they can activate this on Bitcoin with a soft fork and have these zero knowledge proof transactions on Bitcoin. And then there was some sort of disagreement on how to do it and whether or not it's too early or premature for this technology to exist on Bitcoin. So they made Zcash and also Zcoin, which I think right now is called Firo. And it's been a decade, I think almost since that one got launched.

29:04It's pretty exciting. I think what Zcash is doing, they pushed everything to the point where they can have a proper mobile wallet that is trust minimized. They have smaller signatures. They have these mixing pools which have the same transaction size as a normal SegWit one-input, two-output transaction. It's exciting. But at the same time, it seems like Bitcoin is moving philosophically away from these ideas that we need privacy and fungibility. But I do have some hope with these rollups that ironically are coming from Ethereum, which collaborated with the Zcash guys, and they came up with these ZK rollups. And I think that's the most exciting part about Bitcoin development right now.

29:48Mike BelsheDo you know who Vint Cerf is?

29:50Vlad CosteaCan you say the name again?

29:52Mike BelsheVint Cerf, last name C-E-R-F, Vint Cerf.

29:55Vlad CosteaNo.

29:56Mike BelsheYeah, most people don't. He's the creator of TCP/IP. And you know, obviously he started his work way back, probably goes all the way to the 60s. In the early 70s, we had the first transport layers built. By the way, at the time TCP/IP wasn't even split into two layers. That didn't happen until a few years later. All right. Anyway, Vint was the chief protocol officer. He had some sort of name at Google. He's a fantastic guy.

30:26He's taken on a number of leadership roles. I don't know if he's retired at this point, but when we were doing SPDY, what became HTTP 2.0, you know, I had the pleasure of meeting with him and he had some interesting stories. He said like getting TCP/IP off the ground, if they had put security into it, would have been basically a non-starter. Now remember, of course, they dealt with, you know, much more limited computing power at the time of course, but they were aware that TCP/IP doesn't have any security built in.

30:55I think there's a great analogy here for privacy and blockchains. And so you mentioned, you know, Zcash. I agree with you. I love Zcash. I think it's great. It uses the structure of Bitcoin, but it adds the zero-knowledge proof on top of it so that you get some privacy that Bitcoin simply can't offer. And I think we've kind of been in the same problem, the same dilemma space, that Vint Cerf had trying to get TCP/IP off the ground. And if he had had to go and get, you know, certificates and crypto and, you know, encryption all sorted out into it would have made the job much harder.

31:31I fear we're in a difficult spot in blockchains right now, which is making the jump from, you know, completely non-private networks. And it's ironic that, you know, Bitcoin used to be considered this pseudonymous thing. Like, come on, it's completely non-private, and people should have no debate about that. It could be made so much better if we did have privacy shielded transactions. And Zcash, I think, is interesting. Now, a couple things have gone against it. Number one, people, human nature, we're not very good at security.

32:07We're not very good about thinking about privacy until it's too late. A great example there is Facebook, right? We all watched Facebook for 15 years, you know, take all this information about us, and then somehow we got upset about it in 2012, when we realized, oh my god, they're using the information we've been joking about giving them for so long. Like, again, we're just not very good at privacy. Crypto protocols will have a similar moment at some point. We've all been aware that there is no privacy in Bitcoin, and someday there's gonna be like this massive, like, backlash, like, oh my god, I didn't know my transactions were not private.

32:44So that's one reason. The second thing, though, is, you know, governments and regulators have started to get concerned about, like, wait a minute, if we can't watch these transactions, what if bad guys are out there using these things? Look, I'm sympathetic to them not understanding the technology, that there's a learning curve, etc., but they have actively made it difficult for exchanges, custodians, and, you know, regulated entities to participate with privacy shielded transactions. And this, of course, has nothing to do with our obligations as a regulated entity to make sure that we have KYC and anti-money laundering protections and transaction monitoring.

33:23We can do all of that, and yet we can still have on-chain private transactions. So regulators are confused about this point. Those two things are not mutually exclusive. We can solve both problems. But nonetheless, you know, if you, as a regulated exchange, go to any regulator, and pretty much it's any regulator, and you say, hey, I want to support Zcash in the shielded mode, they're going to look at you like you're trying to do something bad, which, of course, is not the case. So this is, I think, it's an open problem for Bitcoin, for digital assets, for privacy, for consumers, that I hope we get on it faster than not, but it's difficult.

34:02There's a lot of static towards the leave it open and unsecured, and we will have the same problem that poor Vint Cerf had way back in 1975.

34:12Vlad CosteaThat's a very good example. I was not aware of Mr. Cerf, but now I'm gonna have to read his Wikipedia page at least. I have it bookmarked right now. I want to ask you more about your take on the block size wars and what happened with the New York Agreement, but before that, let me read an ad. So, SideShift is this company that, due to regulations, is not available in the U.S.

34:36But if you're not from the U.S., or you know how to use a VPN, but I'm not gonna tell you that you should do that, it's your choice, you can basically swap between all sorts of assets, and if, for example, you're getting paid in stablecoins, and you want to get free market money that will most likely go up over the next year, not financial advice, but it's possible, then you can use SideShift, and they have very great liquidity, and they have low fees, and a lot of trading pairs that you can basically choose from.

35:08So, let's say that someone sends you USDT, and you want to convert that into Bitcoin, you can do that on SideShift.ai. Check them out, and thank you for sponsoring the show. And now, let's talk about this interesting moment in Bitcoin's history, because it was like a civil war. I don't like the idea, or the term block size war, because it was not too much about the block size. It was more about which team of developers is going to be in charge, which companies have political influence over the development of the protocol, and possibly it was also about removing Gavin Andresen, and Mike Hearn, and Jeff Garzik, and the old guard of developers.

35:53Mike BelsheWell, let's see. A bunch of emotions get caught up in this. I think everybody that was out advocating for different ways to handle the block size was a Bitcoin fan, frankly. Nobody was looking to make changes to the supply, or other things, but somehow it turned into a concern about that, which I do appreciate. And so, look, at the end of the day, we humans get to decide which chain we follow, under which set of rules. And it is an ultimate human backstop that we have behind Bitcoin.

36:29And the humans voted, and the humans spoke. And so, those of us that had different proposals, I don't feel any animosity or negativity toward it. I don't think necessarily the best solution was picked, but what's most important is that we have Bitcoin, and we have everybody as a community on it. So, I was never very excited about the forks or the alternative versions that tried to compete. Anyway, yeah, if you go back, Adam Back had a zillion proposals. Gavin Andresen had a zillion proposals. Heck, even Satoshi, you can go and find his comments about, and people trying to interpret after the fact, what was his true vision for block size.

37:10What we had was we had a couple of spikes where we were clearly out of block space, and it caused the fees to go through the roof, and it made the chain very difficult to use, and it made it very difficult for people to rely on it for business or for their own personal use. So, as we saw this, we anticipated it getting worse, and we needed something to offset that. There were several attempts to kind of get agreements. So, the thought was like, okay, there's a lot of different views.

37:38How do we get agreement? There was like a bunch of people, I wasn't one of them, went off to China, talked to a bunch of miners. I think Blockstream and others were in the middle of that. There was the New York Agreement. By the way, BitGo was sort of a participant. We were at some meetings that ultimately led to this New York Agreement, but we didn't sign it. One of the things I didn't like was, and by the way, it's because of my roots from IETF and protocol standardization.

38:05When you're building protocols and you got a bunch of developers, they've all got a bunch of great ideas, and yet somehow you're going to have to harness those great ideas and yet also whittle it down to the final incantation of how you're going to use those great ideas. People get incredibly vitriolic and passionate, and they fight. This happens at the IETF. You go to these meetings, and sometimes it'll be down at the low level. People will be talking about this but what byte order do you want to use and which size is best?

38:40I wasn't surprised at all that we had any of that with Bitcoin. Of course, we're going to have it. People are very passionate about Bitcoin. We love this thing. We all want to make it better. After a number of failed attempts to get agreements, different factions went in different groups. We never signed anything from the BitGo side because we didn't think that was the right way to do it. We thought the right way to do it was to have lead by developers and code and then get it out there.

39:04In the end, the decision was made to use SegWit. There was a SegWit2x thing, which I thought would have been a little bit better. I think if we had had more block size, I think we might have seen things play out a little bit differently for Bitcoin over the last few years, but it doesn't matter. There's always a could have, would have, should have. I'm not unhappy with the result that we had. Anyway, it's interesting that we got into such a polarized debate, like you said, of which team are you on as opposed to what's the technology merit of what we're trying to do.

39:37It is true, by the way, anytime you're changing the Bitcoin protocol, you have to be very careful. If you're able to insert this type of change, would it mean that in the future, a change that you may not like could that also get inserted? To some degree, the need to ossify the protocol, there's some reality to it. We cannot change things like the supply. I think that's the thing that everyone agrees makes the protocol as good as it is. I would go to the mat fighting for not changing the supply ever, of course, as would many.

40:12But how do we also accommodate some things which are more practical, small improvements and bring a community along without raising fear that it could change the fundamental properties of the chain?

40:26Vlad CosteaYeah, I think the way in which you presented it is very diplomatic. And I do like the way in which you described it. I was so confused back when it was happening, as in, what's the SegWit2x thing? Why should I listen to this? I think BitPay was a big supporter of that one. And also Bitcoin.com, they were really pushing it. And I was just starting and I wasn't, I didn't have any qualified opinions. I was not in a position to actually express my opinion. I'm not sure if the outcome is the best one we could have had.

41:05I wish the community was not split in two camps. But when you think about it, and I also said this to Amir, I think Gavin Andresen has played such a major role in the history of Bitcoin that he was the shepherd of the project. He was like the main advocate for a while, going on every podcast, talking to libertarian groups, answering to messages on the BitcoinTalk forum, and trying to coordinate new developers like BlueMatt and a bunch of others who came along. But at the same time, he's also the one who made Craig Wright a legit contender for being Satoshi.

41:44And Craig Wright has caused a lot of destruction. He destroyed any opportunity for big block Bitcoin to stand a chance to get adopted because his presence made everything less legitimate or more questionable. And he has been so destructive.

42:03Mike BelsheLook, the guy's a liar and a fraud, right?

42:07Vlad CosteaAnd we finally have a lot of liars and frauds in this industry, but none of them caused this much damage.

42:13Mike BelsheBut Bitcoin stands. I mean, I think we should actually be proud that Bitcoin is designed to work even in light of the worst adversaries trying to steal it, right? And I think we should be proud that actually it withstood it. It's unfortunate that he got so much leverage in the legal system for so long. Gavin was an engineer, right? And I don't want to speak for him on this topic, but I don't hold him liable or accountable for it. He got tricked. And I'm sure that he's kicking himself more than anybody else about that.

42:52It is what it is. Now, by the way, on the flip side of this coin, for those that are new to Bitcoin, although I said we should be proud that it stood up against the adversary of Craig Wright, we still have more work to do. And we still need to keep working on making it more decentralized and more resistant to bigger and bigger attacks. As the value of something goes up, the amount of effort with which you spend to secure it needs to also go up, right?

43:23And so if Bitcoin still had a $100 million market cap, obviously the level of security that would be needed from a chain level, from a wallet level, et cetera, would be much smaller. But we're now at the trillion level, the trillion dollar level, and it's growing fast. And we're going to have entities, it could be exchanges, it could be custodians, governments, people that have trillion dollar wallets. And it may not be a single wallet, but you know what I mean, a virtual wallet. And we need to be securing the whole thing to much greater adversaries.

43:58And when the nation states start deciding that they want to attack Bitcoin, which could happen, are we ready? So if there's one place where we need to fortify Bitcoin, I think we should really be thinking about the security elements of it. If DeFi and payments and other things ended up being on different blockchains on a different time scale in terms of their security, I think that would be fine. But if Bitcoin were ever compromiseable at a nation state level, that's what we should really be worried about.

44:30Vlad CosteaYeah. Speaking of which, what's your take on the strategic Bitcoin reserve and this whole Trump situation? Would BitGo participate and help the US government custody the funds?

44:43Mike BelsheTo be honest, I don't think it really makes a lot of sense. I think Nick Carter highlighted it pretty well, which is like the US has reserve currency status around the world, which obviously is trying to preserve. Obviously anybody that's in the Bitcoin space has some concern that it won't be able to do that for very long, and that may be true. But when you have the ability to print US dollars, you're the one guy that doesn't need to have Bitcoin in the reserves. I don't even know what sense it makes.

45:14So I can understand why El Salvador does. I can understand why Panama does. Most people kind of looking at money probably don't know quite the history here, but El Salvador does not produce its own currency. Panama does not produce its own currency. There's basically a number of countries where they went through the cycle of have their own currency, bad dictator comes in play, prints too much money, hyperinflation, and then reset the currency and do it all over again. And it creates a lot of turmoil. It really ruins the lives of those people.

45:48I think it was around, my history is failing me here. I think it's around 2000 that El Salvador moved off of their own currency and said, you know what, we want to make this future proof against future dictators in El Salvador. We're just going to peg our currency to the dollar. And since the dollar was reasonably well maintained and much more stable relative to their own currency, this worked very well for a while for them. They are now in a boat, which is a little bit different.

46:17So when the US government now starts printing at accelerated levels, it's the foreign holders that get hit the worst by inflation. And you're familiar with the Cantillon effect, right? Right. So like, you know, when we print the money, we print it and we distribute it here. So you know, look, if the government were to print a billion dollars and give it to you personally, Vlad, you would be wealthier. Yes, the rest of us would be diluted, but you would be much wealthier. And that's effectively what happens to poor El Salvador.

46:47They tied their currency to dollars, but they are not the immediate recipients of that. So their effects from inflation is much greater than Americans' effects from inflation. And so now they're kind of in a troubled spot. Anyway, how did we get on this topic? I forgot. I drifted. I weaved.

47:06Vlad CosteaIt was about the strategic reserve and you.

47:08Mike BelsheOh, the strategic reserve. Yeah, yeah. So look, for El Salvador to have a bet other than a US dollar, because they don't have the ability to print dollars, that makes sense. But for the US to do it, I think it's a little bit silly. The other thing which I think politicians continually forget is like the money that the US holds is largely confiscated coins that technically belongs to other people. So I'm not sure how you incorporate that in your reserve. I mean, just as an example, you know, the Bitfinex hack, we've still got a bunch of recovered coins.

47:41Doesn't that ultimately go back to Bitfinex? What's the plan here, guys? You can't just confiscate it and stick it in your reserves.

47:51Vlad CosteaAnd they can always do another 6102 order for Bitcoin holders. And if people are holding their coins on exchanges, they can seize them. So that can get added to reserves if that's what the desired result is.

48:09Mike BelsheYeah, look, now you're getting into the straight self-custody versus custody debate. And this is a tough one. So you know, there's pros and cons to each. So the pro of having self-custody is, I mean, you just hit it, right? Like nobody can take your money, not even the government, which is great. And then the con is that, well, now you're subject to the $5 wrench attack, right? And so how do you prevent that? And stepping away from what it should be for a second, let's talk about just humans.

48:39I've seen this over the last 10 years. Maybe BitGo's been in a better vantage point to see this, which is how much risk are people willing to take kind of in their own custody? And I have these conversations literally every month. I had one just within the last month. The guy who's had Bitcoin for a long time, it's now worth quite a bit. He did self-custody and his wife is now like, well, look, if you pass on, I'm not sure I'm going to know how to get this and recover this.

49:04And so like, you know, how do we deal with that? And should we be considering a combination of custodial services in addition to non-custody? But all the way back in 2014 was the first time I saw it. The guy had self-custody at BitGo and, you know, it grew to be a few hundred thousand. It wasn't that much. But for him, he was scared to death that he was responsible for a few hundred thousand dollars of his family's net worth. And if anything happened to him, you know, he would lose it or he could get robbed, etc.

49:35All right. So everybody's got a different comfort level. And the thing you should ask yourself as a Bitcoin holder, you know, what percentage of your net worth are you comfortable protecting by yourself? So is it 90% of your net worth that you would be comfortable with? Is it 10%? And everybody's got a different answer. There's no wrong answer here. But whatever your comfort level is, you then have to figure out how to build to that. And then depending on the size of it, you need to think about that.

50:02So if you do decide to secure 90% of your net worth or more on your own, what level of depth are you willing to go to? Are you going to put bars on your windows? Are you going to hire a security guard? How are you going to protect against fire? How are you going to protect against all the different elements? And if you screw up anything, you're worth nothing. And your family's got nothing. And now, you know, you may have a hard time putting your kids in college.

50:27I shouldn't laugh. This is serious stuff. Anyway, you end up with this debate of self-custody versus custody. And at least right now, I think everybody's got to answer that on their own. But yeah, once you get into custody, how do you protect yourself against governments?

50:43Vlad CosteaIt's a good question. I guess privacy is part of the solution because you don't have to protect something that someone else does not know it exists. So not talking about how much Bitcoin you own is a good start. Not sharing the amounts of Bitcoin you own with a third party is also a good start. You know, running your own node and stuff like this, just to make sure that you have much better privacy.

51:14Mike BelsheWell, there's some self-custody options that you can do, which do help. But it gets pretty technical. I think it's not ready for mainstream yet. But some of the easy ones are making sure you've got time locks on your large deposits of Bitcoin so that even if somebody does deploy the $5 wrench attack, they can't get anything. So let me just take this opportunity. I've said this on podcasts for my own safety and my family's safety. You can kidnap me, you can, you know, hold the wrench to me, you can rip my fingernails out, you're gonna get nothing.

51:45And the reason is because I do not have any of it on my own. And that's important. Not only for not being able to get mine, but also for not even be able to get my clients. But yeah, privacy helps a little bit. I think it doesn't really solve the human nature problem, which is, how am I going to pass this on to my family? Should something unexpected happened to me? And right now, we as an industry, BitGo included, to some degree, we failed, right? Like, we have not made this easy enough to use, that you can share it with your spouse, in all cases, that you can share it with your grandparents or your kids, and have it, you know, seamlessly transfer.

52:24So I don't know, it's an area that we need to continue to work on.

52:28Vlad CosteaAnd once again, the innovation that came with Bitcoin, and I guess the Austrian economists are going to disagree with me, they say that the innovation is the 21 million supply cap. I believe that the innovation is just this public key cryptography that enables you to custody your own coins by just holding maybe a file on a flash drive or 12 words that you can write on a piece of paper. You can choose your own standard, but basically you can self-custody, you can own your own funds, and you can set up your own whatever.

53:05You can have something very complex, you can have a multisig, you can have a treasure map, you can have everything in one place, you can have in your lawyer's office, whatever. Coincidentally, one of the show's sponsors is Hodling.ch, which is a company from Switzerland, and they do exactly this. They advise you what you should do with your stash, depending on your threat model, and depending on where you live, and you benefit from the privacy of a Swiss company. So you go, you talk with these guys, and they're going to tell you if you need a hardware wallet, if you need a multisig, if, for example, you're part of a company that's buying Bitcoin, it's a good idea to talk with them, and they're going to teach you how to set up a quorum and stuff like this.

53:47I'm not sure if they compete directly with BitGo, but they're sponsoring this show, so I've got to give them a shout out. So check out Hodling.ch.

53:56Mike BelsheHey, most of our clients two years from now don't even touch Bitcoin yet or digital assets, so it's a growing space, which is why it's so exciting to be here. Well, actually, let's talk about how do you protect against governments a little bit, because you mentioned Wrapped Bitcoin. Oh, you need something?

54:15Vlad CosteaNo, I have a Trezor that I'm holding.

54:19Mike BelsheOkay. So with Wrapped Bitcoin, we just moved to a multi-custodial model where the other holders of keys are actually in different nations, and this is something that we can do for protecting everyone's coin. So BitGo thinks about this quite a lot. How do we take your wallet and protect it by way of having keys in multiple countries around the world so that any one country going bananas and trying to issue a 6102 order to collect all the Bitcoin would not be able to do so? They wouldn't have jurisdictional reach.

54:57It's doable. It's part of what we explicitly wanted to do with Wrapped Bitcoin is to make it protected around the globe. I think we can do even more. So BitGo holds seven. We have seven custodians that are operating. A couple of them are still in the final phases of regulatory approval. But imagine having a key in each of seven different countries, and each of those seven countries is a regulated entity, and the regulators in those countries care very much about having oversight and control over what those keys are used for, which clients are used for, etc.

55:34So any one country coming and saying you have to use your key to sign a transaction to confiscate this money on behalf of the users of that land, that regulator is going to say no. So anyway, this is doable. It's not perfect yet, but we are moving in that direction. And I think it's actually really important to think about so that you can have both custodial and self-custodial options that protect against the 6102 type of attack.

56:02Vlad CosteaYou mentioned locking your coins, which is something that you can do in Bitcoin. But what do you think about hardware wallets and all of the I think there's an entire industry of devices that are catered towards helping you secure your coins. And is there anything that you like or that you dislike about these products? I showed you a Trezor. The way that I use it is basically it never has anything in the flash memory. So you can take it from me. You're just going to see the setup screen when you plug this because you need to recover from seed.

56:37And where's the seed? That's another question which I'm not going to answer. But basically, I don't leave anything on the device. I just use it for signing transactions and then I erase it. I also know of devices that promise to have a permanent memory that you cannot erase. And to me, that's confusing. Why would you want to have that? But they market towards newbies who maybe forget how to set up stuff or maybe that they're trying to sell multiple devices to them. I don't know. There's an entire market out there of devices.

57:14Some of them are open source. Others are not open source or pretend to be open source, but they're not. Do you have any experience with this?

57:22Mike BelsheI mean, of course. I've used them myself. I used a dozen of them. Look, I love hardware wallets. If you are taking your own custody, you should definitely be using a hardware wallet. Please don't use browser wallets, etc. But everything has trade-offs. Oftentimes, those are single-signature devices, which means you've got a single seed phrase, etc. And now you run into the whole problem of do humans know how to secure their own stuff if they need to? On this podcast, I'm sure there's a zillion people that have put their seed phrase into a safety deposit box.

58:03Pretty common advice is to do so. Good news, if you do put it into a safety deposit box, you're protected against fire at your home. So that's good news. You've got your device at home. You've got the seed phrase in the bank. On the other hand, we just talked about 6102 confiscations. Do you think that's going to protect you against a 6102? I don't think so. They're going to go to the bank. They're going to freeze all of the safety deposit boxes for people that they know have Bitcoin, and they're going to go take those seed phrases.

58:36The challenge there is you just kind of keep running into these new different ways that you'll run into problems. At the end of the day, it's trade-offs. Individuals need to think about which trade-offs are the highest priority to them and which ones they want to solve first. And whichever route they pick, they're going to then have some new edge cases that they're going to have to deal with. And then how do they want to protect against that? So anyway, yes, I love hardware devices. You should be using them.

59:03I think most people probably want a combination of some self-custody and some custody. And then as we build custody and we build regulated custody, it'd be great if regulators actually got on board with this idea of how do you hold keys, not just in a single jurisdiction, but in all jurisdictions so that we can be safe against both physical threats as well as government threats.

59:27Vlad CosteaIt's a complicated idea, mostly because there is no single answer. There's no glove type of solution, like one size fits all.

59:37Mike BelsheOh, that's right. Look, I mean, when you get into the depths of security and you want to deal with all the physical attacks and virtual attacks, you know, banking and regulators, OCC, others, they look and scrutinize heavily. What are your virtual and physical security protections? Doesn't matter whether you're talking about digital assets or banks. You know, these things exist. And when you're talking about billions and billions of dollars, of course, you have to go to the nth degree. So now how do you translate that all down to individual people? It gets hard.

1:00:09Vlad CosteaYeah. And honestly, it's so complicated. And I think I agree with you, because generally in the hardcore Bitcoin community, people say, oh, it's not so hard to back up your wallet file on a bunch of flash drives or to use this BIP39 seed phrase system and store 12 words or 24 somewhere. But it is complicated. And I do see people when they ask me, how should I get set up? I want to buy some Bitcoin. And I tell them, okay, set up a wallet on your phone.

1:00:42And then they have to write the 12 words. Their first instinct is to take a screenshot of it. And I tell them, you know, are you aware of how many applications in your phone have access to your pictures and that they can basically access your funds just because they have this? And then they take a napkin or something and they write down the words and they're not happy about it. They feel like it's such a backward solution. Like why would you want for the money of the future to get pen and paper and write something and then have to store that and make sure that you don't lose it?

1:01:12Because people are used to losing passwords, and they expect to have some sort of recovery system where they just click forgot password, they type in their email address, and they get some sort of link on their email to reset. And it's difficult. And I do like I think Edge Wallet has the simplest way to onboard right now. You just set up a password or you can use face ID. And for recovery, you have this master code. It's like a private key that you write down. And it's associated with that specific wallet.

1:01:48The problem is that I'm not aware of any other mobile wallet or desktop wallet using the same standard. So it's security by obscurity. And if that one stops being maintained, and you don't find any new versions of that or whatever, you might get screwed. Usually the most used standard is the most secure. But I do like that one for being simple. I'm not aware of any other schemes. I know that there's WalletConnect in Ethereum. I know that there's MetaMask, and that one is convenient. But I would never use that sort of solution, at least not without a hardware wallet to act as the signing device.

1:02:27Mike BelsheWell, you can use it for small amounts of money or experiments. I mean, that can be convenient and good. Just the challenge is that sometimes people start out with a small experiment and then it grows into a big one. And it's hard to retrofit your security later. But actually, by the way, this is a place you mentioned a little bit of like Edge Wallet, and if they were to disappear, and I don't know it, I assume it's a great product. But one of the things that BitGo, you know, we were so early, we're hitting some of the early paranoias.

1:02:55And I wanted to make sure that people knew that even if BitGo disappeared off the face of the earth, you still had access to your money. So from the very beginning days, we created, it's still there, it's called the Wallet Recovery Wizard. It's been enhanced and expanded and enhanced and expanded, where you can get a GitHub set of software, and you can recover your wallet at BitGo using zero services from BitGo. And this turns out, it sounds like it should be easy. It turns out to be a really hard problem.

1:03:24Because remember, in addition to the private key management, the other thing you have to do is you have to go find your transactions on the blockchain. And oftentimes, we're doing key derivation to create multiple addresses, etc. How do you find all of that stuff? Well, if you're not going to use the BitGo services, you got to be using somebody else's services, and those things come and go. So we've used, you know, various open ledger APIs to be able to scan the blockchain. Anyway, very few people do this.

1:03:52I don't know if Edge Wallet has it. We spend a lot of time and money just investing in this thing, which is a tool that you can go and recover your funds without touching a single BitGo server. And it's a tough problem. But if you are going to go self-custody, and you're worried about that type of thing, you do have to worry about it. The second thing you have to worry about, you know, we talked about multisig versus MPC. A nice thing about multisig, or single sig even, super straightforward technology, we got ECDSA signatures, we have, you know, just a few mechanisms, not much you have to deal with.

1:04:26And you can find lots of libraries that support it. When you get into MPC systems, everybody, everybody's protocol is proprietary, every single one. So you know, we talked about how I do a little math, I send it to you, and you do, what's the protocol for how that works? Right? There's no HTTP for the standard of MPC protocols. So as such, no two MPC implementations are compatible, you always have to use one guy. And then what's the backup look like? So anyway, this is a very real issue when you use MPC type wallets, which is how are you going to make sure that you can actually recover those funds in light of the vendor, BitGo or otherwise being gone.

1:05:07Vlad CosteaWhen we started the interview, I mentioned that I watched your panel from Nashville, which was with some other guys from Casa and from Unchained. And I know that BitGo is providing a very similar service with Casa. And Jameson used to work for BitGo. He was the CTO for a while, and then he went and started Casa with a bunch of others. And I wanted to ask you, what's the difference between BitGo and Casa? Is it just the type of customers which you're trying to provide the service because Casa is for individuals and BitGo is for institutions? Or is there something more to it?

1:05:45Mike BelsheBy the way, Jameson and I are still friends. I'm super impressed with everything he's always done, even when he was at BitGo and then beyond. One of the most gratifying things about having a company that's run for 10 years is getting to see where people go and the things they do. One thing we've wrestled with is, people come to BitGo for a lot of reasons. Usually, they're passionate about what you can do with digital assets, what kinds of things are in our future for how we deal with money, etc.

1:06:12And they like what we're working on. Some people come in because they like Bitcoin. Some people come in because they like Ethereum. Some people come in because they're interested in new innovations on key management. So we've got all kinds here. But every single employee at BitGo is excited about crypto in some way or another. Jameson in particular, he's Bitcoin only. That's the thing he likes the most. I'm not sure if he would quite say it that way, but he definitely wanted to be working on Bitcoin stuff and not others.

1:06:38We had a number of people over the years that also came in, joined the BitGo team, helped scale that wallet. We were the first to do all kinds of interesting features. We were the first to write Taproot transactions onto the blockchain, etc. We've always tried to keep up with the cutting edge of the latest, which is available on Bitcoin. But a lot of times, those guys end up wanting to go do their own thing. So initially, Casa was Bitcoin only. I think a couple of years ago, they might have started to support Ethereum.

1:07:05I think they took some flak for that. Because some of their clients are Bitcoin only. But the thing that he wanted to solve was they did a three out of five multisig, whereas BitGo does a two out of three multisig. And then you can ask yourself, well, which one's better? Look, from a straight up security perspective, I'm not going to say three out of five isn't more secure. It is when all you say is, is it two out of three or three out of five? But now you run into a different thing, which is complexity.

1:07:36And remember, complexity also impacts the security of what you do. We like to think that this is all cryptography. But when you're talking about securing digital assets, your operations matter absolutely just as much as the technology you're using. Example is very simple. You know, the analogy that somebody has the greatest encryption in the world, but they write their password and they stick it underneath their keyboard, right? So the encryption was fine. It was the OPSEC that was bad. And when you look across all of the exchange hacks, almost every single one of them, if not every single one of them, is OPSEC, not crypto, being hacked, right?

1:08:18So anyway, it's very important. So when you go from a two out of three system where you've got to secure three keys to a three out of five, now you're going to secure five keys. And now you have to think about the OPSEC for every single one of those five keys. So look, I like the Casa guys. I'm definitely not trying to criticize them at all. It's a different approach. Everything's got trade offs. I think if I recall correctly, they have a mechanism where they say, okay, you're going to have three.

1:08:40So if you want to be self-custody and three out of five, that means the end user has responsibility for three keys, right? So they say, hey, store one on your laptop or phone, store one at your work, and then store one offline, probably print it. I think those are the three keys that the user has. And then Casa is able to take two keys. But yeah, anyway, the difference between a Casa and a BitGo style multisig is the OPSEC that's going to be required for those two things.

1:09:09Vlad CosteaAt the same time, I do have to praise the fact that you believed in multisig so early on, when it was not even standardized. And I think nowadays, it's more usable than it was back in the day. But you took a risk, and you managed to get a lot of clients using this. And as far as I know, BitGo never lost any funds, which is good. I hope I don't jinx it now.

1:09:32Mike BelsheAnd we've never lost funds, we take a lot of precautions to secure those keys. We use the two out of three model, by the way, thank you for your compliment. Back, if you just dial the clock back to 2013, I mean, people were losing their money all time. They would either, you know, lose it themselves, or maybe on a hard disk that they stuck in the landfill. They would forget their passwords. And then they would try to use web wallets. And so web wallets became this thing had a nasty name, because you know, they get rug pulled, or like the server gets hacked, etc.

1:10:05By the way, when BitGo first launched, there was a lot of confusion, people like, oh, that's a web wallet, because it looks like a web wallet. Right? I mean, you log into a website, and it provisions a wallet for you. Now, of course, it's provisioning it on your computer, and you're getting a printed out paper backup, as well as your your backup key. But it looks like it's a web interface. And then to sign, you just sign in your browser, it looks like it works. And then the browser cosigns the other side, but it's actually a hybrid model.

1:10:32And it can be used both with cold storage and with hot storage. So it's very flexible in terms of what it can do. But yeah, you're right, the usability can look very, very smooth and transparent.

1:10:46Vlad CosteaI want to move the conversation towards the topic of Bitcoin narratives and what it was like in the early days. And sometimes that, sometimes we forget what it was about. And before I do that, I got to present the, not the last sponsor, but the last that I will be introducing in this episode. It's NoOnes. It's created by Ray Youssef, formerly of Paxful. And he's trying to follow this philosophy of banking the unbanked. And this is the idea from which I'm starting, because he has this ideology and he basically lives it.

1:11:23He embraced it as his lifestyle, that he wants to help the global South. He believes that there are a bunch of lions, that's what he calls them, who are hustling and working very hard to make some money. And he created this peer to peer platform that enables them to trade gift cards and use various cryptocurrencies for this purpose. And everyone gets a cut of the deal. The way that he likes to present it is that everyone eats. And it's pretty cool. Check it out. If you have gift cards to sell or you want to provide liquidity to the platform and you like the cause and you want to support it's not just humanitarian.

1:12:03You can actually build a business out of it. It's called NoOnes.com. And it follows this ideology that Bitcoiners used to have, that Bitcoin can be used to bank the unbanked. Nowadays, I think that the people who change their minds, we're just using this as a way to virtue signal, to say, you know, Bitcoin is not just for my personal enrichment, I'm also using this for something like a greater cause, something good. But that use case never really took off.

1:12:35Mike BelsheLet's see a couple of things. I've known Ray for a long time. So he and Artur, the founders of Paxful, became BitGo clients, I think back in 2015, by the way. They came on for a slightly different reason, which is, turns out, running all the infrastructure and the nodes and getting transactions and things. They came over to BitGo in part for the multisig, but also for the stability that they got once they moved on to a more professionally managed service. So they were able to hold full self-custody at Paxful and yet get some of the benefits of having somebody else operate the infrastructure.

1:13:08Let's see, in terms of the why are we here, I think it means a lot of things for different people. I think we are building a new monetary system. I think we're building a global monetary system. I think we've never had one before. And then some of that is empowerment to the individual user for storing their wealth as a hedge against inflation. Some of it is for having private transactions, which we talked about earlier, which I think we haven't fully realized that vision yet. Some of it is for having the power where you can get off of the banking system so that they can't block you for arbitrary reasons, which happens all the time.

1:13:47And we haven't made great progress there. I don't think any of these things are bad, and I don't think you have to be in favor of all of them. And if you're not in favor of all of them, then that means that you're somehow bad yourself. Look, the great thing about this decentralized technology is that you are not subject to the whims of adversaries trying to attack you, and you get to do with it as you will. So I do hope that we get further on the privacy and self-sovereignty components than we are.

1:14:17I do agree a little bit for myself that protecting against inflation is good, but we have a lot more to go. I think the regulatory turn that we just saw in the last few months here in the US is going to have tremendous impact towards starting to go down the next path. I mean, the governments really have been fighting us and they've been unwilling to think, is there another way to do this? And all of a sudden we have, I think, a far more cooperative administration, which will allow us to unlock a new wave of innovation that simply wasn't possible under the previous administration.

1:14:56I know you're in Romania, but I've spoken to regulators, we've got regulators in Germany and Dubai and Singapore and all this. All of the regulators very much look to the US for some amount of leadership and guidance. It's not to say that they follow exactly what the US does, but definitely having the US finally be open to digital assets allows them to think about it with a little bit more safety in how they move forward. So I think I think this could be good and we get to start talking about more self-sovereignty now.

1:15:32Vlad CosteaYeah. Here in Europe, we have MiCA, M-I-C-A, right now, which is entering full force and it's going to make custodians share more information between each other. So if you are registered, for example, to Coinbase or Bitstamp or whatever, they're going to share more information about yourself, even with other exchanges and the tax authorities. It's going to be tough. I think there is mandatory KYC AML for all on-ramps right now and off-ramps. It used to not be like this. There used to be some limits.

1:16:04So, for example, if you use an ATM, you could make transactions up to two thousand euro without KYC. And that was pretty cool. Now that's going to be a thing of the past.

1:16:18Mike BelsheYeah. Look, there's been a war on cash not related to crypto for a long time. Right. Like most people don't know this. But in France, like it's illegal to buy something that costs more than a thousand euros in cash. You're not allowed to do it. And it's the same here. Okay, so... So there's been an increasing war on cash. And then I think regulators need to start having an open mind, legislators as well, as to what are they really trying to accomplish.

1:16:46And so you will hear, you know, it's about money laundering. It's about terrorist financing. Nobody wants money laundering. Nobody wants terrorist financing. BitGo, Coinbase, you know, all the big companies are here to help make sure that that doesn't happen. However, the idea that the only way to accomplish it is by surveilling the innocent. I'm not sure I agree with that statement. Things like travel rule. I'm not sure that they work at all. And there's definitely ways that we can start to deal with this problem that's very different.

1:17:18And we can put it into cryptography. We can put it into zero-knowledge proofs. There's ways to deal with this. So then the second layer you have is that while regulators might say it's about money laundering and terrorist financing, they might actually believe it's about control and protecting the state. And it's really hard to kind of actually think Bitcoiners see it pretty obviously when a legislator is on one side or the other. But to a lot of people, they're still confused. Right. And so for the statist, they're really about control.

1:17:52And they will hold up money laundering and terrorist financing as the reasons they need the control. However, those money laundering programs and terrorist financing programs may not even work. And I think we're seeing this play out at the global scene today. Prior to digital assets, regulators had the benefit of kind of creating these little gatekeepers around each of their jurisdictions. So in Europe, you mentioned MiCA, that's the latest set of gatekeepers. In the U.S. we've had them. It's been the banks. But these things have really inhibited us having true global financial systems.

1:18:29For instance, you can't send money from the U.S. to a vendor in Europe overnight on a Sunday night to get business going the next day. That is something that the banking system ought to be able to provide. A truly global financial system ought to be able to do this. Bitcoin, digital assets, of course we can do it. Now the question is, how do we do it in a way that regulators get comfortable with? And so the regulators have never had to address the question before because they just prevented any of this from happening.

1:18:59But the cat's out of the bag. Right. At this point, we do have digital assets, which people can use in peer-to-peer systems. That is not stoppable. The next thing is, how are we going to evolve it? I do think regulators get to come back to table. One last thing for the regulators, because I don't mean to criticize them so much. Remember, they created these rules, even when they're well-intended, at a time when we didn't even have computers. Like, how do these systems work? These are all designed from 1990, 1980, 1970.

1:19:31Of course we can do better in 2025 than we did back then, given the machine horsepower we have. Of course we should have zero-knowledge proofs that provide privacy. Of course we should have zero-knowledge proofs that help make sure that we don't have money laundering and terrorist financing.

1:19:49Vlad CosteaYeah, well said. And I'm actually curious, you mentioned zero-knowledge proofs. What is it in Bitcoin that gets you excited? And what is it in altcoins that gets you excited? You can name a few that you think are interesting.

1:20:03Mike BelsheWell, we talked about Zcash already. I mean, it's funny, you and I seem to be like-minded on this one. I've been a fan of Zcash for a long time. I spoke with Zooko not too long ago asking about, was it the big flaw in Zcash that shielding is optional? You know, you can have a shielded mode and an unshielded mode. Here at BitGo and our regulated entities, we only support the unshielded mode as opposed to the shielded mode. And that's because the regulators prevent us from doing so.

1:20:30He doesn't agree, by the way. He thinks that actually having both is still good. He might be right. He's a smart guy. So, look, I'm excited about getting privacy fixed. And I'm also excited about, you know, using multi-party or multisig technologies to solve the jurisdictional type problems. We really do need this. You can see it coming. You might be able to see it for yourself. Anyone listening should be thinking about, like, again, what's the threshold of net worth that they're willing to have self responsibility for?

1:21:00It is a big, daunting responsibility with a lot of different ways to screw it up. And, you know, we need to make it way, way easier to use and yet still protect against all of the principles we were trying to protect all the way back to before I was even in Bitcoin in 2010 and 2011.

1:21:18Vlad CosteaI remember when Taproot was being promoted and there was consensus that there's a need for Schnorr signatures in Bitcoin. Taproot did not deliver all of that, but it brought some part of it. And one of the arguments being presented by some overly enthusiastic people was that Chainalysis is going to become obsolete because all transactions will look the same on the blockchain. But, of course, this was only about a specific type of multisig using Taproot, which looks like any other type of multisig or any other type of Taproot transaction.

1:21:52But we never got much network effect for that one. And it's funny that we fall for these narratives and we believe that something is really going to push innovation forward. And then we discovered that reality is different. And I was going to ask you if you're excited about privacy technologies as much as I am on Bitcoin, not just Zcash. And what else? No, this one first and I'll get to the second one.

1:22:25Mike BelsheWell, let's see, I think as long as you've got transactions and trading going through regulated entities, Taproot doesn't really matter. They will be able to trace those transactions after the fact. As you know, once the Taproot is revealed, when you sign a transaction, it gives a lot of information back. So I like Taproot technology, don't get me wrong. I think it is enabling. Schnorr signatures, I think, are also good. Signature aggregation scheme, you can't tell who provided the signature with Schnorr, but then with Taproot, you can kind of put the two together and get some accountability.

1:23:02So I think it's a good tool, but just because it's a good tool and a good step forward doesn't mean it's sufficient. So I do not think transactions need to be traceable on chain. I think you ought to be able to deal with this from the regulated side and that just like you should be able to pay cash. And I guess here I'm getting into my own philosophy. Look, I think two parties entering into a transaction as free individuals to enter into whatever that transaction may be, of course that should be allowed.

1:23:32And so I disagree with the war on cash. And so the strongest way to solve this is by having all transactions on chain fully private, whether that's with the zero knowledge proof or other, I guess, is yet to be seen. Back to the human nature problem, there's so many problems to tackle with just getting into digital assets, understanding what it is. How am I going to secure mine? Do I use a hardware wallet? What's this crazy mining stuff? Who is that Satoshi guy? And there's so much that goes into just like getting your head in the space.

1:24:04Frankly, most people don't differentiate what we're talking about here with privacy of, OK, well, after I actually start using it 10 years from now, I'm getting my paycheck in it. You know, am I going to are all my peers going to know how much money I make just because I received my payment, my paycheck in Bitcoin? Yes, they are. Unless we change it. So I think most people just haven't grappled with that problem. It's a second-tier problem from getting in digital assets in the first place.

1:24:34Vlad CosteaDo you believe that this increase in regulations is also going to create more demand for privacy and I mean stuff like Monero, Zcash, whatever, and it's going to build parallel economies? Or do you believe that this is still a niche that will not really take off?

1:24:51Mike BelsheI'm afraid it's going to go like TCP/IP where it just takes so long to get applications and things built on top of the protocol when it's in an un-private form, a public form, that by the time we actually have the applications and things on top of it that need the privacy, it's going to be really difficult to retrofit. The good news is, you know, look, our software does move faster than it did in 1970 or 1980, but I'm afraid we're going to miss on privacy for decades.

1:25:28Vlad CosteaReally? That's your prediction that we're not really going to have privacy for decades?

1:25:33Mike BelsheYeah, I hate to say it. I don't want that to be the answer, but I think, you know, too few people understand and too few people care until it's just too late. Just let me give you an example of what it means to not have security in TCP/IP. So TCP/IP and starts off, we're making connection to each other. I'm going to talk to you. I send you a little message, a SYN, a SYN message. You're going to send me an ACK message. I'm going to send a SYN-ACK message.

1:26:00These things are completely unprotected. So anybody that wants to just interfere with communications, SSL or not, doesn't help you. You can just muck with the TCP transport layer and prevent you and I from talking to each other. And that's today. And then, you know, there's been a little bit of transport improvement on the heels of HTTP/2 and the SPDY protocol. There's another one called QUIC, came along, Q-U-I-C, which is a little bit better for being fast, but it has the same problem. Anybody can inject packets into the middle of it and trick both sides, basically preventing communication.

1:26:33This is not really a solvable problem without completely upending. There's one secure transport protocol I'm aware of that use public key cryptography from a guy named Dan, gosh, University of Illinois, Dan... Okay, it'll come back. Anyway, it would basically completely upend the control of TCP/IP, but it's really hard to retrofit. And how do you get people, all these machines, all these computers that have been deployed with TCP/IP to ever deploy with another transport protocol? By the way, it's part of why they said that you'd never be able to get HTTP 2.0 done either.

1:27:06You can get it done, but it takes a lot of will and it takes a lot of work and it takes 10 years. So the fact that we are basically, you know, the starting gate for privacy today tells me that we are at least 10 years away from anything meaningful on blockchain privacy.

1:27:24Vlad CosteaAre you excited about Bitcoin becoming digital gold, but gold meaning that it's held by custodians and it's expensive to transfer? Or do you believe that digital gold should have had an entirely different meaning that you can mine it yourself and it's free to move around?

1:27:44Mike BelsheWell, I think you created a little bit of a false dichotomy there. Like there's a blend of these elements. But the number one thing that Bitcoin does today is the fixed supply, the immutable protocol that you can count on for storing value. We talked about a number of threats for how people might take that value from you, but the value part is absolutely there. So it is digital gold and it's purely better than a physical manifestation of a gold bar. It's more difficult to counterfeit. It's more easy to transport.

1:28:22It's not subject to increased inflation just because the price of the gold asset goes up. Most people don't know this, but you get about one and a half, two percent inflation in gold every year with new gold being mined. And as the price goes up, what do people do? Well, they go mine in harder to reach places. So the inflation goes up, too, right? So inflation on gold keeps going higher as the price of gold goes up. Bitcoin is the only product that is immune to that effort.

1:28:48So I am very excited about Bitcoin as digital gold, and I think it does a great job of it. So I guess I disagree with your characterization on the alternate side of it.

1:29:01Vlad CosteaWhat would you say is the role of Ethereum? Because when it was launched, it was promoted as a platform to build anything like a world computer, something for unstoppable applications. And then it became the DeFi machine, the ICO layer and stuff like this. But it seems to also play a role in the development of Bitcoin directly with WBTC and indirectly with the rollups and the stuff that's coming back from Ethereum to Bitcoin. And I'm wondering, what is your perception of it? How do you understand it? Because BitGo has been one of the first companies to embrace Ethereum and integrate some of its features to the customers.

1:29:44Mike BelsheWell, look, I mean, having the smart contract programmability is incredibly empowering. I mentioned earlier the DeFi thing. I'll go deeper on that in a second. The first incantation of Ethereum was interesting. I think it was Ben Davenport, my co-founder here at BitGo. He first said the problem with Ethereum is that it's one viral app away from collapsing. And he was absolutely right. Do you remember CryptoKitties? Right? So CryptoKitties came along and it just crushed Ethereum. It couldn't scale. And, you know, the everything computer, unfortunately, it has problems.

1:30:26Now, of course, Ethereum's come a long way since then, right? So we've got sharding and all kinds of other things. We've got layer twos coming in. So some of those problems are addressed. But I think Ethereum provides a really valid role for how do you put smart contracts into a decentralized ledger. I think they've achieved a decent level. You know, nobody's perfect on decentralization, but a decent level of decentralization. And so that's good. And then for running DeFi products, I mean, it's way better than I'm sorry, I shouldn't say way better.

1:30:56It does something that Bitcoin simply can't do. So why is DeFi important? Let's talk about that for a second. You know, regulators have regulated here in the United States this thing called a broker dealer for decades. And the problem is, is that, you know, forget about crypto, digital assets. Years ago, broker dealers, you know, they're responsible for helping you buy and sell your stocks. All right. And then sometimes those guys lie. It turns out they might tell you to buy a particular stock while behind the scenes they're selling it. Right. So these regulations come in play to prevent against pump and dump schemes, prevent against lying, et cetera.

1:31:37And once you're a broker dealer, you are subject to all of these regulations for your entire lifetime. So, you know, you might be a 25-year-old, you know, starting out, you're a very honest broker dealer, you're great. But when you're 45, who knows, maybe you have some kids going to college, some pressures, you know, and all of a sudden you feel a need to rip off some of your clients. Because of this, when FINRA and the SEC monitor you as a broker dealer, you have to be monitored for your entire career.

1:32:08Every year, every few years you got to take new tests, you got to be updated on regulations, every message you ever send off your phone, they monitor it all. Okay. You put a broker dealer into a smart contract and now the regulator, all they have to do is audit the code. They become code reviewers and you do that once and you don't change the code and then all of a sudden that is a better broker dealer than we have in the existing stock market. And you don't have to do any work for the rest of the life of that smart contract.

1:32:38If you change the smart contract, you got to re-audit it, but that's it. Otherwise it just works dutifully, right? So DeFi has tremendous opportunity for regulators to make better financial products, right? You don't have to worry about is the guy ripping you off. It's just a piece of code and it's already been audited and reviewed. So anyway, Ethereum, I think bringing DeFi to life is an incredibly massive value add. I think regulators, if they were smart, they would be embracing it. I suspect regulators are getting mixed messages from banks and traditional finance people that are like, wait, that's my job.

1:33:15No, you need me in that job. Those crypto guys, you know, they're going to, they're going to, no, no, no, no. The code won't steal, but the people will.

1:33:25Vlad CosteaTomorrow is January the third, which marks the 16th anniversary of Bitcoin. And it's really impressive how far we have come in only 16 years. I guess most people who got in maybe in the first two or three years, they didn't see it as potentially becoming this big. Maybe that they were thinking it's just for some libertarians to trade some coins over the Internet. And then they set up some marketplaces. And here we are in 2025 with such a big project that's very difficult to improve because it's hard to get people to agree even on inserting some extra opcodes.

1:34:05And it's also so big that governments are getting involved to purchase and not just regulate. And I want to ask you, based on the experience that you have with TCP/IP and the Internet and the way that it was developed, where do you think Bitcoin is heading in the next 10 years or 16 years or even more?

1:34:30Mike BelsheWell, let's see. I think I think Bitcoin is mostly locked. I don't think it's going to change a whole lot. There will be, I think, some incremental changes on the edges. Some of the software and hardware will get better. I think for the most part, the improvements for Bitcoin are going to be at the wallet and application layers so that we can keep the safety that comes with the immutability of Bitcoin and then move the innovation into higher levels. For smart contracts and other things, those are going to continue to evolve.

1:35:01For privacy, that's going to continue to evolve. Those things are going to evolve outside of the Bitcoin layer one and instead evolve on top of it. So 10 years from now, I think we're still going to have Bitcoin. I think it's still going to be growing strong. And I think we'll have much better adoption. Lightning might be farther along at that point. I think Lightning is now competing basically against all of the other L1s and L2s for is it a good payment system or not? And I think it's not been moving fast enough, frankly. So the other, the other chains are moving more quickly and could win.

1:35:37Vlad CosteaI think I'm the guy who paid one of the biggest fees for a Lightning transaction. Basically, I had funds on Phoenix Wallet. And what Phoenix Wallet does, I'm not sure if it's recent or not, but it's an LSP, which is a Lightning service provider. And I sent, I think, $10,000 worth of Bitcoin from Phoenix to Kraken because Kraken is one of the exchanges that does support Lightning Network. And I was super concerned that maybe the blockchain analysis, whatever, is going to flag my coin. And I couldn't wait for whatever compliance team to reach out to me because I've had some bad experiences.

1:36:18I also use CoinJoins sometimes for reasons which concern my own privacy, not that I have actually I have everything to hide. That's my answer to it. But I'm not a bad actor in any way. So I use the Lightning Network and I sent this $10,000 transaction and I paid a one percent fee. And that turned out to be a $100 fee on the Lightning Network when the fees on chain were like two cents. And it was super frustrating to realize that I could have done that.

1:36:48But it's just that I wanted to use Lightning for this particular use case and I lost $100 in fees. And that's one of the reasons why Lightning was created in a high fee environment. It was supposed to be this tool that actually provides you an alternative that's cheaper. And I posted this on Twitter and people started telling me, you know, Lightning is not made for these large transactions. You know, you didn't actually pay a lightning fee. You paid a fee to the service provider because they are providing the infrastructure for you.

1:37:22And you should run your own node, which I did for a while, but it's very cumbersome and not the nicest user experience. And it was like a full time job. Sometimes you had to rebalance, reopen, and I just didn't enjoy it. But it is disappointing that even if they do fix lightning and it becomes more reliable, it just means it becomes more centralized. And the fact that if you want to use it non-custodially, if you want to receive one Bitcoin, you need to already have one Bitcoin in the channel. That's also roadblock in adoption unless you're going to go full custodial.

1:38:00Mike BelsheYeah. Look, I mean, first off, sorry. Sorry that you lost some money there. Consider it tuition that you just paid into kind of learning how finance and things work. There's going to be a lot more of these that come in the next 10 years. So more is going to move on chain. I think stablecoins are going to take over a ton of regular day-to-day payments, and they are just better. But they are peer-to-peer, which means it displaces some of the benefits that we've built into the system we always complain about.

1:38:34So let's talk about credit cards for a second, because you mentioned a 1% fee or whatever there. Well, credit cards take insane 3% to 5% fees, right? And the credit card company will say, no, it's not 5%, but it is. When you add up the side that the merchant pays and the side that you pay, it is. But what do you get for that? You get insurance as a retail user that if you're getting scammed, they're going to get your money back. There's all kinds of consumer protection laws that you can buy anything.

1:39:02And all you have to do is say, uh-uh, that was fraud, and then the credit card company is going to go look into that. So you're paying insurance in order to have a safer system. Now, maybe you don't want that insurance. Maybe when it's a large transaction, like a $10,000 transaction, maybe $100 is too much insurance. Maybe you would have done it for a smaller transaction, et cetera. But you actually are making trade-offs. So imagine the world where everything is digital, and when you do it peer-to-peer, if you get scammed, you're on your own, nobody can help you, sorry.

1:39:34Or if you accidentally got tricked or you fat-fingered it, you put an extra zero on it, there's going to be zero protection for you. Now, if you use a credit card, there's protection for you. So to some degree, I mean, you have a question for yourself. Which one is more worrisome to you? Are you more worried about the government coming and confiscating all of your Bitcoin? Or are you more concerned about, you know, you got ripped off and you had to pay $100 on a lightning fee, or that the guy that you're sending to, you accidentally messed up the transaction, you sent too much, and you want that insured?

1:40:09Which one do you want to protect against more? Government intervention or consumer protection insurance for small transactions?

1:40:17Vlad CosteaYeah, that's a good point. But I use these protocols sometimes for the sake of it, just because I want to support them and increase demand for them. And if I use a credit card, even if it's cheaper, and I do not fear that my government is going to get involved, I'm not really helping anything develop, you know, I'm not supporting any sort of exciting technology.

1:40:39Mike BelsheI understand. Yeah, you know, your case here was not exactly the case that I threw back at you. But I do think the trade offs of different types of money is a real one. And it makes it difficult for all of us to think about what's the right evolution for a Bitcoin and Ethereum and all the other digital assets. And I think we might need to stratify our different use cases a little bit differently than we do in the in the old school monetary system. So my thought, number one, I want my Bitcoin to be a hedge against inflation, the government can go do whatever it wants with its money.

1:41:16That's my number one. That's my personal choice. Other people put self sovereignty right up there, not being confiscated by the government. It's a legitimate thing to be worried about. If you try to create a single system that is both going to protect against the inflation and government seizure and also covers all of your just day to day purchases, you're buying some art, I don't know, et cetera, fat fingered if you want some insurance on, et cetera, it's going to be a very difficult thing. So we probably just want to split these cases apart and provide slightly different solutions.

1:41:49I think that'll be the easiest answer, even though that isn't the answer we had in the past. In the past, you've got your money and your money is your money. And, you know, it's kind of usable in multiple purposes.

1:42:01Vlad CosteaYeah, but at the same time, there's always a price to pay when you're an early adopter. Sometimes I watch videos and I get a lot of inspiration from the Computer Chronicles, which is a show from the US that was presenting computer hardware and software in the late 80s, early 90s. And I think they ran up until the 2000s. And I do see people buying Apple computers and all the accessories and all the stuff that they were paying for. And not just Apple, also IBM ones, like they were paying thousands of dollars for something which today is like 200 for printer or whatever.

1:42:39The cost of being an early adopter is always high. But sometimes there are benefits. There are educational benefits for yourself because you're ready for what's coming. And you can understand what's the next wave or you can provide useful feedback to improve the technology. Or you can just think about it as a way of supporting these businesses because you do want this to be part of the future. And if you do want this to be part of the future, sometimes you have to pay for it. I do see a freeloading problem in Bitcoin with people who say, yeah, I'm never going to spend any Bitcoin.

1:43:14I'm not going to send any transaction. They're secure in cold storage. And the question is, OK, but when you need to send them for whatever reason, what if you're not going to have the kind of security? You're not going to have miners that do want to put your transaction in a block because you're not incentivizing miners to keep going right now. So if you're not paying for security today, you're just hoping that tomorrow there's going to be someone else who is going to provide the service to you.

1:43:42And I don't know if there's any solution to it. The most extreme one is proposed by Peter Todd and he said we should just add tail emission. He calls it demurrage and he published a paper about it and said, you know, it's not really inflation because there are a bunch of lost coins and whatever gets mined extra is just accounting for these lost coins. It's super controversial. I don't think I would support it unless it was like a life and death situation for the Bitcoin network. But it is a problem because there needs to be someone and maybe that this is the biggest gamble in the way in which Bitcoin was designed, that there is an assumption that's also outlined in the white paper that after the entire supply gets mined, after the inflation runs out, there's going to be this market for transaction fees that's going to support the miners.

1:44:38And we're not sure if we're really going to have that. And if you look today at the mempool, I think it was close to clearing. There are one satoshi per vbyte transactions getting gathered into blocks. I think Ethereum has a much healthier fee market, which is ironic, and I don't really know what's the solution for this. Is it a culture problem or is it a technical problem?

1:45:02Mike BelsheWell, there's I think a number of unanswered questions about how does Bitcoin continue to work perfectly over the long term. One of them is just like, how do we pay for the miners, as you're talking about? I think that in general, we're going to sort this out and it's not going to be so bad. But I think the industry is evolving in ways that, you know, certainly Satoshi didn't know exactly how things would unfold. For example, you know, you mentioned wrapped Bitcoin a number of times. If the usable version of Bitcoin ends up being on other chains, then is the transaction capability all moved off of the Bitcoin chain entirely?

1:45:49And then nobody's transacting on the Bitcoin chain. And, you know, as the block reward goes down, will it be able to sustain itself to pay for all the hashrate? It's an unknown question. I think it's going to solve itself. I don't think it's actually an issue. I think it's going to be a constant kind of pendulum swinging where as the fees get too high, people look to offload those fees. As the fees come back down, you know, the usability comes back into play and then they're able to use it again.

1:46:17So, you know, that's an unanswered question and I think it just gets figured out. Another one that you hear people talking about a lot lately is, you know, quantum computers and will they be able to crack cryptography? Like, I think it's good that people are worried about that and thinking about it early. I think the quantum technology is actually quite a bit less far along than we need to worry about. I think we got at least 10 years. This is something that probably does have to get addressed at some point.

1:46:45So I don't know the fact that there's unknown things that still need to be solved. To some degree, every financial system is a bit of an experiment. Heck, even U.S.-based capitalism, you know, it's a bit of an experiment. You know, there's this experiment called democracy, it's called capitalism, and is it going to work over the long term? We tweak the rules a little bit to try to make it better. Sometimes we agree with those rules, sometimes we don't agree with the rules. I think Bitcoin goes through some of those same gyrations.

1:47:17Vlad CosteaYou're super optimistic, but you did outline something which Paul Puey was more, how should I put it? He was more nonchalant in saying, you know, I believe the future of Bitcoin is not on the Bitcoin blockchain. It's going to be on some other layer, some other network that's going to perpetuate the system with 21 million coins. And maybe it's going to be another blockchain, maybe it's going to be a database or something else entirely. But Bitcoin can survive on the long term without miners or can survive as a proof of stake system.

1:47:51That's another conversation that's interesting because this can be an option, even though it's ideologically opposed to what Bitcoin is and the hardcore users of today oppose it. You never know what's going to happen in the future and to which extent. Right now, mining is a race to the bottom, searching for the cheapest form of electricity. You never know when that point of, you know, the lowest denominator is going to be hit. And under that threshold, you cannot get cheaper electricity and maybe industrial mining is not going to be profitable anymore.

1:48:27And you go back to home mining and it's going to be the users sustaining the network, possibly out of support, you know, as a hobby, because they do want the network to succeed and they do need the transactions to get mined.

1:48:43Mike BelsheYeah, look, I do. I do agree that I think Satoshi was, if Satoshi existed today, I think he would be surprised by how quickly. Mining got centralized, I think that was an unanticipated thing. Pools of mining pools also, I think, caught him off guard. I think he really did envision a world where everybody's contributing to mining, everybody's contributing to their own nodes, and yet we're far away from that today. So, look, this thing's going to evolve and we are humans and we get to figure out how the path goes forward.

1:49:16Good news is it's highly decentralized, so it takes a lot of humans coming together. But, you know, some people that think it's completely, I guess, invulnerable to change, I think they're partly mistaken. Here's an example. Let's say tomorrow a hack was found for ECDSA signatures and, you know, large amounts of money can be taken from kind of anywhere. What would happen? Well, we all just say, ah, shucks, I guess Bitcoin didn't work. And when it goes to zero, we'll go back to something. No, that's not what's going to happen.

1:49:50What's going to happen, the community will come together, they will figure out a fix, a patch. Some people will be vehemently upset about it. Other people will be like, this is the best compromise. But in general, I think we will get to a decent solution. We will fix the vulnerability. Bitcoin would take a big hit, but we would move forward and move on. And look, I don't think that's going to happen. It's going to be a cryptographic hit against Bitcoin. So please don't take it the wrong way.

1:50:12I think actually to some degree, people should take comfort in the fact that like, you know, this is what would happen if we got to the point where something was just completely broken with Bitcoin. People would get together and we would figure it out and we would rebuild the blockchain. So Bitcoin has a huge network effect, which is being built and getting stronger by the day. And it makes it very difficult to replace. So while I do understand, And in fact, I commented on it, Paul Puey, you said he mentioned it as well.

1:50:43Like maybe all the Bitcoin transactions end up moving or significant numbers of them get moving on to kind of other L1s or L2s in certain ways. It's pretty hard to replace the Bitcoin brand. So I think whatever does end up happening, even if it's at the application level where it's tied together, I think that concept of Bitcoin remains for a very, very long time, potentially even longer than the technical implementation.

1:51:10Vlad CosteaOn the topic of mining, Satoshi even wrote emails to Laszlo Hanyecz, who was one of the first people to figure out how to use GPUs for mining, and told him something like, you know, I'm not a socialist of any kind, but I would really appreciate if you gave back some of these coins to the community, or you stopped mining. Please stop mining with your GPUs, because I want more users to join and use their computer, their CPU, their processor, to get some coins and understand how the network works. And I think that the pizza purchase that he famously made was part of his campaign to give back to the community.

1:51:50Mike BelsheWell, you know, that part of the history a little better than I do. So I think it supports this idea that like, I think Satoshi was caught a little bit off guard, how quickly, you know, mining advancements with moving GPUs into ASICs with pooling, etc.

1:52:05Vlad CosteaYeah, that was a big deal. He actually responded to the first mining pool on BitcoinTalk. It was Slush, bitcoin.cz, I think, at the time, or something along these lines, Slush Pool. He did say that it's pretty cool, but he did not foresee this. He had this naive idea that everyone was going to compete with each other and use their own CPU individually and not put their hashing together.

1:52:31Mike BelsheYep.

1:52:32Vlad CosteaSo yeah, whenever you look for wisdom in Satoshi's words, just remember that he did not know everything. He was not a time traveling alien. He was not some sort of genius. You know, he put together some elements, you know, and it's very hard to believe that he was a three letter agency, because if you look at the 0.1 version of the code, it's basically super hacky and uses instructions in C++ like printf. Which are a bit amateurish. And the way that the structure of the code...

1:53:11Mike BelsheAre you a C++ programmer?

1:53:14Vlad CosteaSort of. Amateur. Like, I would never claim to be this smart to invent Bitcoin, but the way that it's structured is not something that you see, you know, when you're looking at the code of a seasoned C++ programmer. And also he had a broken marketplace and the game of poker that were part of the code and later removed. Why the hell would some sort of professional NSA cryptographer who publishes this to enslave the entire world with this surveillance system put that in there? And of course, there is a conspiratorial answer for that.

1:53:49But I do believe that Satoshi was well intended and was this pure researcher, possibly an academic, who found ways to combine existing elements into something that previously did not exist, or at least not in this way. And it's pretty cool. We're still using it today. It's still getting improved and refined. Maybe it's going to make the world a better place or maybe not. Who knows?

1:54:15Mike BelsheI did a lot of coding when I was coding, and probably the language I coded in the most was C and C++. I think I have empathy for Satoshi, whatever he did. I don't know that it's evidence of much other than one thing. You've probably read the book Zero to One, Peter Thiel?

1:54:36Vlad CosteaNo, I haven't.

1:54:38Mike BelsheOkay, well, Zero to One discusses a number of things. But one of them is that there's builders that take something from nothing and turn it into something, one, and then there's people that can take that something that works and turn it into something that's scalable and awesome and great.

1:54:56And the challenges that you encounter in each phase, that is going from the zero to one phase, getting off the ground, and then going from the one to making operationally efficient and scalable and performant is very different. So to me, I think Satoshi looks like he was very much a zero to one type of guy. It's a unique talent. It's something that some people that can be very polished, great programmers, figure out exactly how to scale these things, et cetera, the zero to one people often don't have.

1:55:26But those one to many guys often don't have the ability to kind of take these primitives like, oh, wait, how do I do this SHA hashing? Combine that, you know, with cryptography, combine that to make mining, you know, they wouldn't have that skill, per se. So I think it is two different skills. He clearly was on the zero to one side, whether he's from the government or not. I don't know. I don't really think so. But I try to not pay too much attention to it. You probably know more.

1:55:53Vlad CosteaYeah. And tomorrow is the anniversary of this beautiful network, which is still running. You mentioned something about coordinating to fix issues, possibly having a hard fork. I guess that's what it means. There was such a case in 2014 when Mike Hearn tried to upgrade the database system. It was Berkeley DB, if I'm not mistaken. And basically the network went down for a few hours and everyone got together, all the mining pools, all the nodes. And they ran the previous version and then they upgraded to one that was actually working and found consensus.

1:56:32And I do remember Vitalik, I think, published an article in Bitcoin Magazine and argued that this was a proof of centralization in Bitcoin. But it's the kind of centralization that's acceptable when people find the situation of crisis and they finally agree on something, that they need to do something to address whatever the case was.

1:56:57Mike BelsheLook, there's been a few of those. And, you know, sometimes you find a bug which could be critical. And fortunately, it's often found by white hat hackers. And so they fix it and figure out how to streamline that in. We talked about Zcash a fair amount. You're probably aware that there was a bug in the zero knowledge proof side where somebody could have, in the early days, printed themselves as many Zcashes as they wanted. And an interesting element of this, which is different than the public chains.

1:57:32When you have a bug like that, that's behind the zero knowledge proof, there's actually now no way to tell if anybody ever exploited that bug because it's inside of a zero knowledge proof. Now, at this point, it's been many years and they fixed it. Zooko could give you all the details in crazy depths. But it's a fascinating concept, which is that that same bug happening in an unshielded program, blockchain like Bitcoin, you can tell if anybody exploits it. If somebody goes and gives themselves a bunch of Bitcoin, you know. But in the, in the privacy world, it actually gets harder, from the technology perspective anyway.

1:58:12Vlad CosteaYeah, I have been dragging this interview for two hours, but I tried really to pick your mind on some topics and maybe ask you the same question twice to see if your answer is going to have any inconsistencies. I'm just crazy about this, but I do appreciate your time. Mike Belshe, you are a visionary in this field. I respect your work very much and I wish you best of luck with BitGo. Last question, how can people follow your work or sign up to BitGo to use the service or anything that you want to promote?

1:58:45Mike BelsheWow, thank you. Look, I'm on Twitter a fair amount. So my handle is @mikebelshe. You'll probably put it in the links. BitGo is bitgo.com. We're all around the globe. So look, if there's anything we could do, check out the site, reach out to us. We're always looking for feedback from people to make the product better. So if it doesn't work perfectly for you, let us know. And then I think lastly, something that I hope that we do, I guess this is New Year's time, maybe a little bit of New Year's resolution.

1:59:15I do hope that we get a better balance in the ecosystem between self-custody and custody. I think there's a few hardcore folks that really enjoy self-custody and probably mostly do it right. I still see a lot of mistakes made, but it's so hard to use. Most people can't. And I think we can build systems that blend, you know, self-custody principles, if not self-custody directly, with full access to custody. So I hope we make big inroads there at BitGo and in the ecosystem.

1:59:47Vlad CosteaThank you very much, Mike. And hope I'll get you again on the show someday.

1:59:52Mike BelsheThanks, Vlad. Take care.

1:59:52================================================================================ Production notes: CORRECTED. Base: the pipeline's local Whisper medium.en transcript (int16) with automatic speaker diarization (Guest A -> Mike Belshe, Vlad Costea). Unlabeled continuation paragraphs inherited the previous speaker. Restored from the audio: the opening 'Hello there and welcome to the Bitcoin Takeover Podcast' (dropped by the base ASR), 'here's a question, Vlad' (15:53), 'anything meaningful on blockchain privacy' (1:27:23). Turn-boundary fragments were moved to the right speaker after checking them with local TitaNet speaker embeddings (2 s windows, 1 s hop, reference voices for Vlad and Mike) and word timestamps: 'But it's bigger than Lightning' (Vlad, 9:01), 'But that's not really what they described in the white paper' (Vlad, 23:58), 'I guess a better example is Zerocoin' (Vlad), 'Can you say the name again?' and 'No.' (Vlad, 29:50-29:55), 'Look, the guy's a liar and a fraud, right?' (Mike, 42:03) and 'And we finally have a lot of liars and frauds in this industry' (Vlad), 'Okay' (Mike, 54:19), 'One size fits all' (Vlad) and 'Oh, that's right' (Mike), 'Are you excited about Bitcoin becoming digital gold' (Vlad), 'I think I'm the guy who paid one of the biggest fees' (Vlad), 'On the topic of mining' and 'Yeah, that was a big deal' (Vlad), 'Yep' (Mike, 1:52:31), 'I did a lot of coding' (Mike), 'No, I haven't' (Vlad, 1:54:36), 'Thanks, Vlad. Take care.' (Mike, 1:59:52). Second opinion: YouTube auto-captions of the published video (read-only copies: zcash-documentary S16E1.en.vtt / S16E1.timed.txt, and btctkvr-transcripts S16E1_Mike_Belshe.en.srt / _by_minute.txt, which are the same captions; YT time = mp3 time + about 0.2 s, steady across the episode), word-aligned to the ASR (about 740 differences reviewed by hand; the captions mangle most names, e.g. 'bito', 'Vio', 'Kasa', 'Vince surf', so the base ASR was kept where they disagreed without support); doubtful stretches were re-transcribed from the published mp3 with local faster-whisper (small.en, int16, VAD) on 61 short clips (about 13 min). Notable name/jargon fixes: Biko/Bitco/Vico/Pico -> BitGo; Kasa/CASA -> Casa ('the Casa guys', raw 'the cost of guys'); Dan Bonet -> Dan Boneh (BLS signatures); Snore -> Schnorr; Flip -> FLIP, Speedy/SPDI -> SPDY, HTTP2 -> HTTP/2; IETF/RFC; TCPIP -> TCP/IP; SIN -> SYN, SYN-ACK; QIC -> QUIC; Vint Cerf; Brendan Eich; Tim Berners-Lee (HTTP 402 Payment Required); Ben Chan (Zero Click); Gavin Andreessen -> Gavin Andresen; BlueMat -> BlueMatt; drive chains -> drivechains (LayerTwo Labs); Rootstock; Blockstream Liquid, de-pegging; Aave; Omni/Mastercoin, colored coins; Fero -> Firo (Zcoin); one-input, two-output SegWit transaction; Cancel on -> Cantillon effect; $5 rent -> $5 wrench attack; 6102; Hodlink.ch -> Hodling.ch; Know Once/KnowOnce.com -> NoOnes/NoOnes.com (Ray Youssef, Paxful); Archer -> Artur (Paxful co-founder); global self -> global South; MICA/Mika -> MiCA; zero and honest proofs -> zero-knowledge proofs; Zuko -> Zooko; ECSA -> ECDSA; WalletConnect, MetaMask; Edge Wallet, Paul Puey; Wallet Recovery Wizard; Taylor Mission -> tail emission (Peter Todd, demurrage); Laszlo Hanyet -> Laszlo Hanyecz; Slush Pool; up codes -> opcodes; CoinJoins; Phoenix Wallet, Kraken; Computer Chronicles; Ben Davenport, CryptoKitties; Peter Thiel, Zero to One; Berkeley DB; Mike Belsey -> Mike Belshe (@mikebelshe). Light clean verbatim: ums, false starts, stutters and cross-talk fragments removed; speakers' own slips are kept as spoken (e.g. Vlad's '2014' for the Berkeley DB fork, Mike's SYN/ACK order). Intro music is omitted. Remaining doubtful words are marked best guess.

← Back to the episode · All transcripts · Ask the Archive

0:000:00