Loading summary
Transcript243 lines
- [00:00]
A
It makes sense to prepare for the worst case scenario.
- [00:03]
B
Yeah.
- [00:03]
A
And by the way, quantum computing developments are funded in large part by governments who have in mind espionage type use cases. So there's really no incentive for them to showcase the latest and greatest.
- [00:13]
B
Welcome to Galaxy Brains.
- [00:15]
C
An infinite amount of cash.
- [00:16]
B
Cash. I'm your host, Alex Thornton. The US banking system is sound and resilient. Bitcoin meeting new, all time high.
- [00:24]
A
If you're not long. If you're not long, you're short.
- [00:27]
B
Satoshi is going to come on there,
- [00:28]
A
laugh hysterically, go quiet.
- [00:30]
B
All bitcoin's going to be erased. Bitcoin. Bitcoin's the best crypto.
- [00:36]
A
Bitcoin is going to zero.
- [00:39]
B
Welcome back to Galaxy Brains. As always, I'm your host, Alex Thorne, head of Firm Wide Research at Galaxy Bitcoin. Not Zero. We have a great episode for you this week. Alex Pruden, co founder and CEO of Project 11, joins us to talk about the threat that quantum computing might pose to public blockchains like bitcoin. Alex is a very public expert in quantum and the overlap with public blockchains, and he's been part of a growing discussion in the bitcoin and cryptocurrency ecosystem about what to do to mitigate the threat of a cryptographically relevant quantum computer. Some say there is no threat, others say the threat's very far away. Alex says, you know, if there's any risk at all worth working on it now rather than risk doing nothing. We'll also talk with our good friend Bimnet, a BB from Galaxy Trading, as always, about markets, about the threat of, of private credit, you know, decaying, about inflation, commodity inflation, what the Fed can and can't do and won't do, will do. And some updates about what we think might happen as it relates to a ceasefire between the us, Iran and Israel in the Strait of Hormuz. Before we get to all of that, I need to remind you to please refer to the link to the disclaimer in the podcast notes. And note that none of the information in this podcast constitutes investment advice or an offer recommendation or solicitation by Galaxy or any of its affiliates to buy or sell any, any securities. It was a good episode there with Alex that we just recorded. Really a complicated issue, but I think he breaks it down pretty well. And I'm actually moderating a panel on the main stage of the Bitcoin conference coming up in about two weeks in Las Vegas about Quantum, on which Alex Pruden from Project 11, our guest today, will join us along with James O' Byrne and Reardon Code 3 really prominent quantum researchers. Quantum or commentators at least. And also of course, our team, Will Owens on Galaxy Research published a great report documenting the work that bitcoin developers are already doing to mitigate and research the quantum threat. So very interesting and timely topic after those Google and oratomic papers were released two and a half weeks ago. Let's hop right into it with Bimnet. Absolutely. Let's go now to our friend Bimnet Abibi from Galaxy Trading. As always, Bimnet, welcome to Galaxy Brains.
- [02:57]
C
Thank you for having me.
- [02:59]
B
Bitcoin at the highest level in what, four or five weeks? I guess just below 75k as we record on Tuesday, this week, April 14th. A little bit early. I'll be traveling tomorrow. So we want to get this on the books. Equities higher as well, markets higher. What's driving it?
- [03:20]
C
Yeah, I think it's a renewed sense of optimism around a ceasefire agreement being more permanent. I think what the market is latching onto is progress in terms of the Iranian regime's willingness to potentially stop refining uranium for potentially five years, maybe even longer.
- [03:45]
B
US had asked for 20 years, reportedly, apparently they'd countered with five.
- [03:51]
C
Well, the issue was there weren't proper communication channels between the representatives of Iran that were in Islamabad and the decision makers that were in Tehran and in Iran. And so they couldn't actually get to an agreement because of some of those communication challenges. But right now it seems like we're on pace for a second round of in person talks and there is traction on the nuclear front. And according to Trump, the rest of the stuff went reasonably well. And so I think we're in a really good position to have a favorable outcome in terms of a ceasefire. Now the question is, has the market gone a little too far in terms of baking that situation in? I think it might have, but I still think that there's more to price in.
- [04:52]
B
If we really get it, it would be even more positive.
- [04:55]
C
Correct. What you're seeing happen in the market is you formed a base when the optimism really started to build in earnest. Then you're kind of in the late stages of this rally where a lot of junk and lower quality assets are rallying now. And that's the beaten down SAS names, that's the quantum computing names, that's like nuclear stocks, memes, bitcoin sensitive equities, stuff like that is really performing now. And I think that's because there was a healthy short base. And it's also like if you're short,
- [05:34]
B
meaning there were short squeezes happening and then the longs or the momentum traders, oil's already come down or bitcoin's already rallied. So now they're looking down the spectrum of what hasn't rallied yet.
- [05:46]
C
Yeah, I mean there's also a little bit of a fundamental component where the SaaS names have been beaten down because of things like Mythos and that questions kind of the terminal value of a lot of these SaaS companies. But in terms of near term catalysts, it's earnings and these companies still haven't been negatively impacted from an earnings standpoint
- [06:10]
B
from AI and from this situation.
- [06:12]
C
Right. And so you can't get rid of your two year contract right now because Claude is like, it's amazing for the
- [06:20]
B
last month and a half or two months.
- [06:21]
C
And it's also.
- [06:22]
A
Yeah.
- [06:23]
C
And so there's a bit of covering in some of the beaten up like SAS names. Yeah, Mag7 is doing real well again today. I think what's happening is you're getting some short covering, some fundamental longs being put on as well. Then you also just have some technical things where tech companies, for example, report in a couple weeks, stock buybacks, resume. CTAs are generally better buyers. Now that you're above key technical levels, you've got CTA's stock buybacks. And then April 15th is the tax day. So today is the last day to sell stocks in order to make your tax payment on April 15.
- [07:08]
B
You mean to generate the cash to make the payment because it's T plus
- [07:11]
C
one settle and it's April 14th. So 15th you get the cash, you can send it to the Treasury. In theory, you've got some structural elements that are really supportive of the rally. And also you've had a decent sell off in the dollar, which is again risk positive, factually speaking, because a lot of the earnings of these companies are actually from abroad. There's just a lot of good things happening right now from a flow of funds perspective. The market is probably still going to go higher as long as this good backdrop is there with the Iranian ceasefire. Now, in terms of what the medium term picture is, I'm about as concerned as I've ever been. I think tactically it's higher before, lower, just to backtrack a little bit. I think Bitcoin is going to trade well in that environment. I think 76k is a good area to pause. And that's what we've seen today, where we've just basically sold off from that level. You've got the 100 day moving average. That's around here. It was the high from a couple weeks ago. I think the bid in the market is largely coming from dats. This STRC paper has been very hot in the market. Today's the last day to buy it. If you want to collect the dividend, there's a lot of buy flow there. And the MSTR spread has been expanding with some short covering in MSTR and some longs being added. And so bitcoin's had the sweet spot where it's gotten back into this positive reflexive loop where spot goes higher. That allows the DATs to issue more paper which allows them to buy more bitcoin. Bitcoin goes higher, you get some positives in the underlying stocks. You're in this positive flywheel situation right now and I think there's a reasonable chance that you go back up to like 80, 84K. It's reasonable. Like, I don't think it's out of the question. Would I necessarily be gung ho on playing for that? You know, I think there's some smart ways to play for that without like risking too much call spreads, you know, think things like that. But overall I'm cautiously optimistic. This market, there's the chance that the ceasefire is a little bit of a delayed buy. The rumor sell the news type of event where even on a deal you should definitely rally. But the upside from that rally, how
- [09:56]
B
much is already baked in.
- [09:58]
C
Correct. I think your risk reward starting to get a little bit more unfavorable, especially after a deal gets announced. I think stocks may have a couple percent in them left. Call it anywhere from 2% to 5%, which takes you to all time highs. But in terms of what I'm seeing structurally.
- [10:21]
B
Talk about. Yeah, sorry you were going to say before you. That was good near term and tactical explanation and caveat. But before that you said medium term. You haven't been this concerned ever or in a long time.
- [10:33]
C
Correct.
- [10:33]
B
Yeah. Tell us about that.
- [10:36]
C
AI is probably at the forefront of my mind. The Mythos advancements were so concerning that Secretary Besant and Chairman Powell literally had to sit down the largest banks in the US and tell them about the risks. You also had the same thing happen in Canada. These are the models that were trained on the latest batch of chips. I actually don't know if it was Blackwell's or the Amazon ones. But long story short, I think the advancements of this Mythos model suggests that the trend, or, sorry, the phenomenon of throwing more chips, more power and more resources at AI and getting smarter. AI that is still present.
- [11:21]
B
Yeah. We are not. There's not evidence of plateauing yet.
- [11:25]
C
Correct.
- [11:25]
B
I still think they will eventually plateau,
- [11:28]
C
but right now we don't have the evidence does not suggest it. In fact, the AI people are getting scared about the capabilities from a cybersecurity standpoint.
- [11:38]
B
No. And to be clear, my theory that frontier models will plateau, it's not empirical. There's no evidence that it's happening yet. My idea is solely based on the fact that AI will run out of data to train on, that AI hasn't itself created and that's sort of garbage in, garbage out, and that there will be compute bottlenecks that also become, at least in the near term, prohibitive for training. But again, yeah, I mean, every of these big. For all we know, you know, we haven't what, gotten a big new model from OpenAI yet since ChatGPT, what, 5. And you know, for all we know, they'll release ChatGPT 6 and the, you know, within months, most likely. And it too will show a step function increase in quality. So we're still in the major increase in quality and efficiency.
- [12:24]
C
And I, and I think it's just like really hard for like the everyday person to quantify or fathom the magnitude of these improvements. And so imagine the way I would describe it is, let's say you're in a really hard college class. The difference between an A minus and an a isn't the 3%. That is mathematically the difference. It's really a big change in logic and quality of thinking. And so you're at this point where the AI is just getting so insanely smart and capable. And we know that there aren't that many models trained on the best chips yet.
- [13:10]
B
Right. What happens when more of them are. And so this leads you to what, like job loss?
- [13:18]
C
Job loss, absolutely.
- [13:19]
B
Yeah.
- [13:19]
C
What are the. How do you put a terminal value on a software company?
- [13:23]
B
Yeah, how. When it can be its products can be rebuilt or they can be exploited with vulnerabilities? Yeah, there's. So there's. Would you say there's like Mythos in particular and other models that may come out that are like it, but this incident with Mythos and Project Glasswing, which is them giving the preview to those 12 companies and the threat from its cybersecurity and pen testing capabilities, would you say that, like, is the right way to think about it, that it increases the potential for like, Black swan cybersecurity events? So now it's hard to put a value.
- [13:58]
C
I Think in terms of a probability of a black swan cybersecurity event. You've never seen it higher. In fact, the CCP got hacked like a month ago where there was a vulnerability exploited for like a month before they even noticed terabytes of data, I assume. Yeah, Dubai got hacked recently as well. And so there's some questionable things happening. And yeah, Bessant and Powell met with all the bank CEOs.
- [14:23]
B
Yeah, that's rare.
- [14:24]
C
That is a big deal. And I don't think the market's fully appreciating that. But taking a step back, say you can't place a comfortable terminal value on a lot of software names. So that's one part of the equation. The other part of the equation has been kind of like the private credit story, where the software companies represent a large portion of private credit balance sheets. The logic here is if you're marking down some of the software debt from par to $0.90 or lower, what does that mean for the equity values? In theory, the debt is senior to the equity in the. These software companies. Oh my God, maybe the software companies aren't worth that much as much now. Who owns the equity? Private equity. What does that mean for private equity valuations? Where are they going to get mark them on June 30, et cetera? That's a huge part of the investment world. If you look at any pension endowment, insurance company, you're talking about 10 to 20% allocations in this very opaque private equity world. Now, let's say that stage is set. What do you have happening in fixed income? Fixed income is beholden to inflation and central bank response functions. My central bank response function, in terms of what I think their response function is, is that historically they've always been kind of motivated by the last mistake that they made. They're not able to dynamically think through how to approach a new set of conditions.
- [16:15]
B
You can't reset their context window by starting a new session.
- [16:19]
C
Exactly.
- [16:20]
B
Using AI terms now.
- [16:21]
C
Exactly. In Europe, in England, in Australia, New Zealand, et cetera, they're all expected to hike interest rates as a function of where commodity prices have been and subsequently the inflation expectations are and where they're likely to realize.
- [16:36]
B
You're saying they're short term, they're never as long term as they should be. They're too reactive to the current and the very recent. Rather than trying to take a deep, thoughtful, methodical.
- [16:48]
C
The classic example is Trichet that hiked in 2008 to a supply shock and when the world was about to implode. Classic Policy error type of stuff. I think that's where we're headed. Because in my head, inflation is now driven by the commodities and the supply chain. If you list through all of the implications of the Strait of Hormuz, everything from helium prices, sulfuric acid, polyethylene, fertilizer, all these things, these petrochemicals, they've increased in price a ton of. And you will see that feed into inflation eventually.
- [17:30]
B
Right. Food, plastics, all consumer goods, let alone energy.
- [17:34]
C
Correct.
- [17:34]
B
Chips. Helium is in chips.
- [17:36]
C
And what I'm convinced of is even with an open Strait of Hormuz, every molecule that is transferring out of the strain of Hormuz will have a material bid to it. Right.
- [17:46]
B
I saw some people have been talking about all these empty tankers around the world are like booking it across the world to go into the Gulf of Mexico, Gulf of America, excuse me, to buy our supply. And that a huge portion now of LNG servicing Europe, partially because of Hormuz, but also because of the pipeline having blown up between Russia and Germany, but also in general, US LNG has one of the largest market shares in Europe. It's had in decades because. So. But there's a finite amount of ships. There's some that are empty. Sure, maybe they can buy American petrochemicals, but the ones in the Gulf are full and waiting to get out. And it's not like, oh, yay, a bunch more supply. This is currently like, supply that's off the market. It's not new supply. And you're saying they're bid. So it's not like all of a sudden prices come down that'll get gobbled up right away. As soon as those things stream out of there 100%, it eases the, the tension in the market a little bit. But it's not like a fresh wave of new supply. We're basically like.
- [18:51]
C
And you're still at deficit.
- [18:52]
B
Yeah, we're like a month and a half behind.
- [18:54]
C
Correct.
- [18:55]
B
In oil supply.
- [18:56]
C
Yeah.
- [18:56]
B
So it's like, let it out. Great. It's no longer being blocked. You can maybe start the clock on slowly getting back to par. But it's not like there is more ships that are empty in there waiting to be loaded with Saudi crude. There's no. The ships are full there.
- [19:10]
C
Correct.
- [19:11]
B
Yeah.
- [19:11]
C
So, yeah, it's going to be a very tricky commodity story to navigate, pardon the pun.
- [19:17]
B
And by the way, you've made this point many times to navigate. That's good. And so has the Fed. I think Jay Powell has made this point too. And maybe just recently Maybe it wasn't Powell, but commodity based price inflation is not something the Fed can really materially affect with their blunt levers of monetary policy.
- [19:36]
C
Correct.
- [19:36]
B
So hiking cutting probably doesn't make much of a difference.
- [19:40]
C
But at the same time, what I know for sure is that fixed income markets are not going to start aggressively pricing in cuts in this type of inflationary environment. What does that lead to? That leads to tight financial conditions where all of the corporate debt that's got to get financed this year, all of the sovereign debt that has to get. And these sovereigns, they're going to have to issue more because their defense spending is increasing. Right. And the outright level of rates that they have to pay to finance that defense spending is higher. Right.
- [20:16]
B
And so less money for the economy, for growth, for new jobs, for whatever, all the above. Right.
- [20:22]
C
And they're also, you know, some places they're going to try to subsidize the commodity prices, etc.
- [20:27]
B
Brother, I don't know how many times we've sat here and I've marveled at the complexity of the situation. A long, for the days I wish
- [20:33]
C
it was just remember like, you know,
- [20:35]
B
like, I don't know, 2016, you're just like, well, you know, software is eating the world, it's long everything.
- [20:41]
C
Yeah.
- [20:41]
B
You know, every dip gets bought. Not that easy right now, not quite at all, is it?
- [20:47]
C
And so you've got this scenario where like high yields globally are going to keep a lid on risk prices and then you've got this structural trend of AI that should like consume jobs and you've got heightened risk of cybersecurity and then await, yeah, you might get a permanent ceasefire deal, but in my lifetime, Middle east has always had flare ups, etc. I don't know if this is truly the end of the geopolitical warfare that's going on.
- [21:22]
B
Tupac was calling for peace in the Middle east in 1998. So I agree, even if this resolves, it's a tenuous situation and is, I mean, has been my whole lifetime as well. So let's stop here. It was a great conversation today. Bim Bim Nibi from Galaxy Trading.
- [21:40]
C
Thanks for having me.
- [21:43]
B
Let's go now to our guest, Alex Pruden, Co founder and CEO of Project 11. Alex, welcome to Galaxy Brains.
- [21:48]
A
It's amazing to be here. Thanks for having me.
- [21:49]
B
I'm excited to have you here. You Quantum. Project 11 works on quantum computing. Specifically what? Mitigating the threat of a cryptographically relevant quantum computer to public blockchains.
- [22:02]
A
Yeah. So Future proofing, future building is kind of the tagline. So I guess psa. Quantum computers are coming and they're good for many things, but the thing that we know that they are good at is breaking all the underlying cryptographic assumptions that protocols like Bitcoin use. So there's a necessity to basically future proof digital assets like Bitcoin and others, which, by the way, like most blockchains, rely on elliptic curve cryptography. So. So you got to future proof them. But then I guess, you know, in the farther future, we see an opportunity to leverage quantum mechanics and these quantum computing platforms to basically provide, you know, new ways or give new ways to secure and send value in the future.
- [22:42]
B
Better blockchains, Is that what you're saying?
- [22:43]
A
Better blockchains, better wallets, better blockchains, better key generation, A lot of things.
- [22:47]
B
Well, we're going to get into a bunch of topics related to this and. But maybe just to set a high level, I don't think Phineas, have we talked about. I don't think I've quite done an episode on Quantum yet. Yeah, we've talked about it a little bit. So maybe just set the stage and explain, like, the basics of why a quantum computer or a sufficiently good quantum computer could threaten the cryptography that, by the way, the whole world uses, but also basically all the blockchains or Bitcoin specifically.
- [23:17]
A
Yeah. So the super simple way to think about it is the quantum computer is a very special kind of computer that's not as good as a classical computer for many things, but it's far better than a classical computer for some things. And one of those things is basically the problem of factoring numbers, or in the cryptographic context, what's called the discrete log problem. You can kind of think about it as, you know, when you're. When you multiply two numbers, it's kind of easy to easy, algorithmically to go that way, but dividing and trying to get the factors of the product is kind of harder. So. So cryptography leverages this asymmetry in certain problem types to basically build cryptographic primitives like public key cryptography, where you can compute a public key quite easily from a private key. But it's really, really, really hard, at least for a classical computer to go the other way down the one way road. But it just so happens that we know of a quantum algorithm to go the wrong way down that one way road.
- [24:12]
B
This was invented decades ago. One of them, at least, was the most prominent one by Peter Shor at mit. Right. Or I guess he's At MIT now, I believe.
- [24:20]
A
Yeah, yeah.
- [24:20]
B
This is Shor's algorithm. This was the mathematics behind how one of these then very theoretical now, much less theoretical computers could. Is it like a brute force attack, basically?
- [24:33]
A
Kind of, yeah. And so quantum was actually coming. Like, the concept of quantum computer was invented by Richard Feynman. It's one of the greatest physicists of all time. But it was all in the context of modeling quantum physics. Right. Because quantum physics is famously kind of hard to understand and model. And so he's like, well, we need to build a quantum computer to, like, model the physical universe. But people were like, ah, this seems abstract. Like, is this thing good for anything at all? And Peter Shor actually showed this. He created this algorithm that said, hey, there's a classical problem that quantum computers can solve. And then he also did one other thing that's important, which is he came up with a theory of error correction, because quantum mechanics, very fragile. And quantum computers that are built on quantum mechanics, therefore very fragile. So Peter Shor said, not only do we have an algorithm that we can basically apply to a classical problem, but we also, it is theoretically possible to make these fragile states less fragile, and therefore we can actually plausibly do this.
- [25:25]
B
But no one has run Shor's algorithm on one of these machines yet. Right.
- [25:30]
A
So there have been. People have run variants of Shor's algorithm to do very small instances of, say, factoring integers, et cetera. You know, I say variance because, you know, quantum computers use. And you asked something earlier about brute force. I mean, quantum computers use, you know, quantum mechanics and properties like entanglement, where, you know, you have two quantum bits can kind of share parts of the same state without being physically connected. And so these. These weird quantum mechanical properties that, frankly, physicists have scratched their heads about for decades are kind of the secret power behind these things. And a way to think about it for, you know, kind of a layman is like, they kind of try every possible path to. To the answer in parallel. I mean, some people like to make commentary about, like, it's a multiverse. I don't know. Whatever your interpretation is, I think the
- [26:14]
B
multiverse almost kind of helps it. It's helped me grok, like, the conceptual nature of it. I say brute force today because, like, if you want to derive, let's just say you got an encrypted zip file and it's only four characters. The password. Like, I used to do this. I led forensics at Fidelity Investments, and, like, sometimes there's evidence that a person had encrypted and, you know, know I'm not going to devote like that much time or compute to it, but I'll run like a six or seven character brute force which could take like four or five hours. If it's, you know, four numbers, we're going to break that in 30 seconds or less with, with a normal computer. If it's, you know, 25 characters that are alphanumeric with symbols, it's so long with the classical computer that it's impractical to do.
- [26:55]
A
Exactly.
- [26:55]
B
And that's. You're trying, you're. But you're. The brute force is you're mechanically trying every possibility. My understanding, again, very conceptually of the way the quantum would attack this is, let's say it was a maze and rather than like a mouse will try every single possibility until it eventually finds the cheese. The quantum computer kind of tries it all at once and tries sort of every possibility at once. But then once it finds the answer just collapses down to the only one that it worked in. It is almost like the others never were tried. So it is like a super, like brute force, but it's not like linearly going through. It's sort of all at once.
- [27:31]
A
Yeah, no, the way you describe it is perfect, I think, for, you know, it's obviously more complicated, you know, with the physics, but like at a high level, conceptual. I think that's exactly right. I mean, it's like, you know, you're trying, you know, the password example, you're breaking. You're trying every possibility in serial. Right. It's like one after the other after the other, versus this lets you kind of brute force it in parallel and then you ultimately kind of converge on the right answer at the end. And that requires this concept of entanglement. Right. This is why classical computers can't do it, because you can't entangle two classical bits. Each classical bit is independent from a
- [28:01]
B
1 or a 0 at all times.
- [28:02]
A
Yeah, yeah, exactly. So that's kind of the magic around it. And again, like I said, there's. Quantum computers can be used for other things besides cryptography, but this is the problem that a lot of people care about.
- [28:11]
B
Yeah. Honestly, like, quantum might be like one of the only one of the industries that's worse at telling the positive story or as bad as AI, which also is really not, on balance, very good at telling the positive that it will bring just some. That I know about that just before we dive deeper into the, you know, potentially scary threat that a cryptographically relevant quantum computer poses could map like whole genomes, like really quickly create new drugs that are customized for each individual. Like you think about like an astronomically multiple orders of magnitude more compute. Now like another one that people, somebody mentioned that I thought was really interesting was like logistics modeling. So like, you know, FedEx has to like move like you know, a million trucks and boats and planes around while simultaneously like in the most efficient way to deliver like 50 million packages with certain energy required. When you start to realize that like actually that problem is so unbelievably massive that they actually probably aren't doing that good a job. And they do a pretty good job because I get my packages. But they could maybe be. Be so much more efficient. So complex systems analysis and design.
- [29:16]
A
Yeah, because that's an optimization problem. Right. This is like complex known colloquially as the traveling salesman problem. And so some kind of what are called noisy intermediate scale quantum computers already are used for optimization problems like you just described.
- [29:28]
B
Yeah. And I've heard that some can do sort of even toy still sort of toy level, but physics problems like trying to model a black hole or something
- [29:37]
A
like that, or material science, like discovering new chemical compounds that can be used to make new materials. And that's. So yeah, there's a lot of potential amazing use cases. And I think it's like a good point to just remind everyone before we talk about all the doom stuff.
- [29:49]
B
Right.
- [29:49]
A
Like we as humanity should kind of want this to exist, but we gotta prepare for it.
- [29:53]
B
Yeah. Similar in that sense. A little bit similar to AI. Absolutely enormous opportunity. You know, whether people are protesting data center build out and energy use or you know, fears about job loss and stuff like that. You know, the, you know, the cobbler, the stable owner also feared cars. And like I personally as an optimist have to believe that these technological advancements are beneficial, have always been beneficial. But the quantum one is like you said, magical. I mean again, maybe to like a senior PhD in physics, it's actually mathematics to them. To me it is magical.
- [30:29]
A
I mean to most physicists. Like physicists still have philosophical arguments about this. I mean this thought experiment of Schrodinger's cat.
- [30:36]
B
Yes.
- [30:36]
A
Is the cat alive or dead in the box? Right. And when you open the box, it's alive. Is it only alive because you measured it or was it always alive in the box? Like, like physicists can't even really on the answer to that question. So it is magical in that sense.
- [30:49]
B
Yeah. So. All right, let's, well, let's focus it on, on blockchains for a bit with, with Quantum. So, you know, take Bitcoin as an example. Bitcoin uses ecdsa. It's an elliptic curve cryptography for address signing, signature schemes, for transacting. Right. So one of the things people always ask is like, are we talking about SHA256? Are we talking about mining people going to mine Bitcoin with a quantum computer, going to have a very powerful asic? Maybe, but that's not really what we're talking about. Right. The primary fear is elliptic curve cryptography. So it's the crypto that secures coins and transactions, right?
- [31:23]
A
Yeah, exactly. So the way I like to frame it is what makes digital assets like Bitcoin different from any other financial system. Right. And I basically boil it down to two things. There's decentralization, right. Which is kind of. And that's consensus. Right. Consensus is the mining and that makes sure there's no one single entity that controls the network. Mining is driven all by SHA 256. And technically there is a quantum attack on SHA 256. But I want everyone who's watching or listening to this show to put that out of their minds.
- [31:47]
B
Yeah. Because I was told by a quantum researcher that there is a technically, but it might take, currently it could take a quantum computer the size of the moon is what I was told.
- [31:55]
A
Yeah. And it's because the advantage of the quantum algorithm over the classical alternative is pretty small. So you would need a gigantic quantum computer in practice. And so just forget about that. The issue though is, so the other thing that makes blockchains unique is this concept of self sovereignty, Right? Only you can send your funds. Only I can send my funds. Like this is a bearer, these are bearer assets.
- [32:15]
B
That's cryptographic certainty.
- [32:16]
A
With cryptographic certainty, that cryptographic certainty is guaranteed by these digital signatures. Like if I'm able to produce a digital signature for a given utxo, that is how we all agree that I own these assets. Right. And again, these digital signatures are based around this public key cryptography where you're supposed to in the classical sense only be able to go from private to public and not the other way. But, but if you have a quantum computer, you can go the other way and get the private key. And then potentially, if your public key is exposed on chain, I can sign, even though I didn't originally, it wasn't my utxo, I can sign for you. And then from the network's perspective, the question of ownership goes out the window. Basically. Because if a quantum computer with sufficient scale and speed existed in Some cryptographic sense, they own whatever percent of the public keys on the network.
- [33:03]
B
Yeah. And this is actually a problem too with like plausible deniability, generally speaking. Right. Somebody will say, oh no, I was hacked.
- [33:09]
A
Yeah, exactly.
- [33:10]
B
My coins. I actually, I don't know, tax man, I don't have those coins. I didn't sell them. Actually. The fact that they left my wallet was a hack.
- [33:17]
A
Exactly.
- [33:17]
B
And you really can't prove that you weren't hacked. So what the network would know is that somebody signed with your private key. And similarly here, like, and I get maybe this is a good segue, although there's so many aspects to talk about. You know, if somebody did derive my private key from say, an exposed public key that I had and move my coins, would would that be clear to the, you know, blockchain viewing public, those that analyze the blockchain, would we know that those coins were hacked?
- [33:47]
A
No. I mean, it would look like, well, I guess again, to your point of deniability, it's the exact same point. Right. It's like, do they know that? How do we know that you didn't just send them? How do we know? So how do, how do we know you didn't just lose your key? We, or how do we know that you weren't hacked? Or how do we know that it wasn't a quantum computer? And this is like we're getting to the kind of, the broader topic, but I think this is one of the areas that I think people underestimate about this threat, is that there's a lot of uncertainty with how quickly quantum computers are going to advance, how specifically they'll be used, who will have them. And I think all that uncertainty I think kind of creates or it basically implies that we don't know what might happen, but it makes sense to prepare for the worst case scenario.
- [34:22]
B
Yeah, I think that's one thing that I totally agree. In the case that this becomes possible, it's not like a giant alarm bell is going to spin around and we're all going to know, or the very first time some coin moves, we're going to be like, oh no, quantum is here because it looks like any other signature. Yeah.
- [34:39]
A
And by the way, quantum computing developments are funded in large part by governments who have in mind espionage type use cases. So there's really no incentive for them to showcase the latest and greatest. And so we just may not actually see the state of the art. And that's just one more element of us.
- [34:54]
B
That's a fair point too. There was the recent paper from Google and Stanford and Ethereum foundation developers and researchers. Right. Referred to from a couple weeks ago, generally referred to as the new Google paper. The new Google paper. Just because the Google guys like Craig Gidney were like first in the, in the byline, but that one they released, they proved their mathematics that they. So that and also this oratomic paper for different reasons, both basically proved or posited proved that the complexity of a quantum computer required to break such cryptography was less than previously expected. But the Google paper didn't actually provide their data. They merely proved its accuracy with a zero knowledge proof.
- [35:41]
A
Specifically why exactly what I said and they commented when presenting this proof. And by the way, what did they prove? They proved that they had compiled a circuit. And so you can think of a program, any program can be kind of represented as a circuit and you can more or less think of it as the number of the steps to go along the way to compute the answer. And so what they proved was they had massively reduced, I think by four orders of magnitude from the prior state of the art, the number of steps that their circuit would have taken. But they didn't want to release that. And the comment around it was if we release this compiled circuit, others with maybe less ethical boundaries or constraints might use this to do bad things. And therefore we think it's responsible as scientists to publish this proof to show the world that we actually did this, but to not reveal what we actually did. Which I think is really interesting. There's a parallel here to you go back to the 30s when scientists learned about the power of potentially splitting the atom. And at a certain point there was a degree of first off, the government, it got involved around what people were allowed to say and not say. But even among the scientists themselves, there started to become a degree of self censorship.
- [36:50]
B
Yeah, it makes a lot of sense, especially if we want to see but these many sort of step function increases or decreases, however you want to describe it improvements. To me, one of the things that, that really raised in my mind as an alarm bell was as other researchers or those same researchers further make these improvement in capabilities for quantum. We may hear less about it. The closer they get, the less we may hear because there are significant national security implications and I don't know, safety implications well beyond Bitcoin and crypto.
- [37:27]
A
Yeah. Both from an offensive and a defensive side. Right. Like maybe if you have this weapon as a country, you want to use it for a variety of things. And by the way, potentially there are countries in the world that have a vested interest in destabilizing a network like Bitcoin, either because they have strong capital controls or because they're authoritarian regimes and they want to take the US down a peg. So I don't think again, this is uncertainty around how this will play out and we can't really rule out any possibility.
- [37:52]
B
Yeah. So we're going to get to a little bit more like the debate that's been going on not just in crypto and blockchain land, but in Bitcoin specifically, but also in the investment community. But before I wanna ask you, so just again, back up here for a second. Public private key cryptography. Asymmetric cryptography. The private key is a closely held secret. From the private key you can derive or produce many public keys. Public keys are, this is not, you know, don't get mad at me like someone like John Carvalho always gets mad at me when I say things like this. It's kind of like an email address. You can give it out publicly. Like the MIT is a famous public key key server. I, you can upload a public key and say this is Alex Thorne's key, you can give it away freely. Your Bitcoin address, you can publish on your website and say, send me donations. Because of the difficulty in reverse engineering the private key from the public key.
- [38:42]
A
Yeah, exactly. And I think actually in the original Cypherpunks mailing list, you know, they were using PGP or gpg, I can't remember which, but it's the same idea. They were like they gave away their here's my public key sign messages. So people knew it was them.
- [38:53]
B
Yeah, and you can prove with very low computation that I did accurately sign, but it would take enormous or incomprehensible computation.
- [38:59]
A
That's right. It's trivial to verify, but it's very difficult to sign unless you know, the, the private key, it's called, so this, this, this, this concept in cryptography is called the trapdoor function. So if you, you can go one way if and only if you have the secret, but you can't really go the other way.
- [39:12]
B
So, and that is what Bitcoin, Ethereum, Solana, many other systems including, you know, basic, all the cryptography, securing the vast majority of the Internet, although some, some services like Cloudflare and stuff, have, have offered some post quantum stuff. It all uses that basic concept. So again, if your public key is, I don't know, exposed in some way, viewable to an attacker, we'll just say generally, but you know, in public, then if one of these quantum computers existed, it could be at risk. How many such Bitcoin Bitcoins maybe. Is that the right either? I mean, number of addresses. How many bitcoins are sitting in such exposed public key addresses?
- [39:54]
A
Yeah, yeah. So we maintain a tracker called the risk list on project11.com, and this is constantly updated every day. So you can go see for yourself and then also you can check your address to see if it's ever been exposed. But yeah, the answer is roughly 35% of the overall supply. So call it 6 million total or so is Right. So it's a significant number. It's more than we estimate. 10 to 15% of Bitcoin supply is lost. And that includes Satoshi's early coins where we're mining rewards that were not 1.1
- [40:21]
B
million BTC in my own estimate.
- [40:23]
A
Yeah, so there's, you know, 1. 1 million is Satoshis that are secured under an older address type that is exposed as the public key natively.
- [40:29]
B
Right.
- [40:30]
A
But you know, a lot of them are big exchanges.
- [40:33]
B
Right. And because it's so it's like pay to public key P2PK, which is like the original format which is deprecated. If you create a new wallet today, you're not going to get one of those addresses. I don't think we need to get too much more into that nuance. Other, but other, the other big category is even if you're in a better address format where the public key is not natively exposed in a Bitcoin block, Bitcoin signing schema process requires no matter what format the address is in, when you're sitting on it and holding it, when you transact, you must show your public key. So any address that's been reused is also likely exposed or is exposed.
- [41:14]
A
That's right. And it's actually when you transact, you present the signature, but for you to verify the signature, you have to have the public key. So that's why in the transactions, I mean, it's literally the signature and the public key together in the mempool. And this was the other interesting thing about the Google paper was that they said, hey, based on this architecture that we have, potentially we can, if a quantum computer that could run the circuit that we proved was built, you could actually run this attack in a short enough time to be able to attack
- [41:40]
B
the mempool even before the coin, the transaction is confirmed was finalized.
- [41:45]
A
Right. And you know, there's like, there's a block time and then there's like the heuristic that people use to Consider finalized. Right.
- [41:50]
B
Because it could even be, even if it was an hour or less. Well, let's talk about that in a second. That the attacks. Yeah, but it. So you said exchanges is that. That's likely because they reuse the addresses.
- [42:00]
A
Yeah, because if you think about, you know, and it's really any business that has on chain infrastructure, like just think about if you're a. Like exchange is a good example where it's like, hey, I give you a deposit address and you save the deposit address in your wallet and you send me a bitcoin and you send me a bitcoin again. Right. And so for me to rotate every time. Every time would be a huge.
- [42:17]
B
Even if you, the exchange, the receiving institution, offer that as a service. If you send me use that.
- [42:23]
A
Exactly. Because if you send to the same address again, I have to either abandon those things and still credit you or I just have to tell you I'm sorry.
- [42:31]
B
Right.
- [42:31]
A
And so it's like, but if you're
- [42:32]
B
already holding some of my coins in that address, then by definition I can send to them again anyway. And anyway, people do. But also like exchange Hot wallets and cold wallets is complicated security infrastructure and they don't want to rotate all the time.
- [42:46]
A
Exactly. Because there's a huge security risk in accidentally rotating it if you have this complicated infrastructure or accidentally exposing something as you're rotating it.
- [42:53]
B
Yeah. So luckily, I mean, that's problematic. But I would say luckily for me, those coins aren't lost though. And they're typically used by sophisticated businesses. So if and when a good solution comes, we could probably count on like the big centralized crypto exchanges and custodians to upgrade plausibly.
- [43:10]
A
Yeah, I mean, I think it's. I think, you know. Yes. But you know, we just have to account for the fact that these are complicated system with all their own dependencies and timelines. We have to bake that in over
- [43:22]
B
overall migration timeline and then the other. Okay, so let's just maybe real quick on a short and long range attack, maybe. Can you explain this? Because people hear that term a lot and you were referencing this a minute ago, talking about attacking the mempool.
- [43:35]
A
Yeah. So it's pretty easy. Short range attack is basically an attack like a quantum computer that can run Shor's algorithm and recover a private key from a public key in a short amount of time. Short being defined as basically around a block time. Roughly. And then again. So what this would look like is kind of like if people are familiar with this concept of mev. A Minor extractable value is kind of a concept more from the Ethereum world. But the idea is basically like you can, you can just sign, you, you can recover someone's private key, sign a transaction, you know, based on the public key that someone else was doing a transaction with over here, and then you, you pay a higher fee and then send the amount to yourself. Right.
- [44:11]
B
So even if I'm storing it in the good address, even if you do everything right, but I try to send my bitcoins to a new address and they sit, they see that as a pending transaction and they break it before it even confirms and then they can
- [44:22]
A
front run you potentially.
- [44:23]
B
That's right.
- [44:23]
A
And you know, as long as they, they pay a high enough fee or a higher fee than you, then a miner is incentive incentivized to take that one. And by the way, again, going back to something you said earlier, how does the miner know it's not just you trying to try to reply?
- [44:33]
B
Theme Right? That's an actual feature in Bitcoin.
- [44:36]
A
Exactly right. So that's a short range attack. The long range attacks are where basically the quantum computer can't recover the private key from a public key in a short, less than a block time. So it's more than a block time. So all of those bitcoin that are constantly rotated into fresh UTXOs that aren't, you know, are using addresses or using good public key hygiene are not, are safe. But you know, it's the big wallets like exchanges or the thing that a lot of people fixate on is satoshi coins, are they ever coming back? Who knows. And so that's kind of like arguably the lowest hanging fruit, the easiest.
- [45:09]
B
And you could imagine, like if you're a nation state, let's say hostile or not, setting bitcoin aside as the example the long range attack is going to happen, it would happen first. Now let's say you have a giant pile of, again, maybe not on bitcoin, just generally speaking, you have a giant pile of recovered but encrypted data. Right. They'd say, what do they say, like harvest now, decrypt later.
- [45:28]
C
Right.
- [45:29]
B
So in the past, like this is even in forensics. So like data destruction is very hard. It's basically impossible to prove. So usually what we did when we wanted to destroy data is you just encrypt it like to the maximum percent. Then you also smash the hard drive. But you know, and the idea was that it was the math would protect the data even in the case that the hardware was recovered. So probably Spy agencies and stuff, when they vacuum up a bunch of data from an adversary, if it's fully encrypted and unbreakable, who knows, maybe they just destroyed it probably in the last decades. Plus that they've realized maybe a quantum computer would exist. They probably just been storing it rather than. Sure, we can't get access, but what if one day we could.
- [46:09]
A
Exactly.
- [46:10]
B
And it's those types of things, including something like Satoshi's coins, that the long range attack could cause. You know, fuck it. If it's really big, high quality potential piece of data, who let the quantum computer work on it for a year?
- [46:22]
A
Yeah, exactly.
- [46:22]
B
If it's that valuable. So you prioritize.
- [46:24]
A
Exactly. But I think there's an important nuance here because I think harvest now, decrypt later, which you referenced a second ago, is kind of what the broader cybersecurity world talks about when they think about the risk of a quantum computer. And by the way, Google and Cloudflare and others on the Internet have rolled out post quantum cryptography specifically to address this risk. But you said something important which is prioritizing based on the economic incentive. And the hard thing about an attacker, if you're doing the harvest now, decrypt later is you have a pile of locked boxes.
- [46:50]
B
Yeah. You don't know what's in them.
- [46:51]
A
You know what's in them beforehand. So you're going to try one and maybe it takes you five days to practice. But then like you just, your probability of getting something may not be very high.
- [46:58]
B
And you're not going to have like 20 quantum computers.
- [47:00]
A
Probably not initially.
- [47:01]
B
Right.
- [47:01]
A
So, but, but I, on the other hand, you can just look at a block explorer and you can basically know with certainty exactly what the payoff of that attack will be.
- [47:10]
B
Yeah.
- [47:10]
A
And so I think, I think for that reason that's like another aspect of the exposure that blockchains uniquely face.
- [47:16]
B
Yeah, that's super interesting. That does make it even look even. And if it's not economical, great, then you don't. Was quantum computer on that block of 50 coins? So, so Satoshi's coins, one of the interesting things. Tell me if this is a mitigation or at least a thing that makes a little less scary. They're not all sitting in one address. Right. Whereas on Ethereum everyone's just like for the most part, I mean you can have multiple accounts but like they, they reuse in giant address one address. So you know, for, I would say the average. Well, not even. I don't want to generalize because it's plenty different. But satoshis coins, I was just doing this on my own in my own bitcoin project. Looking at the Patoshi pattern to the extent that they are his. But most people agree he really didn't use them at all. Almost at all. Very, very minimally. So they're actually. He's just got a. You know, I forget how many tens of thousands of unique bitcoin addresses that all just hold the coinbase reward from having mined a block.
- [48:09]
A
Definitely.
- [48:09]
B
So but each one would be equally as hard to break as the others. So you're. This is both, I think positives and negatives. One is that it's not like you do your attack and get all 1.1 million bitcoin. It's many attacks, concurrent or subsequent attacks. But on the other hand, you're also potentially less likely to. To notice. Right. It's sort of like a drip. If you tax Atoshi's coin, you don't even have to comment like that's not that interesting. I'm just making this point, but it's
- [48:38]
A
the only comment I'll make is I think people would notice even one if even one moves.
- [48:42]
B
You're right. I actually just set up real time alerting again on my own node. So you're right, people would notice. But that's why it actually makes like attacking Satoshi's coins. Probably not the main. You probably find like the highest value, lowest famous address you can find theoretically.
- [48:58]
C
Right.
- [48:58]
A
Yeah. If you're trying to economically optimize probably you don't want to be so obvious that you give yourself away, but you want to get a good enough, you know, haul that it was worth it.
- [49:07]
B
Yeah.
- [49:08]
A
I think the one thing about Satoshi's coins that I think is interesting. I mean there are quantum computing companies that we talked to that explicitly have made this part of the roadmap.
- [49:15]
B
I heard this. Greg Maxwell actually wrote about this as well.
- [49:18]
A
And the reason is because the way that some of these folks think about this is they're like, well, you know, the Spanish were transporting gold back from the Caribbean in the 1400s and the ship sank. And you can go, it's legal for me to go recover this as digital salvage. If I just put on a scuba diving, you know, scuba diving gear and that there's a notion that Satoshi's coins are digital salvage. You know, look, I think there's sort of this relies. That relies on some untested legal theories.
- [49:44]
B
Yeah.
- [49:44]
A
But it doesn't mean people may not try because there is a lot of Upside, even if you have to go sequentially.
- [49:50]
B
Okay, so let's talk about pretty well covered a lot. Is there anything else specifically sort of on the background of the problem that we should talk about?
- [49:58]
A
I would just say the one thing I think is really important for people to understand about quantum computers is, you know, quantum computer is an abstract concept. Just like a computer is an abstract concept. It can be realized in any number of ways. And, and so what Google is doing versus what Oratomic another company that you mentioned versus what IonQ is doing are all actually pretty distinct tech trees. And so even though, you know, people may read about or hear about limitations that Google is having or someone is like, it doesn't necessarily mean that progress in the field is blocked as a whole. I think that's really important because you know, there's effectively a race going on.
- [50:28]
B
Yes.
- [50:29]
A
And that is kind of adding to the worry that potentially someone is going to fight figure out how to get there first. It doesn't have. It's not like all bottlenecked on one path.
- [50:37]
B
And in fact I even saw one of the interesting things I think worth noting that another quantum researcher pointed out to me. So Google is using a superconducting computer, whereas aero atomics using a neutral atom computer. I'm not even going to begin to try to explain the difference. I know generally what a superconductor is. I have no godly idea what a neutral atom computer is. But it's a different hardware architecture. Right. Than what Google's doing. But just like a couple weeks before Google put out their paper, which is about software and mathematics, not about hardware, they actually spun up apparently another division in DeepMind that's actually doing the way or atomic does.
- [51:16]
A
Yeah, neutral atom.
- [51:17]
B
Yeah. And some have pointed out that maybe that's actually a recognition or a fear or hedging. Yeah. Google might be saying, you know what, maybe superconducting isn't going to be possible. Maybe we should do it automobile. So still, even on the hardware side of actually building the machine.
- [51:30]
A
Yeah.
- [51:30]
B
A lot of questions about what is the path. But your point seems to be like, listen though, there's a bunch of paths that have been theorized and people are working on.
- [51:37]
A
That's right. And I think that just the only thing that anyone needs to know about this is that the neutral atom, like the kind of the gen 2 that are not Google's approach, the upshot there is that they're much more stable and potentially easier to scale, but they're probably not going to be capable of the Short range attacks in the short term now, maybe over time, but that's kind of the upshot.
- [51:57]
B
It's a lot of nuance.
- [51:59]
C
Yeah.
- [52:00]
B
I want to get to some of the work that Project 11 has been doing, but first let's address some of the debate that's been happening, particularly in bitcoin land. You know, I feel like other cryptocurrency ecosystems, some are saying, have been more proactive in addressing this potential threat, whether that's actually initiating something or deploying some mitigation. But mostly mo. No one's really, not many. I mean, I, I can't tell you how many blockchains claim to be post quantum. I can tell you none of the major ones that I know of claim that. But some of them, like Ethereum and Solana, have done some real work or released roadmaps or at least, you know, formulations of a plan. One of the criticisms that our mutual friend Nick Carter has made, and I think many have made fairly, perhaps fairly, I should say, is that I don't think bitcoin developers or the bitcoin community is taking it seriously enough. Do you think they're taking it seriously enough? And how would you think about that criticism?
- [53:03]
A
I think the way I would answer that is I think there are definitely individuals taking it seriously and the bitcoin core developer community is composed of a bunch of individuals without really an authority structure around it. And that was by design. I think that's kind of inherent from feature, a positive feature of this community. I think the challenge is individual efforts for a cryptographic migration like this may not be sufficient, you know, which is, which is why I think that's where the criticisms stem from. Where if you have, let's say, the Ethereum foundation, basically all on the same page and being like, hey, we have a broad plan, it's 50 steps. You're going to do this, you're going to do this, you're going to do this. I mean, I think it at least acknowledges that the scale of the problem is quite big and it's probably bigger than any one person or even one team. So I do believe that for bitcoin to solve this problem will require more than the efforts of an individual core dev. I mean, I think it will require significantly more because we talked a lot about Quantum in the timeline. But I think the piece, one of the motivations for starting Project 11 is when you think about the work that has to go into actually fixing these decentralized blockchains, even after you decide to fix it, once you fix it just the work to actually fix it is extremely intensive because of the distributed nature of the software and because so much relies on the, that strength of the cryptography.
- [54:27]
B
Yeah. And there's a lot of money at stake.
- [54:28]
A
So there's a lot of money if it goes wrong.
- [54:30]
B
Yeah. And so, so I think we published a report which I know you commented on, really just highlighting the work that is being done in the bitcoin community. I feel like the difference is what you said. It's much more decentralized than other communities. Even Ethereum, which is in my view quite decentralized as a blockchain. It's got a foundation though that pays for development. Where's that? I mean bitcoin, to the extent development is funded, it's several nonprofits funding volunteers basically. And so it's harder.
- [55:00]
A
Yeah. And two things can be true. That model can be great and can be a strength of bitcoin. But that model can also make it hard in the face of big challenges like this. But it doesn't mean that it's impossible to solve this problem if people come together. But I think that's just what they have to do.
- [55:14]
B
Yeah. So I agree. I think that to me, even if I want to. I'm going to ask you the big question in a second about timelines. But even if it's extremely unlikely that that quantum computer ever gets built because like, you know, at a high level we talked about factoring earlier, like what is the highest number that has been factored by an existing quantum computer? It's quite low. Right.
- [55:35]
A
I think technically people like to point to the number 21, although we recently ran little competition where people factored a number Significantly bigger than 21 but still far below a 256 bit number. 256 bits is the number of atoms in the observable universe. Right.
- [55:48]
B
So, and that's what you have to
- [55:49]
A
do to, that's what you have to. So there's still a huge gap between,
- [55:53]
B
I mean it's, it's again, there's been improvements, but it's very early. So. But my point is sort of, and I, and I, I know you, you've said you agree with this even earlier in this conversation. But even if I'm skeptical that it even ever gets built, there's so much money at stake. I care so much about bitcoin and the security of self sovereign digital assets that surely if there's Even like a 1% chance it's worth taking seriously, that's,
- [56:15]
A
that's exactly how I feel about it. I. In my view, you don't need to be convinced that a quantum computer will ever happen. You just need to be convinced that there's enough of a chance that it might in a relevant timeline. And I think that bar is much, much lower. And there's a bunch of reasons why I think that, but we can, we can talk about it.
- [56:31]
B
Well, I feel like the Bitcoin, there has been movement I've observed in the Bitcoin, not just developer community, but in the diaspora, I think, which is because I think if you really talk to someone who's serious and not trolling, I think whether or not they agree, there may ever be a problem. There are plenty of smart people I know who just think it's physically impossible. I'm not qualified to know the physics of that, so I'm not focusing my work there. But most even of them will say, like, look, why I'm pushing back so hard is that like, we could screw up like a hard fork, if it's a hard fork is required or a complex upgrade, or if we're going to rip out the signing algorithms we use now. And that has its own risks. And I think this is where it's sort of like, I want to work on it. I think we should take it seriously. But it doesn't feel like we need. I'm not saying. And you're not either saying roll something out tomorrow. Right. There's still a lot of work that has to be done figuring out what to do. Right. Even if you agree something to do something.
- [57:29]
A
Yeah, yeah. But I think it's interesting because you hear these comments often made together to kind of say, why not to do anything? But I think they actually pull in opposite directions. Right. One is quantum is impossible. Okay, maybe. But you know, there's no proof, there's no physical or mathematical proof that that's true. By the way, there's not one that there's no proof that a quantum computer is possible either. It's just unknown. Right. So you got to roll the dice that you know your bet is right. And then. But then though, if you also say, hey, well, we should. We, you know, we don't know it's possible and we don't want to accidentally screw things up. We should wait to roll out, roll stuff out. But I think then you end up in a world where you risk it is possible and then there is a quantum computer and then what are you going to do at that point? Rush something out? And so it's like, honestly, it's better to Start early when there is no urgency.
- [58:12]
B
That's a fair point.
- [58:13]
A
That's a very fair point. Make sure everything is tested. That's a guaranteed way to rush.
- [58:16]
B
It's very hard to do because of the uncertainty that you said. But basically there should be some equal weighted amount of likelihood and work being done at all times. Problem is we won't ever know the exact likelihood. So I think that's why I focus on really just being supportive of, you know, the equivalent of getting off. Zero. Right. Like, the risk of doing nothing is pretty high. The risk of doing something and taking it slow. It's not that high. I mean, or the cost is not that high.
- [58:43]
A
And when you're doing risk mitigation, you've got to think about the likelihood of the event and, and how serious it is. Right. And think about, I mean, it's. The easiest way to think about it is seat belts. Like, I don't intend to get in an accident every time I get in a car.
- [58:52]
B
It's not that likely.
- [58:53]
A
It's very unlikely, actually. But I still wear a seatbelt.
- [58:56]
B
Because if you do, it could be deadly.
- [58:57]
A
Yeah, you might die. Right. But if you do, if you don't, or if you don't wear one. But if you do wear one, you might live.
- [59:02]
B
So I will say that the institutional investment community is way ahead of the, like, online crypto community in this. And it's really just in the last few months, like I think we talked and I talked with you and Nick about this six months ago, and I went to that. We were at that Presidio. It was a great conference. I was there for exactly like 14 hours. Like just like quick trip out to San Francisco last July. I didn't hear a single peep about Quantum from any of our clients or counterparties by then. And at this point, I wouldn't say it's a deafening roar, but it's a question every single person has. Like when that Google paper came out that was very widely publicized generally, which did deal specifically with cryptos like Bitcoin and Ethereum specifically. I mean, I got, we had questions from everyone about it. So it's, it's, it's at the top of people's mind. I don't think people sold Bitcoin, you know, in October till now because of it. I do think some allocators are using it as a reason to sit on their hands.
- [60:00]
A
Yeah.
- [60:01]
B
Until it's sort of. They feel more comfortable with where, where the, where it is.
- [60:04]
A
I agree with that. And my personal view on Bitcoin's development priorities. I guess I challenge any bitcoin core developer to point to a more existence existential technical issue. I mean there's definitely things you could make Bitcoin better but I guess like this is kind of the elephant in the room.
- [60:18]
B
Yeah. Like what do you suppose? I mean we can improve scaling and do stuff like that but that's new
- [60:21]
A
features or capabilities utilization is barely 1%.
- [60:24]
B
That's what I mean that we can work on new tech or give some new op codes to give more features or not.
- [60:29]
A
Yeah.
- [60:30]
B
The only other one I would add though is that like, like Claude Mythos and like the threat of AI discovering unknown vulnerabilities. Also very high to be clear.
- [60:39]
A
Totally.
- [60:39]
B
And urgent today. Although I mean there's not much to be done except I mean Bitcoin is the most widely reviewed source code in the world probably. Right.
- [60:46]
A
Yeah.
- [60:47]
B
Before we wrap, I know we have, we have a hard stop in about one minute. Project 11, tell me what you guys are doing for this problem. I know you're raising a lot of awareness. I will point out. I know I saw interesting story about Project 11 working with Solana and helping Solana developers test out some PQ crypto. Sort of what is. What do you guys offer today and what's your vision?
- [61:07]
A
Yeah, amazing. So yeah, what do we do? Raise awareness and get yelled at a lot on X. That's thing one. Maybe that's what I do. But yeah, generally look, I mean our future proofing blockchains has been what we've done. So we actually launched the first and only ever in production post quantum secure system for bitcoin. It was a simple cross signing service where you could generate new post quantum keys and attest to owning your bitcoin in a way that could never be forged. Right. So it doesn't really protect your bitcoin, but it's a pretty proof of ownership that can't be hacked. So that was like it was kind of a demo that we did last year called Yellow Pages. Since then we worked with a bunch of different protocols as well as participated in a lot of conversations in the bitcoin community about how should post quantum cryptography be applied. There's a bunch of trade offs that kind of come with a lot of these new post quantum schemes. And so what we aim to do is build reference implementations and work to test those implementations in reality. So the Solana partnership that you just referenced is we took one of the two NIST standard schemes. NIST is the National Institute of Standards Technology. One of their post Quantum recommended algorithms. We plugged it into Solana, and we saw what happened. Spoiler alert. I mean, things work differently and more slowly because these signatures are just bigger and slower in general. And this is going to be a problem that every blockchain has to face. And so these implications are really what Project 11 is all about. Let me make one last comment. Just dovetailing off the Mythos thing. Look, I think obviously Quantum is a threat and it's going to cause a big dislocation in cryptography, but there's no guarantee that'll be the last dislocation. Mythos and AI, as it gets better and better, you might. You might find yourselves in an arms race, or we might find ourselves in an arm race where AI is always getting better and the cryptography is not catching up fast enough. And that requires basically this ability to, you know, to embrace what's called crypto agility. And this, I think, for decentralized communities like Bitcoin, and maybe I'll end here, is like this muscle where we all have to come together and coordinate as not just a group of. Not just a bunch of individuals, but as a group may become relevant again sooner than we think. It may even not necessarily be a quantum computer that forces a change. And so I think it's a good thing to start talking about this now for a variety of reasons.
- [63:05]
B
There we go. Alex Pruden, co founder, CEO of Project 11. Alex, thank you so much. And by the way, you'll see us on stage. Bitcoin conference, I don't know, April something, a couple of weeks talking about this topic. Thank you so much, Alex.
- [63:17]
A
It's amazing to be here. Thanks for having me.
- [63:19]
B
That's it for this week's episode of Galaxy Brains. Thank you to our guest, Alex Prudent, Co founder and CEO of Project 11, and our good friend Bim Nadabi from Galaxy Trading. Everyone have a safe and happy week and I will see you next week. Thank you for listening to Galaxy Brains, the weekly podcast from Galaxy Research. I'm Alex Thorne, head of Firmwide Research at Galaxy. Follow me on X at Intangiblecoins. Follow Galaxy Research on X, L, X, Y Research. Read our written reports@galaxy.com research and don't forget, if you like Galaxy Brains to like and subscribe on your favorite podcast platforms like YouTube, Spotify, Apple Podcasts and more. We'll see you next time.