Balaji Srinivasan: Applications: Today & 2025

This week we're going to hear about applications and first speaker is going to be Balaji Srinivasan who's joining us remotely from Zoom. And he's going to talk to us about applications today and some that we think are exciting on the near term horizon. Okay, so with that I'm going to get straight into the presentation so we can turn it over to Babaji who's going to join us on the screen here. There he is. Hello. Hey, Babaji.

So my name is Balaji Srinivasan. I've got a new project@nakamoto.com, which you can check out basically like a crypto blog. But today I'm going to talk about crypto applications just to give a little overview of who I am. Just a rare photo of me in a suit like last time in a few years and my background. So at Stanford, grad is a PhD in electrical engineering, Ms. In chemical engineering. I founded a genomics company which ended up selling to Myriad for 375. His general partner at Anderson sold a crypto company called Earn.com to Coinbase and served as CTO. And I've basically been an early crypto person and venture investor and angel investor and so on. And I'm at Twitter Oligs. So if you have any questions about the talk, you can just DM me there or ask me. So happy to answer questions. Okay, so just jumping in, say I want to talk about kind of crypto applications in 2020 and then also beyond like things that are at scale in 2020 and then beyond that point, before we get to those applications, I want to talk through a little bit of the history of how we got to the present moment and some kind of underpinning concepts behind all crypto projects and then talk about the applications kind of with that mental framework. So first, history, let's talk briefly about why Bitcoin is invented in the first place. And this may be obvious to some of you, but for those for whom it's not, let's go through this quickly. The idea is if you start with physical cash, something you take for granted with physical cash is if I go and I hand Jesse a dollar bill, then I no longer have that dollar bill and Jesse has it. And we take for granted the fact that that dollar bill can't be in two places at the same time. It's a physical object. It can only be in one place at a time. So if we try to translate that naively into the digital realm and we take the serial numbers on that Federal Reserve note. And we say, hey, well, that's a unique identifier. Why don't we email that? Why don't I email that to Jesse? The issue is that I still have a copy of those serial numbers and I could go and email it to Kim or to Mark or to Ben. There isn't an inherent form of scarcity in the digital realm, unlike in the physical realm, where that dollar bill is actually only in one spot at a time. So naive digital cash doesn't work. This is called the so called double spending problem, because there isn't a native form of scarcity. And the way that we solved this prior to the invention of Bitcoin and still used very frequently today, is to have a central trusted party, a bank or a financial institution which takes a debit from A and then goes and credits B. By doing this, we've now introduced a third party into the equation and we're trusting this central actor, we're trusting them to debit A and to credit B. And with that trust comes power. Because this kind of institution can, for example, deny a payment, or it can credit itself with quantitative easing via central banking system. But from a computer science topology, it's generally inelegant to have a privileged node if you can avoid it. And so what Satoshi Nakamoto did was able to come up with a decentralized data structure, the blockchain, and essentially had something over here where basically miners. So rather than having a centralized actor, a bank that was debiting and crediting, Satoshi replaced it with a decentralized network of these miners who each compete to debit A and credit B. And if any one of them decides for a reason not to try to mine a transaction, the. The other ones will mine that transaction. Because anybody who mines the transactions gets rewarded in bitcoin for doing so. And so a critical mass of them would have to collude in order to stop A from transmitting money to B. And they're highly incentivized to not do that. And so this is kind of the magic of bitcoin is it replaced the centralized actor, a bank, with a bunch of competing entities called miners that carry out the same transaction approval processes and also the money generation process. Okay, so that's kind of why bitcoin was invented, is to give an analog of physical cache that solves the presence of a digital analog of digital cash that solves the presence of the central actor. Now, this entity over here, this blockchain, is something that got abstracted out. And basically once you've solved the problem of having a database of all the money in the world or all the Bitcoin in the world. You can do lots of other things. You can think of the blockchain as a tamper resistant database for storing things of value. Of course, you could use it to send or receive Bitcoin. With the development of Ethereum about five years after the launch of Bitcoin, this is a new blockchain that essentially was more programmable than Bitcoin. So writing a program in Bitcoin is like an assembly language like thing over here. But to use Ethereum solidity, it looks more like JavaScript. And actually it's kind of like JavaScript in the sense of it's easy to shoot yourself in the foot with, but nevertheless it was more accessible to write programs with. And why do you want to write programs? Well, so rather than just debit A and credit B, you might want to debit A and then credit B today and C tomorrow and D next Thursday and E if some condition is hit, and so on and so forth. That's what smart contract programming is. And so because you could write more sophisticated contracts in Ethereum, that led to the ICO boom. And the ICO boom in 2017, 2018 was kind of like the dot com boom, where startups raised millions, in some cases billions of dollars, and folks were able to raise $35 million in 30 seconds. All kinds of money flowed in. And like the dot com boom, there's a boom and bust. And many of those coins that got funded dropped like 90% or more. This though, was something where if you just look at it from a technological standpoint, the capital formation rails that were put in place. An ICO is like the simultaneous disruption of venture capital, swift crowdfunding and cap tables. Because you're doing startup investment, you're taking all these international wires, having large numbers of people do it at the same time. And you have something called a capitalization table or cap table, which is, which is the system of record which says who owns what stock. That kind of thing was also pulled into the blockchain with these ICOs. So it's actually very important technologically, even if not all of these companies are successful. Then after 2018, 2019 to 2020, now we're in the crypto winter in 2019, 2018, and now we're kind of maybe in the summer for innovation, where a lot of the decentralized finance stuff, which we'll get to, is really starting to work. You've got interest, you've got Loans, you've got all these applications and things are starting to work in crypto and it's kind of potentially laying the ground for maybe the next boom. All right, so that gets us to the present day. Now let's talk about some concepts and start with technological concepts. So one thing we touched on earlier is a blockchain is a database for storing things of value. So digital gold and cash, but, but also tokens, like in the ICO boom, digital assets as well. So in game items or real estate, these are called non fungible because one piece of real estate is not the same as another. Whereas $1 bill is fungible, it's equivalent to another. And then there's identity. So present proof of identity to demonstrate that you possess an asset. So cash equity alternatives, digital assets, identity. These are all things where you can kind of store them in a blockchain because they're valuable things with identity. For example, this is kind of a subroutine that was even required for Bitcoin or Ethereum in the first place. You had to present effectively digital signature to show that you owned Bitcoin or Ethereum before you could move it. And so this is being factored out as its own thing. Just like if you can do identity theft, well then identity is a valuable thing to steal. So that's the kind of thing that you might want to store in a blockchain in some fashion. So that's one way of thinking about blockchains. And you can also think of them from a different angle in terms of a protocol. So Bitcoin is a protocol and these crypto protocols have turned payments into packets. And I mean that in the most literal sense. Here is TCP transmission control protocol, here's ip, the Internet protocol, and Bitcoin runs on top of that. And you can actually see a block over here. And it's basically fully packet driven transmission of value without reference to a bank. Because in these bytes you don't see the USA or the Netherlands, you don't see Wells Fargo or hsbc, you don't see countries, you don't see bank, you just see bytes. It's really an Internet native way of sending money, which is cool because once we developed Internet native ways of sending books and music and songs and things like that, well, we got the Internet revolution. So it's pretty impressive. Third concept, once you've turned payments into packets, that means that machines can now hold and send money. So you can paste in a string of bytes over here and voila, now you can actually have a script that sends Money, right? And that's completely new. You can actually have a script that makes money for you when you sleep. This is like an intelligent agent that people have talked about for 20 years. It's actually there now. You can have these primitive intelligent agents that go around or more sophisticated actually, and do trades on Dexs and are essentially decentralized. Exchanges are essentially making money for you just by trading online. It's a script that actually holds the money itself intrinsically. Concept number four. One way of thinking about it is my friend Matt Huang has a great tweet, which is crypto is underrated from the outside and overrated from the inside. So how is it underrated from the outside? Well, in many ways, blockchains have actually already posted technological improvements 10xs in various areas. For example, 10x gold, 10x swift. So 10x gold. Why? Because Bitcoin is lighter, faster, cheaper to transport. It's 10x Swift. Ethereum clears in minutes in any country pair where Swift takes days. Swift being international wire transfers 10x larger than Kickstarter. So a large crowdfund in 2015 was like 10 or 15 million. But in 2017 it was like $4 billion, which is like 100x increase. And regardless of what those things are are that are being funded, simply the mechanism to move that much money and form that pool of capital is remarkable. And then in terms of like Delaware, right? You can hit enter in solidity in seconds, actually have it incorporated into the blockchain in minutes, and you can set up something that looks a lot like a corporation in the sense of being able to receive money, distribute it, have binding contracts that the blockchain enforces and whatnot. Essentially, this is much better than Delaware in many ways, because when people go to Delaware to incorporate a company that can take days in many ways, people talk about a measure of economic freedom being the time to incorporate. With smart contracts, you can 10x that too. This gives you a flavor. Blockchains have 10x many financial primitives, one 10x is a big deal, but a lot of them 10xing gold and wire transfers and VC and crowdfunding and cap tables and incorporation. That's actually a truly structural change and worth worth paying attention to, even if it hasn't rippled out fully yet. Additionally, the crypto space has actually already created many millionaires and a few billionaires. So the founders of coins, the miners who run large server farms, or the exchanges. This analogy is used frequently, but I would compare the present moment to roughly 2,000 in the Internet. Because in the 90s, there was this infrastructure phase. And you had Yahoo and you had AOL and people, everybody had heard of the Internet, but not everybody used the Internet in 2000 on a daily basis. And that's very similar to kind of where crypto is. Everybody in finance, everybody in tech has heard of blockchains. Most people have heard of bitcoin. Even if you're non finance or tech, the infrastructure phase is there. You've got the coins and the miners and exchanges. The applications are being developed, which we'll come to sixth concept. So blockchains give you a choice of who to trust. So before you had to store money at a bank, you didn't really have a choice. And afterwards you can still store it at a bank or at an exchange like Coinbase or even locally on your computer. And I think this is a better framework than saying trust lists. The goal is not to trust no one, it's about having a greater choice of who to trust with custody of valuable assets. So number seven, blockchains enable Internet scale cap tables. So before you'd have something like this, like a spreadsheet, basically, which said, what's a cap table? It's a capitalization table. It is essentially who owns what shares in the company. So Emily Wilson owns 2.618 million shares of common stock. Jake Pappas here owns 180k of series seed, A and B stock. This is what a cap table is. It can be said that a cap table is the most important data structure and technology because it underpins every company. This modest looking Excel type spreadsheet is who owns what percentage of the company. And right now that's basically held on Google Drive or maybe a Carta.com, which is an improvement. And then it's like this file is the basis for millions of dollars being made in ipo. That's how property rights are managed. A far superior alternative is what the blockchain offers, where you can have 50,000, 100,000 millions of people who are token holders and you can actually see every transaction back and forth. You're internationally tradable, transparent, liquid and so on around the clock from day zero. And the blockchain is keeping track of who holds what, rather than someone making a scrivener's error locally with an Excel file. So that's a pretty big deal, not just the automation of it, but the scaling of it from a few hundred people to tens of thousands or beyond. So I think of blockchain first as like the new mobile first, where blockchain competitors are rising for Many Legacy Businesses in P2P Payments and messaging and crowdfunding, all the payment type stuff, but also in other areas like for example, DNS, people wouldn't have thought that you could launch competitors to DNS, but you can. So handshake ens, things like this are actually competitors to the DNS system. And folks like Brave are using this to launch competitors to Chrome which have crypto built in. And you're seeing like a flowering of innovation across many different verticals with crypto powered competitors. And so that's why I'd say like blockchain first is kind of like the new mobile first. It's like a technology that gives you balances, it gives you encryption, it gives you a number of primitives to work with. So as a ninth concept, we think of blockchains as a tool that founders and engineers can use to break network effects. So, so if you take a network like Facebook, it's at scale billion users or what have you, it's got a present value over here. And the thing about that is if you launch a new network, well, it's not Facebook, no one's on it, so why would you use it? Well, now you've got a new tool which is you can issue your users tokens and those tokens have upside which declines as the network gets larger. So now you actually give a financial upside to early adopters and this curve can offset this one. So now you've given some value to people for joining early. That's a very important concept because this is kind of like giving equity out to your employees or selling equity to your investors, except you're issuing tokens to your users. And while tokens aren't equity, the upside of them is something where users can work and they can appreciate. Right. So this is really interesting and it's kind of turning your customers into investor like entities. And that's something I think will be a very big deal for breaking these network effects like the scaled social networks for two set of marketplaces. I think blockchains are going to transform social networks in General. Before last 10 years we've been liking and poking and messaging each other. And afterwards we've been doing paid messaging surveys, tasks, et cetera. I think this is a major transition from, from just making friends to ha ha, making money online. And I think that's important because it starts to actually increase the substance of what we're doing online. It's not just wasting time or communicating, it's not wasting time, but it's pure information goes to transactions as well. Finally just as an 11th point, I think of the blockchain as a partial move away from the cloud towards privacy. So Ledger sold its million hardware wallet for storing Bitcoin. This is something where anything that gets to a million units like this I think will probably eventually get to a billion. Might be bundled with phones. Like you have a phone, you have a charger, and you might get a hardware wallet alongside it. And basically what is a hardware wallet? It keeps your private keys locally, it protects them, and you could think of them as like an anchor that may lead to other data also being encrypted and moved locally, like out of the cloud and onto devices. And the way that would work is basically that your keys might be kept locally. Even if the bulk of the data is remote. It's only decrypted when you present your private keys. For applications like Dropbox or Gmail, that's actually very feasible to do for the most part because you can have a relatively dumb data store in the cloud and then present your private key to decrypt it, such the cloud provider can't see it and it's decrypted. For example, client side, there's many modalities that are like that. I think hardware wallets are going to be a big part of pulling some data and some privacy locally. Now let's talk about some concepts on the other side of things. So we talked on the tech side, let's talk on the community side. So basically building anything in crypto, you have to really understand the communities. Whether it's Bitcoin maximalists or Ethereum holders or the XRP army, these are folks who are. They have a significant part of their identity is bound up in holding this currency. It's actually kind of interesting. It's like the folks who live in the US value the dollar, folks who live in Japan value the yen. Their geographic location is actually more salient for their identity formation. But when you've got folks who are on the Internet, it's actually sort of like currency first as the formation of their identity. They're like Bitcoiners or Ethereum or what have you. And one way of thinking about what that does is if you just took three random people off the street, each of them could win or lose on their own. And so they have no inherent economic linkage. If you have N people, you have two to the end possible win loss combinations. With three people, you have two to the third or eight possible win loss combinations. However, if they all hold Bitcoin, well, suddenly they're economically aligned Such they all lose together or they all win together. And this is an important concept where basically if they're holders, none of them can win unless they all win. Blockchains take us from what I consider the slippery slope to the crypto cliff. Before you'd have a registry like the Spain and Catalonia fight or the CAT domain. Somebody could lose a LOL CAT domain, it could be shut down by the Spanish government and you might still keep your.com domain. So you don't really care if it happens to them. You might feel bad, but it's not really your problem. However, if someone's ENS domain is seized, that really means interfering with the ethereum blockchain. So it really is your problem as an ethereum holder. And I think that's a really important dynamic which turns us from the slippery slope where one individual's rights can be abrogated at a time to the crypto cliff, where to seize one person's crypto is to seize everybody's, right? That's a pretty big deal. So blockchains allow for experiments in self governance. And with blockchains, any sufficiently large group of people can now choose a monetary policy, choose an economic policy that suits them. And so I think we're seeing this really amazing thing where rule of law is being actually encoded. And one of the things this results in is that blockchains make macroeconomics and experimental science. So whatever your thesis on the economy, whether inflationary, deflationary, demarrage or what have you, before this had to be studied in the context of like video games with virtual points. Like Edward Castronova did a lot of pioneering work here. But now crypto has just taken that next level and you've got actual billion dollar or larger economies where you can just set the monetary policy and see what happens. So macroeconomics is finally becoming an experimental science because people are opting in and you can run these experiments at large scale. Very important to think about these things as we get into the applications that they'll come to. So tens of millions of people and billions of dollars are now invested in these experiments. The companies, the wallets, the percentage of folks, the market cap, these are things where lots of folks, the valuations are fairly high. They're not trillions, it's not hundreds of millions or billions using it, but it's definitely tens of millions. What that's resulting in is this concept of the laboratory of the states, that different states in the U.S. experiment with different legal regimes. Some will legalize medical marijuana and others will have different laws on right to try laws or things like that. This is like that, where you have different coins with different monetary policies and they're like a laboratory of the states. That brings us to applications. So everything I just said is useful context furnishing what kinds of applications you'd build with crypto, why was it invented in the first place, what has been done historically, what can be done with it, et cetera. Why don't we go into specific applications, starting with the ones that are already at scale as of 2020. So the first group are the ones I already mentioned. Exchanges, miners and issuers have made millions or billions of dollars. These are folks who are at a billion plus in revenue or in equity sales or what have you. These are like the infrastructure parts of crypto. The second group that is at reasonable scale are the hardware wallets. So ledgers at 100 million plus revenue at least in 2017, and that's about an order of magnitude less. But it's still, it's definitely rising. It's a curve to watch and will probably be built into kind of hardware devices over time. Third group that's already at scale in 2020 that's worth mention of is are stablecoins. So a stablecoin is like what it sounds like, unlike Bitcoin or Ethereum, it doesn't wobble, doesn't have high volatility. A stablecoin is flat pegged at 1.00 USD. And in theory it could be pegged to something else, but usually it means stable in USD value. There's now a few different stablecoins out there and they have different ways of doing that peg. Some will say they have bank account deposits of US dollars that will back the peg. And others will have some asset on hand, whether Ethereum or something else, which is volatile and oscillates where they say that the net value of that is enough to support the peg. So stablecoins are at significant traction. I mean, $5 billion nominal stored value is actually fairly high defi. So decentralized finance is also at reasonable scale. So since late 2018 it's really been soaring. It's now like a billion dollars in stored assets. There's a website, Defi Pulse, which I screenshot a little while ago, and all these things are just rising relatively quietly, but folks are doing well in the space. There's bugs, there's a serious issue like a few days ago, but in general the vector is up. Okay, so those are things that are at scale in 2020, which are exchanges, miners, issuance, hardware, wallets, stablecoins DeFi. Okay, so what's the stuff that is up and coming stuff that has green shoots in 2020 and might be big by 2020 25. This is going to be a long list of things and happy to answer questions. And I'm not saying every single one of these is successful, but some of them may be. Okay, so privacy coins. Oh, and by the way, just assume I'm an investor in everything on this and everything in crypto because I hold lots of privacy coins. I'm an investor in lots of these things, but trying to give a neutral kind of thing. So privacy coins. So-monero, Zcash. These are trying to add a layer of privacy over Bitcoin where essentially Bitcoin transactions are by their nature transparent in the blockchain. Dash Monero and zcash in particular are using different technologies to obscure that so people have actual financial privacy similar to the difference between HTTP versus HTTPs. Right number two, Lending and interest. So compound and maker. Compound is basically somewhere you can earn interest on the blockchain maker you can do lending. And it's actually this interesting system where you can put down some Ethereum and so long as the Ethereum price doesn't go down, you get some US dollar equivalent that you can go and do something with called dai dai. In terms of scaling, there's a ton of stuff that's happening. Lots of companies doing this, some of these are quasi applications in their own right, like the Cosmos SDK. But there's a bunch of different ways to try to scale blockchains. And you know, these technologies may stack on top of each other. You know, some of them are exclusive, but many of them are complementary. So together these will allow for higher throughput blockchains. And that's an important thing to think of as you're developing applications decentralized cold storage. So Casa Hodl. So basically they're gaining some traction. Interesting company where they're similar to Coinbase in some ways, but they basically resolve a paradox where thesis is people want to store at home. The antithesis is they don't know how to do that technologically it's like product wise, not feasible. The synthesis is you store at home, but you have a service that's helping you do that. So decentralized cold storage. Another application, starkware. This is SAS for gas. Basically smart contracts that are on chain that charge for each API call. Right now when you set up an API you have to set up a stripe billing layer and quite a lot of stuff. But it might be the kind of thing where just put a snippet of code there, almost like a lambda AWS lambda function, if you're familiar with that, put a snippet of code on chain and just set up billing and tada. You've got a function that can just execute and make you money. So that's what I call SAS for gas, another kind of thing. So insurance. So basically when smart contracts fail, in theory you can get a payout. Now many of these things are themselves written in smart contracts, so you don't know if that'll actually work. But it's certainly an interesting concept. Multi wallets, right? So this is something my crypto I think is one of the best of these people. But basically beyond just send and receive, there's buy, sell, sign, vote, stake, register and what have you. And these verbs, and by verb I mean action I think are going to come to all wallets. So security. So Xango is the first. It's an innovation on security which uses face ID for save and restore. And I think you're going to see this kind of innovation where it becomes easier and easier and crypto eventually becomes easier to use and harder to lose this than your traditional money over time. New financial instruments so FOMO3D and pulled together, these are two examples. FOMO3D was this interesting game theoretic thing where if you were the last person to deposit money, send money to this address, you'd get the pot. But if somebody else sent it, then they went ahead of you in the queue. And so the game theoretic thing is if nobody else is sending it, then you should send some money. But if other people are sending some money, then you shouldn't send some. So it's kind of unstable equilibrium. Eventually a guy won by just flooding the blockchain with transactions, had the ethereum blockchain with transactions and then putting their transaction in to take the rest of the money out, which is very clever. Pool Together is kind of a different model where it takes folks lottery like instincts where they want to quote, gamble money and instead what they do is they kind of give you lottery like payouts for saving money. So you could win 1644 every week just by saving your money. And it's clever because it's something that takes lottery like instincts and couples them to actually a pro social behavior of saving money, which is very clever. These are things that I think would be difficult to do in the traditional financial system, but that now we're starting to see versions of this and it's a very promising area. So blockchain Games another promising area. So Forte will be talking about this, but there's variants of this, some that have micropayments built in, some of them that merge esports and crypto, others that have in game potions or things like that. And so maybe it gives a new definition to pro gamer. You're just racking up the crypto by playing these games. Crypto social networks like Voice and Twitch are two of these. But there's a ton of them. These are still in their infancy. There's lots of different theories on how decentralized they need to be. Should they be decentralized at just the currency level? Should it be decentralized at the application level? We have to figure all this stuff out. But some of these are things where the application centralized the coin itself is decentralized and they just go with that. Then decentralized DNS, ens, unstoppable block stack, handshake. These are essentially domain names that you can programmatically register and then work with for and you don't have to go through DNS. And that itself is actually kind of useful where boom, you can deploy a website and actually get a custom domain just for that website. So actually fully deployed, which is actually better for testing than the current paradigm of staging and test domains. Automated market making. Now I'm actually kind of bearish on this, but it's worth mention things like Uniswap or the concept of bonding curves. They're trying to give liquidity to the long tail such that anybody can come up to effectively a smart contract and get a fair price from it back. The tricky part with this is there's various kinds of market manipulations in theory that you can do outside of that smart contract. And we'll see if these survive over the long term. I'm wishing them well and it's worth mentioning to understand the concept of an automated market maker. But we'll see if these work. Decentralized identity so three box is good. There's decentralized dot identity. But these are essentially folks who are trying to use Ethereum or other cryptocurrencies for sign on and not just payments. I think this is something where you might need to see the equivalent of a fiat exchange which lets you exchange fiat currency for cryptocurrency you might see something similar which lets you exchange fiat identity like driver's license or passport for crypto identity of some kind. Then personal tokenization. So Ruben Bruman often did something with cnsl. Spencer Dinwiddie has launched a token. So folks are issuing Tokens for their time or some function of your time like a consult. I think this is actually a pretty important and big thing. Mutuals and guilds so Moloch Dao and gitcoin. So these allow people to vote with their funds for like the direction of protocol development and they attempt to incentivize collective action. We'll see if these work. I think the fundamental issue is folks who are investing their money in something to simply improve the protocol. There's always a free rider problem where the person who doesn't invest the money can still benefit from improvements to the protocol. So we haven't yet solved that. One possibility is you're a corporate entity or you're a very interested individual and you want to steer the the protocol in this direction versus that direction. And so voting with your wallet for development is an interesting model like how Git Coin is working Founders rewards so just like scaling this is kind of a technology or business model thing you should be aware of as designing applications. So zcash pioneered this concept of founders award where people get a cut of the zcash mining and other folks are getting this as well bch and others or they're trying something like this. And this is a new business model for funding developers from rewards. And it's interesting because it gives a model for example for progressive decentralization. You can have a founder's reward which in the first few years and then it kind of draws off over time and after 10 years you don't get any more without the community re upping you. And that's kind of like a 10 year vest but for a decentralized protocol. And it's a clever way of making money for the founders for developing this thing but still having it be decentralized in many other ways. So on chain bounties. So Tezos basically had this concept where the developer bounty concept like gitcoin but putting it on chain which I think is a novel kind of concept clients for dapps so like instadapp for example makes it easy to interface with all these defi applications kind of like a coinbase layer on top developer tools. So Alchemy and bison trails improved APIs for GETH general performance improvements efficiency. If you're just sending tons and tons of transactions down to the blockchain you might want some of these as kind of an intermediary Oracle's and prediction markets so erasure augur also like chainlink and stuff like that. Basically right now you can only really bet on prices on the Ethereum blockchain Directly. But if other kinds of data feeds are put in there, then you can place bets on those. And Augur is live and folks are trying that out. Other folks are trying different kinds of oracles as well. Well, to place the data on chain so you can make bets on it. Daos. So this is worth a mention. Kind of. You know, the simplest version of this are the intelligent agents we mentioned earlier. But you could have these organizations with the A in daos, whether they're actually autonomous or what have you is tbd. But essentially it's something where trading money on these exchanges is kind of like pure Internet arbitrage. And I think you're going to see that with intelligent agents and you're going to see more besides that community owned organization. So these are collectives, like cooperatives that are owned by their users. And I think you're going to start seeing that become a lot easier to do with crypto. Like a group, it doesn't have to be like a normal corporation. It could be 1,000 folks in a group decide to set up something that's producing good for the members and that's kind of cool. It's like fusion of groups and capitalism and so on online. So that kind of brings us to the close. You know, I surveyed a ton of different kinds of applications there. The goal was just to give a sense of how broad the space is. And I'm happy to answer any questions or take them. Maybe. Jesse, you can, you can tell me.

Yeah, thanks, Bhaji. I'm going to try and actually join the Zoom meeting so that you can see folks during the Q and A and see if that works.

Oh, wow. Lots of people.

There's lots of people. Okay, cool. Well, thank you for that presentation. That was great. So, yeah, let's open it up to Q and A. Does anyone have questions? And in order for Balaji to hear you, we're gonna have to give you the mic. So hold your question till you get a mic.

Hi, my name's America, one of the Cyberco twins. I'm really interested in the trend of DeFi, but I saw something very curious where an artist token was. They were using her NFT to collateralize for a loan. Do you see a trend of collateralized gaming instead of just collateralized loans? Because I feel like one, you must have money to participate in DeFi, but if you have assets, maybe that's an entry to get into the space. Yeah, it's a great question. Fundamentally that depends upon the price for that asset. And the issue is that currently the way that maker or something like that is working is with Ethereum as collateral. And there's a large and robust and global liquid market for what one eth is worth. But if you have a painting or digital painting or something like that, you can use this collateral if and only if you have a large enough market to give a price on it. Now that might be just one person, but that person would have to have some kind of binding contract. They'll buy it back at a certain amount such that the person will extend you credit. I think it's possible. I think these markets have to become deep enough to provide pricing for non fungible assets.

Hi. In the past 15 years with Internet companies, a lot of value capture has come from either data monopolies or attention. And I'm curious, for the crypto applications.

Coming up over the next five, 10.

Years, do you see any large themes of where value capture is going to come from?

Well, that's a great question. So I'm actually very bullish on tasking in general. So like we did at Earn, Earn has gotten to several hundred million dollars in sales under Coinbase Earn, where all of these crypto issuers are. For example, if you're a new user to Coinbase, you can get like $180 in free crypto at Coinbase Earn. And the reason for that is all of these asset issuers are basically paying users to do tutorials and tasks and learn how their coin works, like send a private transaction with zcash or take out a loan with make or die and so on and so forth. So that's cool because it's the better than free economy. Rather than just try to hack your attention, people are actually paying you for it. And I think that's going to be a major area of tasking. And I think that's going to go from hundreds of millions of sales in the last few years to billions and eventually tens of billions. So that's kind of one area. And crypto uniquely enables that for a lot of reasons, but one that I think is really underappreciated. So if you think of Stripe, Stripe is a multibillion dollar company, like $30 billion company, because it improved over PayPal on pay ins, right? Setting up a form to just receive credit card payments people thought of as a solved problem. It was actually a huge pain with PayPal. And Stripe made that much better and therefore improved pay ins. But if pay ins on the Internet were hard, payouts are even harder. And one way of thinking about that is if you just think to yourselves okay, how many websites have you put a credit card into? It's probably dozens, if not hundreds. We say, how many websites have you put your bank account into to get paid? That's actually much smaller, right? It's probably like maximum 5 to 10, and usually not even that. Maybe if you're an Airbnb host, an Uber driver, maybe PayPal has it, Coinbase has it. But not too many websites have your bank account information, because a website with your bank information can debit rather than credit. So improving payouts on the Internet is something crypto does, since with Earn, for example, we could pay out 10,000 people by hitting enter. So that's why I'm very bullish on tasking and earning and so on in general. And I hope that actually helps make the Internet more positive some, because right now there's a lot of fighting, and it would be better if folks accumulated $10,000 rather than 10,000 likes, I think. So that's like one major area I could give others, but that's one. Hey, Balaji, thank you for the presentation.

I have a question for you with regards to social.

So I see sort of a lot of development on one side with, with Signal and with organizations like Matrix. But there's a version of the future where you'd imagine every email has an amount attached to it, or every tweet has a stake to it and you can filter based off of some kind of economic incentive. Do you think that future will come? Definitely, and here's why. I think over the last several years, different parts of email have been factored out, right? So emails between friends have gone to Facebook, emails within companies have gone to Slack, emails to strangers. I mean, they're substack, and that's kind of coming back. But there's Twitter and so on for that. So really, the primary use of email today is for two things. First, it's for things like receipts, password resets, that type of stuff, and what people call transactional email. And second, it's inter corporate email. So you are a sales rep contacting another person at a dot com, trying to get their attention. Now, I think that channel, because you're proposing a commercial transaction much of the time, is something where legitimately, you could put a dollar sign on the edge. Okay? So because it's a commercial transaction, you could put a dollar sign on the edge and it's not unreasonable. Whereas with a friend, paying them for their attention wouldn't seem right for a coworker, paying them for their attention may not seem right unless it's a very, very large company where maybe there's some bonus or something you're doing, but for a stranger who you're proposing a business relationship with, it's not an unreasonable thing to do. Moreover, if you think about email, your email inbox is a to do list that somebody else is writing. So if they want something to go to the top of your to do list, simply paying you to get to the top of the to do list is actually good. Now, of course there's tweaks to this. Many folks who are wealthy enough may not actually care about it from a money standpoint, but they might care about it from a signaling standpoint. And so they could just elect to donate all the money that they received to charity and maybe donate $50,000 a year to charity simply to sort messages by priority so that a stranger who really needs to reach them can. And so that's a way that you can use it for prioritization and still do good at the same time. And that's actually what a bunch of A6 and Z partners did with Earn.com, they donate to black girls code donate to worthy causes. People get to the top of their inbox. So I do think we're going to see another version of that. When deciding which apps to include on.

The 2025 list, what were some criteria you were using?

Was it mostly based on usage right.

Now or kind of from first principles, et cetera?

Oh, I think so. 2025. It's basically a grab bag of stuff I've been paying attention to. I was trying to be broad and pick a bunch of different things. Everything from hardware, wallets to privacy, coins, defi, new chains, et cetera, et cetera. Actually, probably what I under covered maybe is new chains, but those are almost sometimes the most covered. You can just scroll down CoinMarketCap and take a look at them. Basically I thought of them all as things that I could see some version of that concept working, if not the specific company. I'm pretty sure Crypto games, for example, are going to be a thing by 2025. People want to make money on esports. Crypto lets you monetize better. That seems like an obvious thing. I'm bullish on most of those trends, but it was just a grab bag of things I thought useful to mention. I could probably do another 20 slides frankly, but it was just kind of a selection.

Hey, I'm just wondering what you think crypto is going to do for the future of work and employment, because I know you mentioned tasking. Sounds kind of like you're thinking that employment will be sort of unbundled with bounties and other things like that. So how do you think about that in terms of the way people are going to actually be employed by companies in the future?

Yeah, it's a great question. So I mean, the thing is that, so I do think that there's going to be a huge scaling up of micro tasks for machine learning training. And the reason I say that is we haven't yet. So we've been using machine learning to train things like image recognition and stuff like that. What we haven't done is ask experienced doctors and lawyers and I mean there's folks who have trained, certainly medical imaging, machine learning, learning, but we haven't gotten all this information out of the brains of skilled people and eventually, for example, a bunch of lawyers annotate a contract as that language is non standard, this language is standard. A number of, let's say photographers say, oh, that photo is bad, this photo is good. There's actually a ton of stuff we still are going to be doing in terms of taking skilled information, digitizing it and pulling it in. And eventually what that does, it allows, you know, the doctor just focus on the really hard cases and automated diagnosis can take care of other things. Kozol has talked a lot about this, so I think there will be an avenue for that. Kind of like much, much more in the way of training data. But it's hard to fully, fully, fully factor out a job into only micro tasks because there's a lot of shared context. Right. It's something where I think you could make it work for programming first, where you'd have a GitHub repository, you have 500 open bugs, then you put 100 bucks on each of them and you put $50,000 out there and you let the sharks devour it. That is to say, on each of those, GitHub itself actually has GitHub bounties now, or bounties, it's called something else, GitHub rewards or what have you. And the concept would be you put $50,000 out there, 100 bucks on each ticket, and just let the piranhas, all the open source developers try and go and solve them and maybe that'll work. But you're also going to need some kind of approval process, so maybe you need another $50,000 to pay reviewers of those tasks and what have you. So I think it might work in programming first, it might work in books. I think we need some real innovations in kind of mapreduce for labor before that works. And that's non trivial. I think it's feasible, but non trivial. Hi, thanks for the talk. My name is Jeff. I guess my question is. So blockchain offers a lot of really interesting possibilities for identity and really in a very different way that than real world identity works today. And I was hoping maybe you can just expand on what you think the opportunities there are. Yeah. So I gave a talk on this called the Pseudonymous Economy. And the concept is that in the pseudonymous economy, people earn under one name, they speak under another name, and their real name is yet a third name. Right. So you treat your real name quote unquote, you know, and I put it in quotes, because a real name is a state assigned name. You know, you can't sequence somebody's genome and determine their real name. You can determine, you know, like their blood type and other kinds of things by, you know, from the blood. You cannot determine their real name. So if you treat your real name as a Social Security number and you might say, well, that's really atypical biology. Why would someone do that? Well, I can give you an environment where hundreds of millions of people do that every day, and that's Reddit. Hundreds of millions of people are effectively pseudonymous all day on Reddit and they just use whatever name suits them, they build up reputation under that name and they can switch other names and what have you. And like the one convention all these people on Reddit go by is that doxy doxing is bad, mapping the real name to the physical person is bad. And one of the reasons for this is kind of half joke about this, but I think by 2040 or thereabouts, doxxing will be considered one of the worst things you can do, because if you dox somebody, you get their address, you get their address, then you know who they are, you can cause harm to them in the physical world. So I think that you're going to see a huge burgeoning of pseudonymous identities, not just for communication, which we've already had over the last 20 years, but for earnings.