Is DePIN Overhyped? (Breakpoint 2024)

0:00Hey, everyone. As you know, Solana is one of the ecosystems I'm the most interested in. I've been incredibly bullish. I love to see the vibrancy of what is happening in the space. Unfortunately, I couldn't make Breakpoint this year, but the stories have been amazing. I heard the talks are incredible. And we're really proud of Real Vision to give you talks from Solana Breakpoint 2024 in Singapore. I really hope you enjoy them.

0:37Check, check. All right, sick. Sweet. We're here. I'm super excited to jam on a motion that is very near and dear to my heart. I think I have some version of this debate five times a week. So we have some gigas, as was rightly put, on stage to reason about some of this stuff with us. We've got Matt, head of research at Pintera, and Abhay from Helium. And so we're going to get right into it. We don't have a ton of time. So I'll lay some broad structure and some ground rules. I've been told that I can do whatever I want here. So let me lay it down a little bit. I have this gong. Anytime there's a banger, I'm going to hit the gong.

1:20Please optimize for bangers. Some more bangers. Most debates, crosstalk is not recommended. I highly encourage crosstalk for this debate. So, you know, take your time, but be civil if you can. And so run of show, I think we start with opening statements, as all good debates must. And then we'll kind of get into the specifics, the operative words here, which are hype and capital inefficiency. And then towards the end, we'll try and get into some meta questions about what we're all doing here. So maybe we'll start with opening statements. Matt, as you're defending the motion, we'll start with you and then we'll go from there.

1:58All right. Thank you, everybody. I am your resident D-Pin skeptic here on the left end of the left curve. So thank you all. I'm a research partner at Pantera. We have D-Pin Investments. My life is better. If D-Pin works, I will be working with our portfolio companies to make it work. So I'm in good faith here, but I'm here representing 2019 me. So I was a PhD economist and I came up in the trenches working in tokenomics for IoT, early wave D-Pin sort of projects. And I ran into some problems. And the problem was you would often find these things that had some questionable designs and then you'd sort of have a big Rube Goldberg thing built around it.

2:35and you'd have to talk to the one person in the organization who understood that because they made it. And it'd grown too complicated for anybody to criticize it anymore because only one person understood it. And I'd have to sort of talk them down and find the game theory model it was. And so that's why I'm wearing my token engineering shirt here is that's where I'm coming from. And I would say the reason deep end IoT projects are especially susceptible to this sort of Rube Goldberg buildup is that usually under capitalism, you will find that the capital hires labor, right? And the exception to that, when labor hires capital, which is what Deepin's kind of trying to do, right, your decentralized ownership, is when monitoring costs are really low or when capital costs are really low.

3:13Actually, I should say and. Monitoring and capital costs. So think of like a law firm, right? A law firm, everybody's a lawyer. You can kind of see what each other's are doing, right, because you're all lawyers. Each person's billing by the hour, so there's a sort of market metering of what each person's doing. You're not trying to, like, figure out the complementarities of the skill sets. And so you can be a partnership. Also, what do you have to buy? Paper and a pen, right? A computer, pretty easy, right? Unfortunately, I think D-Pen fails on these two counts. Generally speaking, monitoring costs are high, right?

3:42D-Pen, by its definition, is a big Oracle problem with a token incentive putting pressure on this already fragile Oracle problem, right? And then two, capital costs, which I don't know how I'm doing on time. Might have to talk about it next question. But I think capital costs, it varies, right? Some projects can be a little better and a little worse on this. But my fear with D-Pen is that the overall long-term capital trajectory is still pretty high capital. So I think there are challenges. I think they can be worked through on a case-by-case, kind of like case law basis. But in general, I'm pretty skeptical of D-Pen as the name of a category.

4:11To me, it seems like naming a problem, not a solution. Thank you. All right. Okay. All right. Well, yeah. So, you know, the thing that I was thinking about a lot with this topic is that I think Matt and I actually truly agree on a majority of things, except I totally disagree with you here. So that's what we're supposed to do, right? So I'm going to start with, like, set the tone with a couple of broad statements. You know, I think crypto in general is not a human scale industry. We have not built the thing for every single human on the planet yet. You know, I can tell you a couple of reasons for that.

4:50But I'd rather you just give me your 12 seed words like every other rug site out there, and maybe that's the fastest way to do it. And I think that human needs really... When you're building something that's human scale, you solve something for human needs. And I think it's pretty well represented in Maslow's hierarchy. At the base layer, you want to have physiological things taken care of, food, clothing. I would add connectivity, water, and power to those things, because these are our basic human needs on the physiological sense. Then you need things like safety, shelter. And I actually think crypto does do something here as well, specifically around financial security.

5:30It's not just job security, right? So there's an opportunity there as well. And then you have things like love, belonging. That's where maybe things like SocialFi come into play. Esteem, crypto Twitter, obviously. Self-actualization, I don't know, whatever Vitalik's on right now with that song. Like that's probably it. So, you know, I think crypto has the opportunity to really solve for Maslow's hierarchy. And, you know, for me, I think Deepin is really the only real point of all this infrastructure we're building in crypto. So just with that base argument, I think that there's, of course, it's, we need this capital in here.

6:08We need this hype in here because ultimately crypto is for Deepin, not the other way around. All right. Some bangers in there. All right. So let's start with the operative words here, right? Let's do capital efficiency first, and then we'll talk about hype. Abhay, billions spent. Only millions in revenue. What's the gap? How do we make this happen? Are you talking as an investor right now? No, no, no. I'm talking as a general macro observer. The point stands, right? Tokens are an expense. We're building out all this infrastructure. the demand side seems to be picking up and the ramp is there. I think some people are more optimistic, some people are less.

6:53So tell us a little bit about what gets us there. Yeah, I mean, I guess I'll make a lot of arguments in the connectivity space. It's the thing that I spent most of my time on. I'm sure that we can make similar arguments in other spaces like energy or you know, maybe even water someday, hopefully. You know, I think that one way that I look at this is the US ISPs have basically spent, you know, almost$2 trillion between the 90s and 2020, roughly. It's self-reported, so you got to take that with a grain of salt, but like, that's roughly how much they've spent, right? And like, in the last several years, you're seeing anywhere between like 60 to$80 billion being spent every year, which is insane, right?

7:30What have we actually gotten? I'm just thinking about this in the US, again, like it's, we're in Singapore, so maybe it's not the best argument, but like, you know, like, that's how much money has gone into building connectivity infrastructure. And this is a core, like, again, it's core to what we need as a humanity. And, like, you know, traditionally, you think about this, governments are the ones that are making a lot of these investments as well. And you have things like the FCC spending, you know, anywhere between, like, five to ten billion dollars over the last ten years. You have the USDA trying to get all farmers connected.

8:05You're trying to get all these places connected, But ultimately, like, there's still like almost 5 % of the US population that's uncovered by broadband. That's nuts, right? And, you know, you have the IRA that was recently passed in 2021. And, you know, I forget exactly how much they put in there. But it was on the order of like$50 to$70 billion just for broadband. When they were really focused on something like putting fiber into, you know, into every single home. And fiber is insanely expensive. You know, tens of thousands of dollars just to serve an urban environment per mile. And then if you're thinking about a rural environment, you're talking$50 ,000 plus just per mile to get, you know, one farm broadband connectivity.

8:52If we're going to talk about capital and efficiency, let's talk about that first. I think fundamentally, Deepin is another way of building this kind of infrastructure. It's complementary to what's being done out there. and it is certainly, it seems to be a lot more capital efficient so far. We haven't seen the end yet and that's why it's worth investing in it right now. Matt, can I take the other side? Yeah, I mean, so I'm with you from an empirical standpoint. It's too early to tell, right? We're in venture, it's a time-honored tradition to chase a network effect until it hits an inflection point.

9:22Like, okay, yeah, it doesn't look like you've gotten as much as you've spent so far. I think that's totally fine. I think conceptually, I'm a little more concerned capital-wise because of the way a lot of these are structured. and I'm not talking about healing here, obviously. I think just in general, more of a deep end thing. The issue, right, is that the actual value that accrues to the network is usually not the network itself, meaning it's not what the people who are buying the devices are usually getting, right? And I think the reason for this is most of these industries are kind of dynamic, right?

9:52They require what we call dynamic capabilities where some new tech is going to come and you're going to have to upgrade to 5G or something like that, right? And so you're going to have to coordinate this mass of people who, you know, look, it's already hard in a blockchain space to convince people to accept like a free network upgrade, right? If you're a validator, to make them pay for new hardware is its own new challenge, right? And somebody's coordinating that. And then maybe more importantly, the structure of these things where you're having to design network incentive or incentives that are constantly fighting against smoofs or hackers or spammers, you know, which people do really well, that's a specialty, that's a dynamic capability that is going to remain within the organization.

10:28And that's going to be a hard thing to decentralize. And so that's sort of like in the economic sense, the capital, I think that accrues in the business is one that's going to be hard to decentralize unless you become a DAO, right? And I think that winds up being the kind of like long-term vision for Deepin is that the D is going to have to stand for DAO at the end of the day, right? And that the economic capital in it is going to only, the only way you can decentralize it is if you decentralize something that can then manage these dynamic capabilities, especially the upgrading capability, generating new hardware, convincing people to get it, et cetera, and then defending the Oracle being the two big ones.

11:01But isn't one of the problems that Deepens are trying to solve, and some of the more successful Deepens are effectively telling people, hey, you're going to be deploying this thing for yourself anyway. You're maybe deploying this thing for your neighbors anyway. So I don't completely agree with that upgrade problem argument because I think that, you know, if you have a coffee shop and you want to provide connectivity to your customers and maybe also provide connectivity to your neighbors, like you're going to do that anyway. And you're going to, you know, run really great fiber to the, to the site.

11:31You're going to put up a wifi router. Why not also support Helium mobile at that point? With you 100 % on that. So, so Helium does latent capabilities then? Sorry? It does like latent things. Like if you have a spare compute, if you have a spare device, I do think that's the better approach. Yeah. And I think that is the end goal of all dpens is to align existing latent supply. It's not necessarily... Sure, Greenfield deployment is really, really important, but ultimately that's what it's for. And the best dpens should be doing that. Yeah. You know a thing or two about spoofing. Of course. Of course.

12:05Well, I guess to that point, right? Like, Helium is one of the OGs, but arguably the first in this form of capitalization. Please get my cane off so that I get off stage. I was telling someone the other day, if you compare the deep-end market map from 12 months ago to the one that exists today, the one today is about 10 times as big, with 10 times as many names. How do you feel about that? I mean, I think that, well, one, I love experimentation. I've been in startups since God knows how long. I am also just as old. And, you know, I think that, like, if you're thinking about hype, if that's, like, kind of where we're headed, it is important for the market to support a lot of experiments at the same time so we can finally, like, kind of resolve to some, like, let's say true price or true value, you know, sort of output.

12:56And, you know, I think about, like, I spent some time in the food delivery space and the ride hailing space a little bit. And that was another one of these examples, right? We had, at least in the US, but also I guess there's some great companies here too. DoorDash, Caviar, Postmates, Uber Eats. Here you have Grab. I've been using Grab the entire week and it's been awesome. My God, it's better than Uber. Sorry, Uber. But it's been really great. And they've raised so much money. A former company of mine acquired one of these. That's where I spent my time on this. and like, yeah, they raised like$30 billion and maybe there's like$100 billion of market value right now, right?

13:37And I think this is what happens with some of these cycles. It is necessary for there to be experimentation. You know, different companies are going to take different approaches. DoorDash started in sort of suburban environments, whereas like Caviar and Postmates started in urban environments. And we have seen how that resolves to. DoorDash, at least in the US, is the winner. Grab appears to be the winner here. Like we'll get winners and then you'll get a couple of like other entrants that are sort of support them on the side. So I think this is the nature of experimentation. It's the nature of investment.

14:07And we should see more of these kinds of things. Have you ever wanted to trade Bitcoin but haven't dared try? With Plus500 Futures, you can trade crypto without the hassle of opening a wallet. With just a few clicks, you can register and start practicing with their free and unlimited demo. See a trading opportunity? You'll be able to trade it in just two clicks. Feel ready? You can move to real money with as little as$100 once your account is approved. And the great thing is that in addition to crypto, Plus500 gives you access to a wide range of instruments, S &P 500, NASDAQ, gas, and much more.

14:41Explore equity indices, energy, metals, Forex, and beyond. With a simple and intuitive platform, you could trade anytime, anywhere. Experience the fast, accessible futures trading you've been waiting for with Plus500. With over 20 years of experience, Plus500 is your gateway to the markets. Visit us.plus500.com to learn more. Trading in futures involves the risk of loss and is not suitable for everyone. Not all applicants will qualify. Plus 500. It's trading with a plus.

15:27it. As a meme, I think it's kind of cursed. And so like to go back to what Abby and I were saying earlier, right, like the reason we were talking about latent hardware versus, you know, devices that people already own versus having to acquire them is that's going back to that second point I mentioned at the beginning where I said that these things tend to work only if monitoring costs are low and capital costs are low. One way to make capital costs low is to make them zero because people already have the capital, right? So I do think within the spectrum of dPIN, stuff that is making capital costs low by focusing on latent infrastructure that people already have, that's something that we look favorably on and I look favorably on as a researcher for sure.

16:03But generally speaking, you still do have this other piece, which is the monitoring piece. And I think the monitoring piece, we have no general solution for it. The solutions are case by case. These are Oracle problems and then most of the time we're making them worse with these token incentives. And so you will see deep end projects which have clever solutions. they add frictions in some important or interesting way and that is cool right but generally speaking as a category there's just no answer to that question right it's each case and for that reason I think deep end as a category is a cursed category even though there are some good projects within it I disagree but that was a banger that was a banger that's a banger yeah okay so we have about five minutes left I think it's worth doing definitions for a bit I've had Bitcoin described to me as a deep end.

16:51I've had Visa described to me as a deep end. Let's talk like table stakes a little bit. What does the word mean to you guys? So we can like start defining the category a little bit more. Maybe Matt, you want to take this one to start? Sure. Well, look, the Bitcoin thing I think is interesting. I actually use that as a question to hire researchers is why does Bitcoin use useless work, right? I think one of the first things when you open up your little, what is a blockchain explainer and you read, oh, Bitcoin's wasting all this energy on these puzzles that don't do anything. You're like, well, what if we made that useful, right?

17:20Why don't we do that? Proof of useful work is one of the first things you think about when you learn about, you know, the way we architect consensus. You know, Deepin is trying to do something like proof of useful work. And it's an interesting question to figure out where that, why that's not a good idea or why that's a difficult idea or why, say, Satoshi didn't do it. And look, I said this is a question to hire researchers, but I'll give you, my take on it is, literally when you say useful, what you're trying to do is you're trying to introduce an externality, something that the protocol can't see, right?

17:47That's the Oracle problem that you're introducing literally there, right? The reason you do useless work is because you don't have that. If you do useful work, now you just have a thing that you don't understand at the protocol level, right? And so you have to have somebody sort of telling you about it, right? So I think that winds up being the definitional reason why it's different. So D-Pin, for that reason, I said the D should be a DAO. And I think the physical infrastructure network to combat this Oracle problem, right? I mean, it's ultimately just case by case, right? Ultimately, these things just are whatever people are telling it is in the real world as it lives on the blockchain.

18:20That's the pin part. And so I think it does resolve to just being a DAO if it's going to be decentralized. And so my movement would be to say physical infrastructure network, yeah, if you can solve the Oracle problem. Decentralized, yeah, if you can make it a DAO. These things are DAOs. Okay. Would you say solving the Oracle problem is a skill issue? Would you say that that's just a thing that we can resolve? I mean, I think it is a skill issue on a long enough time scale is maybe the best way that I think about it. I think like all of these networks are trying to actually to Matt's point earlier, trying to take some latent supply and apply it to a new problem.

18:57Not all of them, but yes, a lot of them. Sorry, the good ones. Yeah, the good ones. The good ones, maybe. But like, yeah, that is one of the things we're trying to solve. and in D-PIN specifically, we're trying to take off-the-shelf hardware and say that it's a secure device and try to build a root of trust around that and then eventually solve the Oracle problem through that root of trust. There are other D-PINs out there that are trying to solve that by putting an ECC chip down on the sensor level and maybe that solves the problem. But show me a secure piece of hardware and give me a little bit of time and I can show you how to break it.

19:33Like, so yeah, so of course it is a skill issue. It's also, and it's going to be sort of this constant battle to try to figure out how to solve this problem. And I think it only really is, I think it only really works if you sort of economically align it as well, right? Ultimately, like if you're economically aligned to Sybil something, you're probably going to do that first, right? Or at least some edge of your user base is going to do that. And I don't think this is necessarily a deep end problem. I think that's just a human scale problem. But back to my earlier point, I think that's what we should be building.

20:08Makes sense. All right, we have about 90 seconds left. I think, just to wrap things up, Matt, Abhay, you've taken different sides on this. I think looking forward, the question is, maybe Matt, what would you want to find that's adjacent or relevant or similar? What are you excited about if not for Deepens? What does the application layer, what What is the sort of layer of, like new set of economic coordination problems? What does that look like? Yeah. And if not for Deepin, how would you categorize, what would you call it? And for you, what do you want to see get built in this space going forward?

20:42What problems need to be solved? What categories? How do you want to, you know, continue evangelizing this movement? And we'll start with you, Matt. Yeah, so just quickly, I would say I really love the ambition and energy of Deepin projects. I mean, that's consistently the most exciting thing about them. So it's always very interesting to look and figure out if they do have these other things buttoned up because it's very cool to try and solve these real world issues. And look, I mean, the ambition of these things is fantastic. Aside from that, I'm really bullish in general. I think you asked this, right?

21:11I mean, I think, you know, stable coins and crypto AI and all kinds of other exciting things going on in the blockchain space. So I remain super excited about it. But Deepin in particular, I would say, is especially just case by case. It's one of those things where I'm always excited about the ambition. And then I always want to figure out, have you thought about these two, the monitoring and capital costing especially well? Sweet. As for something very practical and what I want to see more of, I want to see more people exploring the mobile space. I mean, I think that although that sort of talks against my own book, I guess, in some way, I think it's important to have more experimentation, more competition, new ideas of how to deploy these kinds of networks.

21:48Outside of the existing ones in mobile, and there's some hype around energy right now, which is kind of interesting. I mean, I think there's a lot of opportunity in manufacturing. You know, when I build hardware, for example, it takes such a long time to get, like, a prototype in my hands that I can play with. It would be really, really great if there were, you know, many companies like, you know, 3DOS is a good example of this, that are building, that you can just sort of, like, write a little, like, thing in AutoCAD and make a little spec and send it over, and the next morning you have the prototype in your hands.

22:19Like, that, these kinds of things. And I think that could be applied to other use cases as well, not just for physical things, but also for in the biomedical sciences space and more. So I'd love to see things like that in deep end. Amazing. I think we're at time. Thank you so much for taking the time to do this. Thank you, Sean. Great monitoring. Appreciate it. Thanks, Sean.

22:52Have you ever wanted to trade Bitcoin but haven't dared try? With Plus500 Futures, you can trade crypto without the hassle of opening a wallet. With just a few clicks, you can register and start practicing with their free and unlimited demo. See a trading opportunity? You'll be able to trade it in just two clicks. Feel ready? You can move to real money with as little as$100 once your account is approved. And the great thing is that in addition to crypto, Plus500 gives you access to a wide range of instruments. S &P 500, NASDAQ, gas, and much more. Explore equity indices, energy, metals, forex, and beyond.

23:25With a simple and intuitive platform, you could trade anytime, anywhere. Experience the fast, accessible futures trading you've been waiting for with Plus500. With over 20 years of experience, Plus500 is your gateway to the markets. Visit us.plus500.com to learn more. Trading in futures involves the risk of loss and is not suitable for everyone. Not all applicants will qualify. Plus500. It's trading with a plus. Thank you.