Getting To The Bottom Of Quantum w/ Rearden

Okay, guys. Today we have on uh Brandon Black, known on Twitter as Rearding Code. He's one of my longtime internet friends, and I've been wanting to have him on and now we have a subject hopefully that will be invigorating. It's top of mind, and yeah, Rearen is pretty spicy and so yeah, I think this will be a fun conversation. We're going to talk about Quantum today.

[Speaker Name]: Rearen give us a bit of background and then also in doing so give us a bit of background as to how you have been one of like at the forefront of this quantum discussion at least on my timeline.

[Rearden]: Yeah sure. So background on me I mean I've been kind of in computer security for 15 close to 20 years now. Worked in in Amazon payments and at a bank security company before getting into Bitcoin coding. So kind of thinking about security has been my bread and butter for a very long time. So that's like the important background for this discussion I think. And then in terms of quantum, I don't know how I got into the middle of it. I'm a default skeptic. Whenever there's excitement about something, I come at it skeptically. And so just to to give that, you know, full disclosure, I do default the skepticism. And so when I started reading some of the the newer quantum discussions in the last maybe year or so, there's been more noise about it. I really wanted to dig deeper. I wanted to make sure like I tried to check my own biases. And so that's that's actually how this happened is that I went to check my own biases. And the and the deeper I dug into it, the more skeptical I became. So then I I was like, well, if my default skepticism is only getting reinforced as I dig deeper, I'm going to start talking about it. And and at that point, I'd also learned a lot about it. So I learned about, you know, what they're doing and how they're trying to deal with the quantum challenges and everything. And nothing that I read relieved any of my skepticism. it made me more skeptical. So that that's how I got here.

[Speaker Name]: Okay. Awesome. Yeah. And that's basically what I've kind of inferred from the timeline. Like I've always uh you host occasionally these spaces on Twitter. By the way, this is um a shout out. You guys should listen to these um where you guys talk about Bitcoin development. Um Shinobi's on there. Like some of the um yeah, like guys building on Bitcoin are are in there. So I my point is I've respected you as somebody who's like, you know, generally aware of what's going on and also um like you're you're fairly technical. I think that that's something that I would also um you know want to make sure people have some context here. I'm not bringing on like some random guy who just dug into it like you've been you know technical and in Bitcoin for quite some time. Um okay so when you started digging into quantum like what was your like you said a year ago this is really like kind of festered in the past I don't know couple weeks. We'll get into that but um a year ago like what was the initial like okay I want to look into this because of you know the discussion and like what were sort of your initial takeaways?

[Rearden]: Yeah. So, I think I'm trying to I'm trying to remember the timeline, but I know um I kind of got pulled into it partially because people I I respect started looking closer into it. You know, um Hunter Beast is an internet friend. Y and so when he started pushing to to do more work on Bit 360 or or or other quantum resistance in Bitcoin, you know, that catches my attention. Anytime someone wants to change Bitcoin, uh it's an important thing. Uh and frankly, I want to change Bitcoin. So, in another sense, that's another reason I got into the quantum. I was like, "Oh, maybe this is a good reason to to make upgrades to Bitcoin." So, that's what that's what kind of brought me into here.

And so, some of my my early thoughts were basically one in in reviewing the early work that Hunter Beast and others were doing on quantum resistance, I found myself really skeptical of the quantum resistant signature algorithms as appropriate for Bitcoin. I there there are some algorithms out there quantum resistant that that as far as I can tell are cryptographically sound. Um there's there's some noise on the timeline right now of people saying oh if we push risky crypto into Bitcoin it could break it. I think a a carefully chosen algorithm does exist that would be secure. So I just want to be clear. I don't I don't think it's all bad crypto but they don't seem appropriate for Bitcoin because they're large and we know I mean Bitcoin everyone teases Bitcoin for being seven transactions per second. Oh, it's it's not going to be Visa, right? And and if we were to add one of the current quantum resistant algorithms, I think the best out there is almost an order of magnitude larger than elliptic curve signatures and keys. So, we'd be dealing with then less than one transaction per second if we adopted one of those postcrypto algorithms. So, so while I wanted to change Bitcoin, I I then had that, oh, we're not ready for this yet reaction.

Okay. Well, well, if the postquantum cryptography isn't ready yet, look, if the if that was ready, I'd be like, yeah, you know what? I I let's just do it. Let's add it to Bitcoin. Let's make Bitcoin secure against this potential attack. There's no problem here. Okay. So, it's not ready. Is this really a risk? Are we in trouble? Do we have to figure this out or or can we take our time is really the question that then came to mind. So, I started looking deeper into the claims of the quantum uh researchers and and you know what is what what do they base their timelines on? Why is Hunter Beast and others feeling like this is a a soon problem as opposed to a later problem? Um, and that's where I got really really skeptical because the the example I mentioned it on on my timeline not not too long ago. Someone started claiming this thing Nevin's law, right? They're saying, "Okay, Moore's law was for classical computers and Nevin's law is for quantum computers." And Nevin's law basically says that on some, you know, one or twoyear time range, we'll be doubling the number of logical cubits in our quantum circuits.

And because quantum calculations can represent an exponential state within their calculation. Um I mean and this is this is the the fundamental promise of quantum is in here right that classical computers every additional bit you can process is is linear in how much more stuff you can you can actually represent. But because the quantum superposition where each state represents both each bit represents both states of that potential bit. Additional bits represent exponentially more total state space, right? Because if you have, you know, two to the two is four, two to the three is eight. That's how quantum bits represent stuff is they represent all eight of those states. So the Nether law says that we're going to get double exponential processing power increases. And I mean, when you hear the word double exponential increase in something, you've got to be a little skeptical. So that's that was like one of the the the things that got me further down my skepticism. So then I'm like, okay, well, how fast are quantum bits scaling? And and then it gets really confusing. Um, so in the last year, we had like the we've had a bunch of different announcements. Oh, this chip has these many cubits and that one has those many cubits and all these announcements, right? And and so the neans law depends on how many logical cubits are in a chip.

[Speaker Name]: Okay, before sorry to interrupt, but before we go on like maybe for 30 seconds and I'm not trying to dumb this down to a level where like you know nobody would find it interesting, but like explain what a like again just take as short a bit of time as you want. What what is a quantum computer doing and what is its like uh you know three sentence risk to Bitcoin if you will? Potential risk potential.

[Rearden]: Great great sid track. Um, so the thing that a quantum computer can do and and I would say like the the theory here is sound as far as I can tell. If if we understand quantum mechanics correctly, which is a little bit of an if, but you know, I think it might be true, then the what the quantum computer can do is you can load it with some state about the world and kind of a a construct the quantum computer as a circuit that represents something about the real world. And once you've loaded that in, it can represent all the possible final states in its superposition of of cubits. And and so instead of you know again with a classical computer, the computer is always calculating one specific state of the potential of the algorithm you're working on with a quantum computer if you've designed the computer and the circuit within it correctly. It represents all possible outcomes of that calculation. and then you you give it an input for that and it resolves to the correct output for that input. And and so that's kind of the the the trick or the benefit of the quantum computer is that instead of having to let's say iteratively calculate each possible outcome when you load it up and then kind of trigger it or or or prime it, let's say, with the with the state of your system like I've got this public key. then we apply it to the circuit that calculates public keys, the private key to public key conversion, then it'll resolve to the correct private key. And it's a little more complicated than that. And Shor's algorithm, you technically can't do this directly. It it uses a a technique called period finding, but the the general summary is still that you represent the the way in which keys in Bitcoin or other elliptic curve systems rotate around the Oh, man. Okay. I got to explain more here.

[Speaker Name]: Let's take one step back. Okay, you can go ahead. Yeah. Yeah. I want to get into the the elliptic curve just a little bit here. Yeah. Yeah. Perfect.

[Rearden]: The key thing that makes elliptic curve cryptography difficult for classical computers is the fact that it's a discrete elliptic curve. It's not on a all real numbers from 0 to infinity. There's a boundary and the boundary is is some specific number which is called the uh oh shoot what's it I whatever there's a there's a a prime number size of numbers within which the elliptic curve is calculated. So it rotates around as you as you add within the elliptic curve field you're you're moduloing going back down. So you're not just like continually going up through the quant the um elliptic curve. You're rotating around a modulo. And so that is what then lets the quantum computer solve it as well in theory is that the quantum computer can figure out the periodic nature of that elliptic curve and that lets you go backwards in theory again to a private key. Normally public key to private key is unreversible because a classical computer cannot figure out the period with which you move around the elliptic curve. So as you go through it, there's this impossibility of how do I go back to the previous step because I don't know the period. The quantum computer can do period finding using shores algorithm to go backwards in that step of of stepping around the elliptic curve on this uh modulo process.

Got it? Uh, so that's that's the danger to Bitcoin is that if the quantum computer can do this for large enough input sizes, 256 bits, then it can go from a public key to a secret key, which obviously then unsecures all of our Bitcoin. Well, you know, some some details to be worked out there, but uh that's that's the danger is that people could go take your public key that you publish because it's public and find your Bitcoin. Um this is a danger for all other cryptos that are not specifically quantum persistent. You know uh Ethereum also uses elliptic curve discrete or digital signature algorithm. So we're we're all in the boat together essentially.

[Speaker Name]: Okay. Um and maybe you can relate this again back just for a little bit of context and like you were talking about classical computers. One of something I've been seeing you been trying to follow this to the extent that I can on the timeline. And I've been reading a lot of what you've said and I I you mentioned you you personally maybe even had more concerns that a classical computer would um be able to decrypt keys before a quantum computer. Is that correct? Or is it just like uh go ahead but yeah maybe you can relate that also to like again differentiating like this the quantum side versus the classical side why you're saying like look like I think currently it's probably more of a risk that we just get you know extremely powerful classical computers versus even this like quantum idea.

[Rearden]: Yeah. So the the idea there would be I mentioned there's this period finding thing you can do using Shor's algorithm and and we've seen in other crypto systems over the years examples that come to mind are when I was a young man in learning computers and stuff we used uh 1024bit RSA keys yes and algorithms for factoring prime numbers improved to the point that or sorry factoring co-primes improved to the point that RSA24-bit was no longer considered secure and so we went to 2048 bit and and that's remained kind of probably fine but most people now recommend using 496-bit RSA keys. So we've seen the this process of both the computing power increasing and the algorithms improving in classical computing to weaken otherwise secure crypto systems. And this also happened in the AES world. you know, AES uses this um there's a generator function that changes the key material you're encrypting with for each block of data you're encrypting. And people have started to use algorithms to kind of start to work that backwards. So, if you can get even a small portion of the message, you can then find a few bits of the key potentially and start to to reverse that that key rotation function of the AES. So there's techniques that people do in the classical algorithmic work that reduce the difficulty of of of breaking these crypto systems.

And if both computing power increases fast enough and those algorithms improve fast enough, you end up eventually with breaking the crypto uh in this case breaking the uh public to private mapping for the keys. And so what what would that look like for a lift curve? But we would see someone in the classical world figuring out a way to transform the problem from this essentially impossible guessing game, trying to guess where backwards you would go on the curve to some other structure of the problem that is maybe similar to what the quantum computer would do, a problem we know how to solve. So, can you convert it to factoring primes that we're getting better at already? And can you then apply some techniques to simplify that factorization in one way or another? Can you use things you know about the structure of the curve itself in order to to to make the calculation less difficult and over time you know these advances they've been happening in fighting crypto forever that's why we upgrade crypto system that's why we don't use Shaw one in the internet anymore right so so that's the kind of thing I'd be talking about is just the gradual slog through improving the algorithms and growing the compute power.

[Speaker Name]: Got it okay so I'm I'm sure I'm going to come back to more technical questions that you know hopefully you can elucate but let's just now kind of zoom out for a second and and like talk about current context. Where do you find yourself if you had to like selfidentify on the spectrum of on one side? I'm on the spectum. Well, okay. Nice. So, that's the the quantum spectrum. Okay. So, on one side of the spectrum, you have um it's all a complete scam charade. This is all, you know, like basically hype and mania and nobody knows what they're talking about. That's one side. And on the far right, you have this is coming in 5 years. Um, so I know you fall closer to the left side, but like where do you self from what I've, you know, gleaned thus far, where do you like where would you put yourself? And then maybe yeah, like give us a little bit more just, you know, general ideas of like, okay, I think this this is possible or it's not possible. I think you've said yourself like you're skeptical it's even possible. So I I'll let you go, but you get the general idea.

[Rearden]: Totally. So So I I find myself in a superp position of two states on that. All right. where where on the one hand I can say with extremely high confidence you know I've I've looked at a lot of the work going on in quantum and near certainty we will not see a quantum break of elliptic curves or similar crypto systems in the next decade. So that's like near certain. So that's that's one part of me is this I' I've thoroughly researched there's just not a way that that happens.

[Speaker Name]: Good. I like those strong claims. Like those are those are good, you know, sound bites. Like people actually make like bold claims like this, not like some nebula stuff. But so anyways, go on.

[Rearden]: And then my emotional I've talked to these people. I've I've asked them hard questions and the answers I've gotten have been deeply unsatisfying. Uh that part of me wants to say it's an absolute scam. So emotionally all the way to the left, it's just a scam. But the logical I've done research, I've thought about it, I've I've kind of read the papers. That's just I'm certain it's not the next decade.

[Speaker Name]: Okay. So, I love this and I'm also glad I brought you on and maybe I'll get somebody who's more on the person. I'll bring Hunter Beast on cuz this is somebody who's working on um yeah, like BIP 360 and yeah, we can talk also about that in a bit. But, um so where is this disconnect? Like right now, for example, as you know, I've seen you guys back and forth a million times now. Nick Carter is writing this kind of series of posts on quantum and there's like zero from from my perspective there's almost no overlap at least when I see you talking about this it's like there's no coming to an agreement like and it almost appears as though you don't even agree to disagree right it's like no you're wrong and he's like no you're wrong and there's not much of like um yeah there doesn't seem to be much common ground so let me ask you I'm not asking you necessarily to steal man but why is there such a divide like why are There are people like you who I I mean I think you're very smart. I don't know enough about quantum computing to know if you're an expert on quantum computing, but I think you're technically smart. So, but exactly. So, like I don't even know who the experts to listen to are, but like why is there this massive divide? Like I see you making rational points and then I see what I think are rational points from Nick. But, you know, to be quite honest, I can read this stuff and be like, you guys could be speaking it. You guys could be both completely bullshitting. I'd be like, "Yeah, yeah, that sounds good." Like, you know, both you guys are right. Spot on. You know what I So why is there this massive sort of like you know obvious like divide here?

[Rearden]: A big part of of why we see the divide is actually I think there's a few points here. One is the quantum computing companies and researchers are able to put forward this pretty compelling set of evidence of look we've solved this problem. We can put these many physical cubits on a device. we can and we've seen progress pretty decent progress in increasing the number of physical cubits on a device. We've seen pretty decent progress in reducing the error frequency within physical cubits. And so they can they can show here look at all this progress that we're making in these two domains. More physical cubits, fewer errors. And so that's where the I think the excitement comes from that people in the investing world and in the within the quantum field they're like look at all this progress we're making on these on these things. So that that contributes to a divide where whereas from the outside since I'm not a quantum researcher. I'm not a quantum expert. I'm just someone who thinks about things, right? That's that's really the best way to look at it. I'm someone who is very competent at reading research both and I I do this like in my own life for like medical research. Like I go to the source papers for everything that I care about. Um so I just have a competence in reading research. So and then so from my side I I look and I say okay how would I know if this was a problem? And the way that I would know that we have a potential quantum break coming is that I would see evidence of practical scaling against real world problems. And so I would say okay the last number that someone attempted to and published findings on factoring in the quantum computing world was 15. I believe it might have been 21. I get confused about that sometimes. But the point is it's a number that my six-year-old can literally factor. I double checked. I asked him and so and no one's done a bigger number, right?

[Speaker Name]: Yeah. So if if I don't Yeah, you're scaling this and that, but does that matter to the real world where we're going to eventually break a crypto system? And the only way that I again is not someone working in the quantum field could see yes, this is going to be a problem is if it breaks if factors a bigger number, if it does something harder than what it has already done. And so and so that's where the break comes is that difference between okay we're improving these things but do those things map to the real world they don't seem to.

[Speaker Name]: Yeah okay so I want to interject quickly again now back to your technical question and I was trying to understand this before I end so when you talk about this idea of logical cubits my understanding is that there's you can make all these cubits but a lot of them are really screwy right they like mess up all the time and so really you need like errors to be extremely extremely low for this to be viable and so that's one I I think so can you kind of explain this idea of error correction like how in order for these things to be viable for anything the error rate needs to be you know like orders of magnitude lower than it is now or maybe you know just a large decrease from where it is now.

[Rearden]: So the same way we talked earlier about how quantum computers can do exponentially more state mapping per cubit you add because of that property of being in two states at once. The same property means that errors have a similar exponential scaling where if you have one cubit you have some error rate but if you add a second cubit now you have that error rate uh exponentially increased and so each cubit you add whatever error rate you started with the error total error rate grows exponentially as the state space grows. Got it? So in order for a quantum computer to do its I'll call it magic for the moment but to to do the computation it has to have an error rate that's low enough that as we've added enough cubits and to do to break elliptic curve a common 256 elliptic curves you need somewhere around I think it's 2300 so between 2,000 and 3,000 logical cubits are stable enough that you can get to that many cubits and still have something that that's distinguishable from noise. So just thinking kind of the way the math works here. They think you need to have something like 99.999 uh% correct from so 0001% error rate on each logical cubit for you to get to those 2,00ish cubits and get an answer out at the end that is not just noise. Okay, so that's that's what we're talking about here is that the errors scale exponentially with the state. So the the benefit of quantum was also is kind of achilles heel there.

[Speaker Name]: Yeah. Okay. So then go back to this um like the the number that that a quantum computer is factored. One of the other arguments that I see from you and and others on your side. It's kind of like a running joke. Um I think uh Casey Rotormore has this meme of like the the goose like saying factor 21, you know, like the um the goose honking at somebody. So, one of the other um sort of things people bring up is that like when people are showing a lot of this that new research like the breakthroughs, right? A lot of times when people are reading the research like you gave like I'm saying this in a really reductive way, but like you gave the computer the answer already and so so explain that to me like what is that like what is that common trope which is like actually you're you already told it the answer so like that doesn't really count. And it seems like this comes up a lot so I'm not like you know pulling something out of nowhere.

[Rearden]: Yeah, I've been I've been actually reading about that this morning because uh Bob McKelith and the and the quantum researcher are going back and forth this morning about it and that does vary a lot between the different claims you see from the quantum folks but in general because the the quantum field is is I think they know they're struggling to map from their their improvements in the various domains I mentioned to real world results. they're they're trying to find ways to get that mapping to happen. And so so what they've kind of started moving towards now is that you you you try to take as much of the problem you want to solve and move it into the classical domain. Do a bunch of pre-processing on the data that you want to solve for to make it simpler in the quantum sides. like how how can we do as much work on like let's say GPUs before we give it to the quantum computer so that the quantum circuit itself is as small as possible because as we talked about the quantum error rate scales exponentially with the number of cubits we want to have that circuit be as small as possible they're like oh we can we can pre-process this much and then we can actually post-process that much so now the quantum circuit is just this small piece in the middle that does the key thing we need from the quantum circuit which is kind of reduce this exponentially sized problem space down to something the quantum circuit can solve in in hopefully linear number of cubits which you know could be that that that's the theory and so that's what we're talking about is are they cheating essentially in how much pre-processing they do and so so Bob and this other guy this morning are going back and forth with when the quantum computer they they made the claim and it published a paper that it factored or sorry that it um reversed the elliptic curve discrete log problem for a sixbit elliptic curve. So that was the the claim that was published and Bob is claiming that they basically pre-processed the difficult step and they didn't and that and his claim is basically that they they accidentally pre-processed the difficult step and then gave the quantum computer the easy step. They they kind of they didn't realize that the way they had done their pre-processing isn't possible for a larger curve. It was it was possible because on a six-bit curve the the classical computer or my six-year-old can just do the math and and and pre-process or actually could solve the whole thing. And so it's easy on a six- bit curve to accidentally do the hard part in the classical circuit and then just give the easy part to the quantum. Oh, look, it worked. So that's that's basically where we're at here is did the quantum computer do the difficult part for real or did the classical pre-processing do the hard part for it?

[Speaker Name]: Yeah. Okay. Okay. I wanted to go back to something you said earlier about this like you I think it was referred to as double exponential scaling and you you kind of laughed this off and I think there is a tendency right like I see this a lot in crypto um like outside of Bitcoin which is like most people don't understand exponentials right or like exponential growth right and so it's at times it seems like a little bit belittling right and I think this is probably where you fall and and it's like hey guys like it's not that I don't understand exponential growth it's at like what you are showing me is not evidence of exponential growth. Um I'm just laughing because yeah this is a frequent you know and and people love to site Morris law or people love to site you know right before the Wright brothers took off their people said they would never fly. It's like okay yeah so one uh person who is and I'm interested to hear your thoughts on this but is thought of as somewhat of an expert or a you know thought leader is Scott Arensson and he was I think the one who um was talking about this idea of double exponential or what whatever it was he was talking about is there anything that in your mind would make you think oh like maybe we are seeing exponential progress uh maybe this is different and that's that's the main point I'm making th this time is different right this transcends any understanding of you know Moors law or you know mechcast law scaling networks like these kind of understood ideas we have is there anything that would change your mind on that?

[Rearden]: Yeah absolutely and I I made that post specifically in a conversation with Nick this this week. Yes, I I I am on the lookout. I am reading quantum papers when they come out. Looking out for a demonstrable cycle where a quantum technology for cubits has been developed, have been manufactured at one size with some error rate and some applicability to the real world. And then a year and a half later or a year later, another generation of that technology comes out that that is now bigger and can do more real world work. And that so that's the thing that that like the the thing people kind of get wrong about Moore's law is that it was an observation not a projection. Moore observed that transistor density was scaling about double every 18 months. And he made note of that and it was published in a magazine that people took it as Moore's law. and and people because they called it Moore's law that it was kind of a misinterpretation. It wasn't a law. It was Moore's observation. And so that's where the quantum folks I think get ahead of themselves. They say we're going to scale exponentially here without having a de demonstrated cycle of we have scaled exponentially for two years, three years or something like that. So that's the evidence that I'm looking for. And again, I'm I'm actively looking for that evidence because I know that I'm biased toward skeptical and I want to break that bias as early as possible so that we don't end up in a bad situation. But so far there is no evidence of that. There's not been and we've seen you've probably seen it in the quantum papers that that you've kind of just seen the headlines for. Even what we don't see is I we built this technology and then we built that same technology bigger with more real world functions the next year or two years later. Instead, what we see is this new technology and then that new technology and then the third new technology and then the fourth new technology. They keep doing new things because they haven't found a technology that scales. And that's part of what drives my skepticism is that it's not we're not scaling the transistor. We're not scaling the optical cubit or the superconducting cubit. We're not we're we're trying we're throwing things at the wall trying to get something to scale.

[Speaker Name]: Let me take a moment to talk to you about Ellipsus Labs, a team I've known for a while building some of the most interesting products in the space. The Ellipsus team is one of the few that I can consistently point to and say, "These guys really know what they're doing." They're backed by Paradigm, Electric Capital, and Han Ventures. The founders, Eugene and Jerry, have experience at Citadel, Jane Street, and the Salonicore team. Since launching their orderbook decks, Phoenix, they've done over 80 billion in trading volume by making onchain order books competitive with centralized exchanges. They're also the team behind Gavl and Soldfi. I would consider them the foremost experts in crypto market microstructure. The best part is Ellipse is hiring for New York-based engineers. Work with a small focused team who are resultsdriven, collaborative, and use a modern stack. If you're an engineer who wants to work on infrastructure that's already proving itself in the market, go to ellipsuslabs.xyz. Okay, now back to the show.

[Speaker Name]: Scott Arensson is somebody who I'm citing him again. he has his blog and he's more on the skeptical side at least historically but maybe recently given the hype he's made some you know more kind of you know I don't know audacious claims but who let's say that I don't want to read research like you would research um I have different passions and but at the same time I don't want to just throw stuff into catchy and get it to like slop stuff back to me like are there any resources that you think are good but also to the extent then it can sort of gets away from the bag bias. And I'm not necessarily accusing everybody of the bag bias, but like it does seem most of the strong claims come from people who are have some like t I mean especially the the quantum labs and stuff, right? Like obviously they're and they're raising a ton of money, right? Like so um are there any good resources for somebody like me who wants to understand this or at least keep a breast, but I also don't want to read like the 40page research papers that you're reading?

[Rearden]: Yeah, I think I think both the the chain code research paper and the the publication that came out after the Presidio quantum summit that they did have been pretty measured, you know, and they fall more into that unlikely in the next decade, but we'll keep an eye on it camp, which which I think like that's like I said that's the best I can say with like very high confidence. My logical faculties say that makes sense that very unlikely. I would say almost zero likelihood, but you know the claim that's fine. And either one of those are kind of reasonable things to think in the next decade we're going to keep an eye on it. So I would say look at those two resources first and then other resources. The other the other thing to to keep an eye on is just to watch out for people who who who say the sky is falling. And maybe I'm I'm particularly attuned to that because as someone who was a a kind of old school libertarian gold bug for years, you know, like the libertarian gold bugs have been saying that hyperinflation is around the corner since before I was born, since 1970, right? Hyperinflation is around the corner. And those kind of claims just tend to be wrong. I think Scott Adams uh of Dilbert fame has actually talked about this quite a bit where most of the time humanity says the sky is falling and it turns out that it's just an engineering thing that we will solve when the time comes. And we've seen that with Y2K and we've seen that with you know a whole bunch of things in human history where the sky isn't actually falling and we are actually as a species pretty good at solving things in a timely manner but