Speaker unverified: prediction

Transcript context

…yours and can be ported over and over and over again. So I think that's one of the big questions I'm interested in over the next decade: what is the split of custodial versus self-custodial accounts? So this is a very old timey analogy, but many people don't realize that Chase builds their own software, the big banks do. But if you bank with a credit union or a mid-size bank, they absolutely do not build their own banking software for the ledgering and managing the accounting, anything like that. And there's Fiserv, there's Jack Henry, there's First Data, there's a few companies like this who build the bank cores as they're known. And so the banks are actually a balance sheet and a credit strategy and a brand and various things, but in front of all this software that is provided by someone else. And so are you saying that your vision is that there's much more of that, where you can plug in the crypto equivalent of a bank? Again, this is like a total Trad Fi analogy, but you can plug in the crypto equivalent of a bank core into a neobank or maybe Uber and Lyft want to build this for their drivers or something like that. Is that basically your vision of where things go? That's I think the hope in many ways for where things should go or could go. I think it's very much an open question. It's a big part of where we feel at Privy that we have a responsibility to try to make sure that it's kind of an even playing field and there are good opportunities on both sides of the aisle. But yeah, I think that's exactly the point. I think the point would be the ledger is already public because it's on chain. The account is really cryptography, it's like private keys that people should be able to take with them and accordingly, you can move the banking core yourself as a consumer. And I think the Uber and Lyft analogy is at this point, a very tired crypto analogy. I'm pretty sure you've heard it, but the old crypto dream was what if Uber basically enabled you to have a different rating system based on where you were and a different pool of—the core network is shared, but then the actual app and delivery mechanism through which you have is something that you can build on top of much, much more easily. Yeah. Was that a woefully out of touch analogy for me showing just how— No, I've been trying to figure out how we explain self-custody because the only self-custodial asset is cash. And so it is very helpful actually to have more mental models for it. Okay, so we got one Guinness here. Go check out collect. Alright, this is just our regular NFC interface. So what payment method do you have there? I am paying with a Fuse card, which is backed by my stablecoin balance. It's a Visa card issued that is tethered to a balance on a USDC on Solana balance. Okay. So there is no bank account associated with this card. It is Visa backending to a Solana on chain balance. Exactly. Okay, let’s see if it works. Please be accepted. Sweet. You're good for it. How does the onchain transaction work? Does that happen in real time? Visa backending to a Solana on chain balance. Exactly. Okay, let’s see if it works. Please be accepted. Sweet. You're good for it. How does the onchain transaction work? Does that happen in real time? The stablecoins in the balance are then moved immediately to a smart contract and then at the end of the day, all the funds in that smart contract are then settled to the networks. So the balance has actually moved out of the wallet. So I've had $5.36 moved out of my wallet. But is there a real time, can you get race conditions here? These are the age old questions, like, can you overspend? You can double spend. Okay. Yeah, you can't overspend because it checks the balance in the wallet in real time, but you could have a double spend attack? If during the exact same moment two transactions are authorized at the exact same time on the same card from the same balance, then both would in theory be approved. So you're saying we need a blockchain built for payments. That's what I'm saying. And the account layer is something that we've had to build on the wallet side, which is actually the ability to freeze funds based on the first payment to prevent double spend. It's actually something that we've had to build out for some of our customers specifically for that reason. And you're doing this at the Privy level rather than the chain level and then you post that transaction to the chain? Exactly. That's cool. Again, you would see why you want this. So one mistake I think people make when thinking about crypto is they think about it as a discreet invention. Like, one day we had computers, but of course one day we had UNIVAC and then we had the Apple One, and then we had the Macintosh. And actually the computers got much better over time to the point of almost being different things. And I think similarly people think of blockchains happening at one point in time, but we tried for the payments use case, we tried raw Bitcoin, no lightning or anything back in the day in 2013, 2014. That was not a good payments blockchain, I'll tell you that. And despite the fact that the Bitcoin white paper really talked about payments as the core use case rather than many things that have really worked for Bitcoin. And so I'm curious, as you look at the last five to 10 years of blockchain advancements, just how would you guys describe them? We started building payments use cases on top of blockchains like two, three years ago. And it became very clear to us immediately that none of these things were optimized for this use case. And it was a bunch of micro decisions that probably made sense for different use cases that were being optimized for, but made it really hard for us to be successful. So one example is that on some blockchains, in order to make the address, make it possible for that address to accept USDC, you have to fund it and prime it so that it is available to accept that USDC and that might cost 30 cents. You can't send USDC to an address that does not have it? Yeah, that does not have GAS and has not been basically programmed to accept USDC. And what was the bottleneck? C and that might cost 30 cents. You can't send USDC to an address that does not have it? Yeah, that does not have GAS and has not been basically programmed to accept USDC. And what was the bottleneck? It was basically all of the transactions needed to be serially sent and confirmed through the blockchain. And what would end up happening is that there's a relatively high failure rate for these. So a decent number would fail. We'd have to capture them, chronicle them, and then resend them. And it's just because they're not picked up. No worse fate working on a blockchain than having to manually set the nons for the transaction. When you get to that place, it's a dark, dark time. And so we just realized time and time again that there's been blockchains that have been built for trading use cases and blockchains being built for storage use cases, but not many blockchains that were built for payments use cases, which have their own... It's hundreds of very small decisions that add up to a materially improved experience if you want to build payment infrastructure. Won't Solana people say Solana solves this? Like, it's the blockchain built for scalability. Solana is a great blockchain for a lot of use cases, but it's still not great for payments. Maybe a very geeky take on this. If you go through the history of very early Bitcoin,the reemergence of peer-to-peer computing from the late nineties early aughts where instead of these being volunteer networks like Tor or others, this is now we've built incentivization into P2P. So you have a means of incentivizing resource coordination globally through these networks. So that's step one. I think step two with Ethereum is we've gone from a single purpose chain, Bitcoin, to now actually a programmable chain. So this is akin to Von Neumann architecture, you move from fixed programs to you can now store the program separately from the compute and you can have the computer do anything. And so they are actually turning complete blockchains. And Ethereum is the world computer in that regard. And I think the phase we're in now is scalability.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.