You basically allocated 100 Bitcoin to a wallet and you said, "Go hack this," and clearly it's still sitting there in that wallet, right?
One thing that surprised me, how much people are afraid of quantum computers. I said, "Okay, fine. If you're so great, here's 100 Bitcoins, go get it."
You recently went to the White House. BitGo, which is a big deal in the world of that thing, that little thing that's happening out there. What caused President Trump to mention your company?
Trump was discussing kind of all that we've done over the last 18 months in order to completely change the climate politically.
You've been focused on institutions from the beginning.
We strongly believe retail deserves what institutions demand to get the same features that institutions get.
But if quantum became a reality today, wouldn't you be at risk of losing your coins? Hey everyone, welcome back to the show. Joining me this week is Mike Belshe. He is the co-founder and CEO of BitGo, and BitGo's been in the news recently. I think I heard President Trump mention your company in the last two weeks. So, Mike, great to have you.
Thank you, Natalie. Good to be here. It's been way too long because I think we met before you had this podcast, like you were just getting started. How long have you been doing this now?
This is year five, I believe, which is so crazy.
All right. So, I don't know why we—I think it's because you're New York, I don't know.
I record in New York a lot, yeah, but I'm based in the Midwest. I'm going to keep it private for OPSEC. We're going to talk about security here. We're going to talk about a lot. But no, I mean, what caused President Trump to mention your company? You recently went to the White House.
I did. So, look, it's been a big week. I'm actually pretty humbled and impressed. I've been working on a little thing. I'm impressed by the administration and teamwork, which is not something that you really think of with regulators. We had the good fortune—I sat down with Commissioner Gould from the OCC, I sat down with Secretary of Treasury Bessent, of course I met with Trump, also met with Chair of the CFTC Mike Selig, and met with Chair of the SEC Paul Atkins, all in the span of the last four days.
But the teamwork part—this administration is working together. When you think of regulators, there's been infighting and jockeying for positioning and questions about whether this is the CFTC's turf or the SEC's turf. This administration is like, "Okay, let's chop up the work to be done, let's figure out who should do which parts, and let's go figure out how to make sure we have a Bitcoin/crypto agenda for America," and they're making it happen.
So, sometimes you hear, "Oh, thank you, President Trump, for your leadership." I almost feel like it's said too much. But on the other hand, he's assembled a team that's working together. I think it's an example for other administrations in the future. I didn't mean to go so political; I'm actually just genuinely impressed.
I was here for an IAC meeting, which is an advisory committee to the CFTC, and then President Trump was discussing kind of all that we've done over the last 18 months in order to completely change the climate politically. And of course it's true. 18 months ago, did you see DTCC, did you see Morgan Stanley, did you see Citibank saying that they're going to touch Bitcoin? Well, now you do, right? And that's what happens once you shift from a regulator that wants to regulate by enforcement to a regulator that's like, "Wait a minute, let's understand this problem. We might have to change some things, but let's work on it."
And that's what we have now. So anyway, they're working as a team. They're firing on all cylinders. Obviously, there's been a lot of talk about the Clarity Act, which would help all of us, of course, regardless of which part of Bitcoin, crypto, tokenized equities you're on. But they are working well together and we're going to get a good result whether Clarity passes or not.
Well, so you make it sound like there's been a lot of progress. I would say that a lot of my viewers feel the opposite, that they've kind of stalled on Bitcoin. We're not hearing about whether the United States is going to buy or the strategic Bitcoin reserve. When are we going to get an audit of how much the United States actually holds? So what do you say to the folks out there who feel like no, this administration has largely forgotten this space?
Well, look, I'm not here to answer for the administration. I think that Operation Choke Point 2.0 is clearly over. I know there's a couple of people and there's a couple of edge issues that still need to be solved, but compared to where we were, it is black-and-white different, right? So there's been a ton of progress there.
On the Bitcoin reserve, I think they're sorting that out. Remember, there was a little bit of, "Is it going to fall under Commerce Department or is it going to fall under Treasury Department?" I believe that's all been settled now. It's going to be under the Treasury Department, which I think makes sense. Treasury holds assets; it's usually not Commerce. So that is probably an enabler to get things going. There's legal questions that they're sorting through, so what kind of timeframe it should be exactly, I don't remember. It's only been 18 months and there's been a lot to shake out.
Look, I'm mostly interested in the regulatory clarity around how do we treat digital types of digital assets. That's what enables the digital rails to be possible across all these different markets. So, the market that we're familiar with, of course, is crypto and DeFi and that type of thing. We've been talking a lot about tokenized equities, but also banking, right? Stablecoins have been the talk of the town. And what's going to happen? We're going to have stablecoins first—that's your deposits—and we're going to break out lending onto tokenized chains. So there's just a ton of opportunity here, and it couldn't happen if we had regulators that are just like, "Nope, nope, nope, nope, nope."
Are you feeling like Clarity will pass or no?
But it doesn't matter to you.
Look, there's definitely a lot of people that are optimistic. I guess there's a vote that's coming up. I hope that it passes, but this team of regulators is determined to make sure that we have good rulemaking. So I'd rather have the legislation first and the rulemaking second. But as long as the rulemaking gets done, I mean, it's a massive step forward. So anyway—
Today's episode is made possible by our title sponsor, Ledn. If you own Bitcoin but don't want to sell it, Ledn gives holders another option: borrowing against it. Ledn has operated through multiple market cycles and publishes Proof of Reserves so you can verify what they hold. Its Bitcoin-backed loan rates scale with the size of your loan, so the larger the loan, the lower the rate. Your Bitcoin stays custodied and is never lent out for interest. Ledn now also supports tokenized gold, and gold-backed loans are on the roadmap. Get a quarter percentage point off your first loan at ledn.io/natalie. Terms apply. Visit the site for details.
The future of AI and Bitcoin mining depends on access to energy. Bitdeer Technologies Group, publicly traded on NASDAQ under BTDR, has secured a 3-gigawatt global energy portfolio for AI and Bitcoin mining infrastructure. And because Bitdeer is vertically integrated, it manages everything from equipment manufacturing and data center construction to proprietary GPU-native orchestration software. Learn more at bitdeer.com.
I want to pivot to another reason why BitGo's been in the news lately, which is that you basically allocated 100 Bitcoin to a wallet and you said, "Go hack this." So tell me the backstory of that, and clearly it's still sitting there in that wallet, right?
So we went public earlier this year, and going down the public path, one thing that surprised me is how much people are afraid of quantum computers. People are afraid of various types of attacks on Bitcoin. And it does come from a lot of the traditional investors that haven't been very familiar with Bitcoin in the past. It has been off the table—stuff that they couldn't touch just 18 months ago—and now they can start to think about it. Companies can go public; there's been, of course, a whole slew of crypto companies that have gone public.
This fear about getting hacked in Bitcoin, which is a legitimate fear, of course—that's what we've been working on at BitGo since 2013—is real. And then on the AI side, it does look like some of the AI teams want regulation on AI, and there's a big debate in America about are we going to have a regulator tell us which models we should be using and when and what we can do with it, etc., because they're too dangerous.
Look, I'm a technologist by background. AI is an amazing tool. I think it gives us defensive properties. We are going to be able to use it to defend against attacks just as much, if not more, because remember, we've got the code that isn't public and we can run it on our own repos. Attackers can't do that; they're just coming in through the front door. If we use AI appropriately, we end up with much better defenses than we've ever had before.
All right, but nonetheless, Anthropic in this case—OpenAI has been doing it as well—keep announcing they've got this model and it's so dangerous. It's this monster, and they keep trying to put a cage around it, and they can't figure out how to keep it in this cage. And so we've heard that they had a breakout of the sandbox. Again, if you can't create a sandbox that contains your AI, I think you're not creating a very good sandbox. And I don't think the AI is really as powerful as they want you to believe it is. I think it should be open access to everybody. I do not think we have to be afraid of this at all. And we want teams—the BitGos of the world, the Coinbases of the world, all the banks of the world—to start using those tools immediately. And the faster we get it, the better. So anyway, I kind of got tired of the latest claim from Anthropic, and yeah, I said, "Okay, fine. If you're so great, here's 100 Bitcoins. Go get it."
And obviously, no one's hacked it, right?
I'm pretty sure my phone would blow up if somebody got it. Interestingly, at the time I did it, I think Bitcoin was $65k, so it was like a $6.5 million bounty. And then of course we've had a pretty good run. I think we're at 77 today.
Yes, we had some good rallying in Bitcoin.
So yesterday I announced that the bounty has been increased to over $7 million, but it's 100 Bitcoins.
Well, so for people out there that are listening to this and they don't really understand how the technology works behind these AI models, some of it can be confusing, right? Because we had Qwen come out, and I don't think we know for sure whether that contributed to the Coldcard hack, but it seems like the message was we have these very powerful frontier models happening overseas that now people are using here, and they're almost maybe better or in some ways more efficient than the ones released in the US because the US models actually do block you from doing certain things for security reasons, right?
Okay. Look, in the aftermath of me posting those 100 Bitcoins, I had a bunch of people—they're all trying to use their Claude to come and hack that wallet, and they're like, "My Claude says it won't do it. It's hacking. I'm not allowed to hack a US bank." BitGo is a US bank. So look, yes, I think you need to have models that are able to go and do these things, and whether Anthropic wants to participate in that or not kind of doesn't matter, but we need to have places where this can happen because we need to be able to use it for defensive purposes.
We've been building that inside of BitGo for the better part of two years now. If you're running a crypto company and you haven't been running your own software through AI for quite a while, you're not doing your job. And by the way, that is what happened with Coldcard, because Coldcard could have figured this out, right? Coldcard could have gone and looked at their random number generating; AI would tell them. It's just they didn't look. So everybody should be using these tools. I think the more we make it open, the better.
I know that's scary to people, but open source has always been scary to people on different levels—maybe on a business level or an IP level or whatever. Trying to block it isn't going to work. The AI world is moving at breakneck speed. We all agree with that; you can barely keep up. You think a regulator—
—is going to somehow be able to keep up with who should and shouldn't have access to what? There's also been talk about self-regulatory organizations for this. Maybe we could do something like FINRA. I'm also against those. I really think this is not a place where we need regulators. I don't think it's going to help; I think it's going to hold America back. It's really—I have no ifs, ands, or buts about it. We should not have regulators going after a phantom problem.
If you can define a specific "this is the issue I'm going to fix," great, maybe we can talk about that. But when you're just talking about "I'm afraid that this AI is going to be too dangerous," what does that mean? Who decides what's too dangerous?
Too dangerous for whom, right? And by the way, it also creates places where the government now picks winners and losers. The government has done this: there's a list of companies that have access to frontier models and can use it, and a list of companies that can't. Well, if you're not on the list, you can't, right? So we don't want governments picking winners and losers.
And then even if we do that, the hackers are not bound by this because they've got Kimi, K3, and all kinds of models outside the United States. So it's simply a fool's errand to try to do it this way. Actually, at the CFTC meeting yesterday, they were asking for advice on AI and how to regulate. My answer was hold off, don't need to regulate yet. A bunch of people that were there—
—agree with that approach.
And you want the defense to be just as strong, if not stronger, than the offense that's getting all these tools and there's no regulation over what they're using to hack, right?
Okay. So for people who are listening to this who are not technical and maybe some of it does sound scary: the idea of self-custody is scary because that holds risk, but the idea of holding your coins with a custodian that might get hacked sounds risky. I mean, what do you tell them?
Let's back up and maybe take it away from the crypto circles for a second. You could have cash, right? Do you have all of your assets at home as cash?
Yeah, okay. You don't want to hold cash anyway. But let's say cash was not inflating away, you still wouldn't do it, right? 100 years ago plus, the world moved away from holding onto your own gold coins and your old cash; instead, you put these into banks and whatnot. Do you think it took time to build the trust around how do I use these things? Of course it did.
Now, with trust and with centralization, you get some benefits. The benefit is as long as it goes well, you're happier, you sleep easier, you don't have to put bars on your windows, you don't have to have your security people. If something should happen to you, you can pass it to your kids. There's a bunch of good benefits. Then the bad news is, shoot, you just gave it to somebody else. Can you really trust anybody? And the answer is usually no.
Now, all of a sudden, we've got a digital asset. Digital assets are harder to secure than physical assets. They are vulnerable to the entire 6 billion people on the internet 24 hours a day, 7 days a week. With your physical assets, that guy from North Korea has to come to your house.
So he's probably going to have a hard time getting to your house, not too hard to maybe get onto your phone.
So we have the hardest-to-secure thing, and people are trying to secure it. All right, so I guess I'm probably scaring people there. That said, we have some really good stuff. We have cryptography that absolutely works, that's been reviewed and used in all kinds of banking situations for decades, and it can be done. But humans are not good at security. Individuals—all of us—are vulnerable. We are the weak link. Most of these hacks, it's not usually the crypto; it's usually social engineering. Right now, this Coldcard hack is a little different, right? It is the software vendor failed.
So what does BitGo do? We do the exact same tricks that you would do in physical security, which is if you've got a single point of failure—that is one way that someone can steal the money—you break it into multiple ways. Instead of having one lock, you might have two locks. Instead of having one person with the access code, you'd have two people that are required to come together with the access code. Or heck, if it's the nuclear launch codes, maybe you've got five people that all have to do it. But you eliminate any single point of failure so that any single piece that might be broken wouldn't be vulnerable.
For BitGo wallets—and this isn't intended to be an advertisement for BitGo, I would love it for all wallets to replicate this—multisignature is the mechanism by which you take that single key vulnerability and you break it into no single point of failure. You can have two keys or three keys. There's a number of companies that use multisig. BitGo pioneered multisig all the way back in 2013; that's how we got into this business. Casa is another great one. Jameson Lopp used to work with me at BitGo; we're still friends to this day. He went off and created Casa. He wanted to do another variant of multisig, but it's a good model.
Anyway, both of these have the property that you have independent keys that are required in order to unlock the Bitcoin. I don't know about the Casa implementation, but on the BitGo implementation, not only that, it's separate implementations of how do we do our random number generators on each key.
Remember that single point of failure? If you had the same implementation or if you had the same RNG that had the problem that Coldcard did, then multiple keys are vulnerable.
So multiple keys, and use separate stacks for how you implement each one, and then in order to unlock a transaction, you need two of those to come together. If you look at the probabilities of failures, let's say the probability of a failure on a single stack is 0.01% and the other one is 0.01%, the chances of both of them failing are very, very small. So that's how you take problems and you make them really, really unlikely.
Now, all of that said, self-custody is going to be hard probably forever. I think the $5 wrench attack is still probably the biggest vulnerability that we have in crypto, and that'll be here for a while, but we will get through it.
I just want to back up for a second because you were actually part of the team that invented 2-of-3 multisig, right? I mean, there was a time in Bitcoin where you could only secure your coins with a single signature. Can you just walk us through that and maybe in a non-technical way?
Well, first off, I don't want to take too much credit, actually. Gavin Andresen was one of the early Bitcoin Core devs and he actually deserves probably the most credit for really having pushed through what it is. But my story is this: I was getting into Bitcoin, I was excited about it like a lot of people, and I was going through the journey of understanding what it is, what is money, what's great about it. And I was helping some of my friends buy Bitcoin, and being the technical one, I was the one that stored it.
I was using an air-gapped computer. It wasn't very sophisticated. This was all super early, I would say borderline amateur days. But I had removed the networking; it was kind of a cold storage device, etc., very securely stored underneath my couch. And it started to become worth a lot of money. Some of the people that had money on this laptop are Silicon Valley folks that people know. There had to be a better way to store this. I was afraid my kids would literally come in and spill Coke on it or something like that. Of course, I had a backup securely stored in the laundry room, so if the house burned down, I guess that would have been a problem.
Early days, and so I thought there had to be a better way to do this. I started looking around and I found this little corner of Bitcoin called P2SH, which I did not invent, but P2SH is Pay-to-Script-Hash and allows a little bit more flexibility about how you unlock an unspent that's sent to a particular type of address. Gavin Andresen had been the one that really spent the most time making that happen inside of Bitcoin. It had been in Bitcoin, I think, for maybe nine months prior to my coming along.
All I did was I said, "Look, well, maybe if we constructed this 2-of-3 mechanism, that could be pretty good for humans." 2-of-3—I talked about no single point of failure. The two keys mean it protects against theft: no single point of failure, you can't steal from one machine, you have to steal from two. And then also we use a 2-of-3, meaning there's a total of three keys. The third one is a backup key.
A problem people often have is if you have a single key, okay, you've got that securely stored, but where do you put the backup? If you trust that to somebody, well, then they can steal your Bitcoin. Or if you lose that backup, you're out. So the 2-of-3 model: any two keys can sign, and then you always have a backup. So even if you lose one—because humans lose stuff, right? We should be equally worried about us.
So anyway, I came up with that model and I sent it off to the Bitcoin dev list. Some people were very, very positive. It was such a nice community at the time. Everybody was like, "Yeah, go do this, it's great. I would use that, I'd pay for that." So that led to the first implementation I did at BitGo.
Wow. I wish I was early to Bitcoin during those unified days and everyone not trolling each other. That's true.
You just have to wait longer. So I wanted to get your take on how you think self-custody versus institutional custody will shake out over the years, because it's still, I think, a minority of people that do self-custody. Many people, I think wrongly, believe that it's much too technical. But how do you see, as Bitcoin evolves, what percentage of people are going to self-custody versus trust a custodian?
First off, self-custody is critically important. I think BitGo is the only custodial provider that has a full complement of both self-custody and custody solutions. We start out with these wallets, we do this 2-of-3 model. It doesn't matter what blockchain you're on; we always adhere to the 2-of-3 model. Protects against both theft and loss. You hear a lot about MPC; we slot MPC on some of the non-Bitcoin chains. I think we're also the only one that uses multisig, which is squarely—ask your cryptographer friends—it's definitely more secure than using MPC. Using MPC on Bitcoin rather than multisig is terrible; you shouldn't do it. We pull out all the stops to do the most secure solution on each chain, whatever that chain provides. Not all chains have multisig support, so we use MPC.
Back to your question: we need self-custody. And the reason we need it is because if we don't have it, we'll just end up with a centralized banking system all over again. This is what allows all of us as individuals to fight back against the system if the system ends up not working for us. It's happened before; it'll probably happen again. You need that power. You need that ability to say, "Nope, that's not for me. I'm going to be able to hold it on my own, and you can't take it from me. It's mine."
If we lose that hill, then I'm not sure what we're doing, because then I guess we created a new database for the banks to use and the banks are in charge, and if they decide they don't like what you're doing, they can freeze your account and you're done.
So we critically need it. That's number one. Now, is it for everyone? Look, I think as it goes more and more mainstream—we talked about earlier the administration helping it go mainstream by allowing more and more people to use it, institutions to help using it. Charles Schwab says they're going to launch Bitcoin purchases for their clients by the end of next year, sometime relatively soon. So that means you'll have custody at Schwab if you want it. Maybe you'll be able to do custody at Citibank. You can do custody at BitGo right now.
As people have this as part of their mainstream asset portfolio, they're going to have some which they might use in a self-custody capacity, and some which they might say, "You know what, this is a large enough pile, I need help protecting it." Definitely I've seen over the years—I still get them every couple of months—somebody that started out holding their own custody 10 years ago, and they're perfectly happy with it, but their new predicament is, "How do I make sure that my spouse gets it or my kids get it should something happen to me?"
And by the way, there's other things which are kind of hard, like how do you protect against loss and fire? You can have your cold key, you can have your backup—did you take the time to put your backup outside of your house? And if so, where did you put it outside your house, and who has access to it? Maybe put it in a safe deposit box. People say they do these things, but we all know how people are with security. People are not as good at security.
The other thing which is hard about digital asset security: you might buy it and it's worth $1,000, and then you wait 10 years and it's worth a million dollars. The amount of security you would apply for something that costs a million dollars, of course, is higher than the security you would apply for $1,000. So when you bought that Bitcoin, you're like, "Ah, this is good enough for me, it's only $1,000." On the other hand, a few years later, it could be worth something where it's actually material to you, your family, etc. So this makes self-custody hard. I think we're going to see a balance, but BitGo will always remain completely available for self-custody as well because we absolutely need it.
And don't you custody the Ledn Bitcoin? Ledn's our title sponsor.
We do a lot of work with Ledn, yes. They've been a happy client for a long time. I don't want to speak for them, but yes, I think they're quite happy.
Well, before we get into more, you have an announcement that you want to make, don't you? There's going to be an opportunity for people who watch the show and listen to it to maybe start custodying with BitGo if it doesn't all have to be in self-custody. But if they wanted to move to an institution, what are you offering folks?
That's right. Well, by the way, we've been focused on institutions from the beginning. It's not that we started out that way. Actually, the very first variant of BitGo was retail-ish and said you have X amount of Bitcoin secured, things like that. But who gravitated towards that product was businesses that really needed this no-single-point-of-failure. You can't have one guy on the IT team being able to take all your Bitcoin. Also high-net-worth individuals went there, people that had a lot of Bitcoin. And so we quickly ended up starting to cater towards institutions, and of course that's what we've done for the last 12 years.
Through all of that, we strongly believe retail deserves what institutions demand. And it's appalling to me, it's appalling to so many people. By the way, I think our industry has failed on some of our promise. I think part of why Bitcoin has held back a little bit: we have not delivered on the early promise, which is that we're going to eliminate middlemen, we're going to reduce fees, we're going to bank the unbanked. Instead, we've got retail exchanges charging so much it makes Goldman Sachs jealous. I'm not kidding. They were lead bankers on the IPO. They are absolutely jealous at the insane rates that US retail exchanges charge. You can go look it up: go look up the public retail exchanges and you'll see 160 basis points in fees to buy Bitcoin on average.
That's crazy. Yeah, that is crazy.
Institutions don't pay that. Institutions are like 10, 20 basis points.
So what BitGo's been doing in order to satisfy institutions—we built it at scale, it's easy to use. Why do we gate it? We're not going after retail per se as our starting point, but we built a better product. So whether you're talking about trading fees, whether you're talking about staking fees, whether you're talking about the security of it, the ability to do self-custody in addition to custody—exchanges have really developed to just facilitate more trading. That's what they want to do: just drive more trading at 160 basis points. Great. We're not going to do that. So anyway, we're taking our whole product and making sure that it's available to everyone. So you just sign up, use it, it works, and you get the same features that institutions get.
No minimum, nothing, no fees.
Everyone can get their account. Okay, we'll put a link up and we'll put a link in the show notes. I think that that's going to be very interesting. Congratulations on—
Well, give us feedback. We always want to make a better product. There's a lot of capability, and I think one of the hard parts of digital assets is there are so many different aspects that have to be built. Remember, the traditional finance system has been built over decades, right? And so it's kind of segmented into different products and different places, and you're kind of used to it. With crypto, staking is a brand new thing, trading is done differently, it runs 24/7. So anyway, there's a lot to it, but try it. We think it's pretty easy to use and hopefully you like it.
So, Bitcoin, I feel like, has broken up into several different factions, and this bear market with its low sentiment—I don't know, it's like the nerve endings are raw with a lot of the groups. And I think there is a group of people who are so scared about the institutions coming in, and they see companies like BitGo and the Bitcoin treasury companies as sort of the enemy within the gates. They're worried that Bitcoin will somehow end up controlled, end up like fiat paper versions of Bitcoin, kind of destroying the value proposition or manipulating the price. What do you say to them, and how do you address those concerns?
Going all the way back to 2012 and 2013, the early Bitcoiners—you heard more than one discussion about how it's a bit of an experiment. To some degree, we haven't graduated out of all of the experimental phases, but the goal here is to make it something that's trustable because you don't have to trust anyone, right? The whole reason you can trust it is because there is no trust. And that's a hard thing to do, and we talk about centralization a lot.
I don't think we've solved all the problems. I think that's true. I think that the big ETFs do become a bit of a threat to some degree to Bitcoin potentially. I think anyone that's accumulating large amounts of Bitcoin in a single treasury also does. I've asked this to some of the folks that run digital asset treasuries; I've said, "How much is too much Bitcoin?" And they're like, "No, it's never too much Bitcoin." Okay, no, no, no, but really, how much would be too much Bitcoin? Like if you had 95% of the Bitcoin—let's say Natalie's Digital Asset Treasury, you have 95% of all Bitcoins—do you think that'd be a good idea?
It's kind of like when Michael Saylor was asked on CNBC, "Are you trying to get all the Bitcoin?" And it's like, well, there would be no point if one person had all the Bitcoin.
That's correct, that's correct. Okay, so we agree that 95% is too much for Natalie's treasury. Should Natalie's treasury have 50% of the Bitcoin? Probably still too much.
20? Probably impossible at this point to even get to that, right?
Okay, well, look, I don't know. Look, we do have individuals, right? And I like Michael Saylor, so I'm not trying to discredit him, really just talking about the theoretical exercise of how much is too much as a single entity. He's approaching 5%. I think it does make people say, "Wait a minute, is this healthy for me to invest in this asset if one party can affect the market so disproportionately?"
I don't know where the stopping point should be, but I do think that these questions are coming up even at the 4% to 5% level. He's at, I think, 4% right now.
Yeah. So I think it becomes an issue, and you can sort of extend that to an ETF. But actually, I do think there's some amount of comfort you can take in US ETFs: those are piles of large numbers of people all commingled together in a single product, so it's not really a single entity. But because it's a single entity holding custody, they can move quickly or they could have a failure quickly.
Well, and one entity is controlling the keys, right?
Yes. Well, they could use multiple custodians; there's a lot of things that they could do, right? So there's ways to address those problems. But look, I think to some degree it does become a concern. However, we need to build Bitcoin so that it's hardened against that, because no laws you can pass in the US are going to fix this problem. Somebody somewhere else on the planet might do these things, and the way you trust Bitcoin is the fact you don't have to trust anyone.
So, you should always be thinking a little bit. Yes, there's still some experimental components of Bitcoin that we are still figuring out. Quantum hasn't been fully figured out yet. We've got to go do some changes, right? It's hard to change the consensus protocols. What are we going to do?
Are you worried about that at all?
Quantum? No. But look, we're addressing it, actually. We did a super simple thing at BitGo. This gets really technical, but basically the different parts of Bitcoin are vulnerable to quantum—specifically elliptic curve cryptography if the public key is exposed could potentially be hacked by a quantum computer.
So if your public address is out there, that potentially is vulnerable to quantum if we get a quantum computer, right?
Correct, but the hashing algorithm that creates the addresses is not quantum-vulnerable. So what that means is if your unspents have never exposed the same public key, you're okay.
In wallets, this is low-level stuff that you really don't want to have to think about, but we spend time thinking about, "Okay, how do we spend the different parts of your wallet?" You might have received five bucks here, and then $100 there, and then $20 there, and these are these unspents, and we manage these in a way to keep your fees low. That's typically what wallets have been doing over the last 10 years. There are some exceptions for weird wallets, like if you're receiving lots of quarters at a video arcade, you might have large numbers of small deposits, things like that.
With quantum, what we can do is we can change the way we do the unspents so that we optimize for not having any public keys revealed. In the old model, we were minimizing fees. Now we can say we want to minimize—or take to zero, actually—the number of public keys that are exposed. So if you had received two deposits at a single address, instead of only spending one even if we don't need to, we'll spend a little bit more fee and we'll spend them both so you get a new address, and that means the public key of the new address is not exposed.
We've done this today. We launched it. It's got a couple features to it: number one, you can just look at your wallet, it'll tell you if you've got any quantum exposure, and it gives you a click and you can fix it.
And number two, it's got an unspent algorithm where it's not going to exacerbate the problem. Every time you go to spend, it's going to make sure that you stay quantum clean. If everybody did this—and by the way, I think maybe every wallet should do this—remember, with quantum, there's going to be kind of three phases that have to happen:
Number one, how do we change the protocol to have a new quantum-resistant signature? We're going to need that. Still needs to happen, right? It's going to take some time, people are working on it, we're getting there.
Once that's available, all the wallet providers, all of the exchanges, all of the self-custody people have to move to a wallet that supports it.
And then number three, everybody's got to move their assets. I guess there's a fourth phase, which is dealing with Satoshi. Maybe we'll leave that as a last exercise.
So if you think it's going to take two years to get the protocol upgraded, how long is it going to take for everybody to move to the new wallet? Some will move fast; BitGo will be there on day one. I promise you, we were there on day one with Taproot and Schnorr. We do these things, but not everybody will. It takes a long time to upgrade those. And then number three, how long is it going to take for everybody to move? A really long time. So it's that movement risk, that operational risk to quantum, which is probably the biggest risk. The actual getting the signature ready is going to be faster.
Now, if we had a real quantum threat, that would give people some motivation, so they might move a little faster than normal upgrade scenarios, but still this is an issue. So we could do this unspent trick today. We do it at BitGo and it gives you some freedom and latitude: even if the quantum computer shows up tomorrow, your coins are okay.
Coin Stories is powered by BitGo. BitGo offers individuals free, insured, institutional-grade custody. Keep full control of your Bitcoin secured by a publicly traded, federally regulated digital asset bank. Sign up for free at bitgo.com/natalie.
Owning your Bitcoin should feel secure even when life doesn't go as planned. Phones can get lost, devices break, backups go missing, right? Bitkey is built for exactly those moments: no seed phrase, a screen that shows you exactly what you're approving, and recovery and inheritance built right in. No subscription needed. Download Bitkey today and use promo code "STORIES" for 10% off. This episode is sponsored by Bitkey.
Freedom isn't just about protecting your Bitcoin. You have to protect your data and privacy, too. My friends at The Bitcoin Way can help. They offer expert training on straightforward cyber security and privacy tools that actually work for regular people. To learn more, book a
call with them at thebitcoinway.com/natalie. I remember my first news job. I was making like 20-some thousand. I didn't start putting away in my 401(k) when I should have. What's your argument for considering even having an IRA if you're a Bitcoiner?
Taxes. You're going to sell off some Bitcoin and you're going to move it into dirty dollars or whatever currency survives the dollar, and there'll be a taxable event for that liquidation. If you do it in a Roth, you could turn $5,000 into a half million, and you're going to take it all out entirely tax-free.
I think that this does raise some concerns given what happened with Coldcard, that you could do everything right. I mean, because the people who lost the money, they literally, I mean, some of them rolled dice and still lost funds, and the idea—
I thought if they used the dice, they were okay, they did it wrong.
I think some, if they didn't have enough dice rolls, were also compromised. But I think a lot of people are worried that they could do everything right and maybe have multisig, but if those wallets have been publicly, you know, revealed somewhere, that they could be sitting at home one day and the quantum threat suddenly becomes reality and there's not enough time, like all of a sudden your coins are just gone. So, how—I mean, when you think about all the millions of people who own Bitcoin now, we're going to have people who lose more coins, don't you think, in this upgrade?
I don't think it has to be that way. No. And I guess to some degree, back to the are the banks untrustworthy? Like, do you need to have self-custody? Is like giving up custody the same thing as like, you know, giving up all your privileges? The developer of the software you pick, even if it's self-custody, is being trusted—
unless you're reviewing all of their code,
So, that is a trust point that, you know, like we talked about: how do we make something that you can trust because you don't have to trust anything? Well, unless you're coding it yourself,
you're trusting whoever wrote that code. Yeah.
So, look, there's things we could do there. We can do security audits. Obviously, we've been doing them for over 10 years at BitGo over and over and over again. There's SOC audits, there's other things we can do, but to some degree, yeah, implementation risk will always be there. Now, when you use an implementation that's multisig and you can see these keys are provisioned separately, that's an architecture which lends itself to a safer result than not using that architecture. So multisig is a better architecture that should help with this problem. But yeah, you're not going to get away from 100% if you are trusting three implementations of keys.
