Circle's Arc Explained: Building the Economic OS for the Internet

Alright. It has been absolutely massive week for ARC. We're so excited for this conversation. Gagan, how are you doing, man? Thanks for taking the time. Doing great. So relieved after the launch yesterday. Very excited.

It's, like, so much like that. She worked so hard to get us to there, and it's, like feel like, okay. It's live. It's very exciting, and thank you guys for the welcome here. I love the animation.

Let's go. I wanna ask, when when did this conversation start? Like, we we can see on our end. We, you know, we see the test net launch, but behind the scenes, like, how long have you actually been working on this and this been a core focus of your role at Circle?

You really wanna know? It's it's you know, it'll surprise you. We have so I've been at Circle for four and a half years, and we have talked about what is now ARC for almost four of those four and a half years. Wow. Yeah. So we like and and it's it's actually a you know, it's it's something which started when we we had, like, a lot of enterprise partners and institutions. And it started when we these partners would ask us, like, hey. We want to use USDC.

We want to use for various use cases, we want to use a certain stack. Which blockchain should we use? And how who are the validators? How do we go and talk to our compliance team that how are you going to, like how do you measure risk? And how do you how do we tell our CFO, like, which tokens to hold? So these were, like, questions that would come in over and over again, and we would be like we these are, like, starting to become, like, repeated problems that we are hearing, and we need to build a solution for this.

And so over time, the timing lined up. We solved a lot of complex problems all the way from, like, regulatory problems on how do you build a blockchain that is compliant but still has, like, distributed control, how do you have regulated stablecoins as payments instruments, and so on and so forth. So over time, a lot of those learnings went into, like, what you saw with our launch yesterday.

Wow. Yeah. It's almost like a did it feel like a own your own destiny type of moment where it was like, if we don't do this, somebody else is going in and you start to lose control. And control is kind of like a dirty word sometimes in the crypto DeFi space, but I think if we take a pragmatic approach to how you need to build regulated systems, like, somebody has to take control. Somebody has to have ownership. Somebody has to be regulated. And if something goes wrong, somebody has to be held accountable.

I I think it's what you're pointing out is, like, it's the way I would phrase it is, like, who can we solve the very specific customer needs that are unmet today? Right? And so if if no one else is doing it, we will we are happy to do it. And there are a set of use cases that work perfectly fine on Ethereum, and we still continue to support and invest in all the protocols that we have and running the running, like, USDC on Ethereum, running USDC on Solana, 30 other networks.

And there are specific set of problems that we don't see getting solved, and so we are, ARC is, helping solve all of those problems that are very, like, that are bottlenecks to further adoption. And we also are we have a view that, what has been done so far, like, the the emerging requirements from institutions and, for the agentic economy are gonna look different from what exists. So, we are happy to, like, you know, put forth ARC as a solution.

And a lot of learnings that have gone in the reason we phrase it as an economic OS for the Internet is because an operating system is comprised of, like, its builders, the variety of use cases. So the the variety of builders, the comprehensiveness of the use cases it can address, and the clean abstractions that it can offer to different kinds of developers. Right? Like, so the variety of builders, we have institutions.

We have, like, novel app developers building Internet native use cases, and all of them are on the same platform. Like, at launch, we had BlackRock, SBI, DTCC, Ave, Morpho, eToro, Extended, Formo, all these all these the best of Wall Street and the finest of Internet culture, Internet companies in the same room. Right? That is what you want. That's what, like, a good platform looks like. And then we have, like, very clean abstractions and so on.

Yeah. You know, you mentioned, and this is a question we often ask ourselves as we kind of have the perspective of working with different ecosystems and builders, and builders are always asking themselves like, okay, how do I approach the chain? Like you said, like what chain? I might know what wallet I want or I might know what specific partner or infrastructure I want.

Or maybe I'm super, super into a specific chain for whatever reason, but often what we want to unpack is what the use cases might look like or the, you said, bottlenecks that exist that Arc is unlocking. In your view, like, with without Arc, what was prevented and with Arc, what is now possible that may have not been possible or really straightforward before?

Yeah. For institutions who are looking to bring their use table coins, your entire risk team's questions on who is the validator and how is the network secured, instead of saying that it's a random set of nodes on the Internet, you can now say it as the likes of BlackRock, Standard Chartered, Visa, SBI, and Mastercard. These are, like, the validators. Galaxy. So these are the validators, and these are, like, renowned global institutions whom your your risk teams will be very familiar with. So that's, like, one.

The second bottleneck is, okay. Do you have to hold a native token, which fluctuates in value in order to transact and use the network? And you don't, you you need to pull USDC. So that is, like, dollar based accounting that every company is familiar with. And the third most important thing is is fast, finality. So there is no because it is a layer one and not a layer two, the transactions are settled in under a second.

So anything that needs to happen that waits on transactions getting settled, that is not blocked for forty eight hours or whenever the roll up happens, it is within a second.

Zach, I'm sure you you have some questions or things you wanna dive into, but just No. Go for thing I wanted to double click on was the risk, the first thing you said. Like, what what is the difference or comparing and contrasting from a risk standpoint? The obviously, the the the incredible companies that you guys have, BlackRock, DTCC, Galaxy, I mean, there's so many of them. No one would argue that, like, you've assembled an incredible list there.

What what is what what what historically was the situation, and what was risky in terms of the validator sets? Or if I'm a I'm not a risk person, and I've never been in that position, but if I were, what what were they struggling with historically there?

Was it, like, exposure of certain information? Or what does that look like? It's basically like it's, basically, it's like a, you know, risk of insolvency or whenever a transaction you are let's say you are you are making, like, millions of dollars worth of transactions, and you want to ensure that if you are if you have follow-up, like, lending or borrowing against money that is supposed to be in the bank, if the transaction gets reversed and you have lent against that money, then there is, like, insolvency risk.

So who's who's responsible for it, and who is providing the guarantees for it? That that question cannot have fuzzy answers. So that kind of risk is like it almost it it breaks the enterprise grade use cases right at the start. That's a fundamental question. So that is something which gets solved with, like, the institutions as validators.

And historically, my understanding is to become a validator for other chains, it was more about how much of that asset you held, how influential you were in the network versus, like, a corporate or, like, institutional approval process in terms of trust and credibility and things of that nature. Is that is that generally true, how that would work?

Yeah. That's generally true. So we have the the validators who are there. We so also side note, we announced we minted our token yesterday, which is gonna be the the token, the governance asset for the network. So all the validators are gonna also have the the rewards in our token. That happens at proof of stake time frame, which happens sometime next year. So there will be, like, mechanisms, governance mechanisms that are aligned through the our token.

But, again, it is something which is which boils down to, again, like, who is holding the tokens, and these are these are entities that you can you can bank on.

Circle is the first publicly I'm reading this here just from this tweet. Yep. Circle is the first publicly traded company to mint a network token for a new blockchain. What a technical technical milestone here. Zach, what are you

No. Yeah. I was just holding my tongue here. Was gonna ask, like, what why should people care about this? Like, why should we what is the impact going from proof of authority, proof of stake? Again, I think these are generally some terms that people may hear, and, their brains might shut off a little bit. But for like I said, there's I'm getting the sense from you that there is a that this does matter when it comes to validators.

This does matter when people are thinking about which networks do they wanna participate on an enterprise level.

Yeah. It does matter a lot. Right? Like, who is getting the tokens? Are they anonymous and, anonymous entities? Are these nodes running in North Korea? And is there a way for for you to validate that those nodes are not run by entities that you cannot trust? Are they gonna collude? Are they going to attack the network? How are you gonna answer these questions unless you know, like, who these entities are and what kind of guarantees you're gonna offer.

So this entire network whose validators are public, whose node operators are public, the token allocation and distribution based on who has the stake in the network is public. As an enterprise, you have a lot of trust in such a network. So any if you if you don't care about, like, your transactions being validated by nodes node operators anywhere in the world, sanctioned countries, of course, you shouldn't care about it.

But if you are an enterprise that is, that is operating regulated businesses, you will definitely need to care about, where which are who are the node operators and who are the validators of the network.

Wanna shift gears a little bit now, get into the nitty gritty of, like, architecture, development, tooling, and and more of unpacking this question of, like, who is ARC for? We've heard we we also and then the the I think the reasoning for me behind this question is seeing, from an outsider's perspective, this big push for agentic tooling. We've heard p people say, like, Sinclair from said something that stuck out, and I've heard reiterated at conferences that stablecoins aren't for humans. They're for agents.

And if we're taking that that that assumption to be true and that ARC is a chain built around, you know, one of these stablecoins, what tooling needs to exist, and then what do people build with these tools?

Yeah. So the reason we call Arc as economic OS for the Internet is because it is designed to fulfill various use cases of money. And the use cases begin with a payment that settles in your wallet. But once the payment settles in your wallet, that capital needs to be used in a variety of ways. So it needs to be used for earning. It needs to be used for lending or borrowing or trading novel assets or just holding and shopping around for for something.

So there are 100 plus partners that ARC launch with who fulfill all of these different use cases in their own way, which means that any builder who's wants to use their capital for a certain for a in in a certain way tailored to their needs, is able to pick and choose these partners and compose, like, however they want to invest their capital. For instance, if you are if you are a trader who likes to take more leverage, more risk, you can you can use the capital for purse trading.

If you want to just do a spot trading, you can use the decks on ARC to do a spot trading. And those two are, like, different ways to trade a same asset, get risk different risk exposure. And ARC has ARC and its partners are that we went live with, support both. We also went live with, like, Morpho and Away both, who support, like who are support, massive on chain lending markets. So we don't bookmark as, like, for it is designed for any one use case.

We think we truly think of it as an operating system that needs to support, multiple use cases of money. Because once money lands in a wallet, people use it in a variety of different ways.

How do you think people might be using money? I don't wanna say incorrectly, but maybe not to its full potential now with these toolings. Like, you know, we have all these Yeah. Building blocks. We have all these Legos, and, you know, we we get really excited to talk to sometimes consumer apps, not that they can't be b to b where, you know, they are trying to take these pieces and build something that feels new, feels vibrant. How do you hope to see some of these, building blocks come to life?

Oh, that's a that's a wonderful question. So, you know,

I think about this a lot that today, it's not like the the building blocks don't exist. It's that stitching them together personalized to your use case is hard. I'll I'll give a very simple example. Right? Like, every week, we hear about, different treasury rates that fluctuate or, different new, in savings savings rates that become available in on chain markets.

How does a user an end user discover that, oh, you know, if they were to shift their USDC from vault a to vault b, they are taking slightly higher risk, but they'll earn 2% more. Or how do they know that, you know, if they have high risk exposure to one asset and there's an earnings call coming, they might want to diversify before that. Right? So all these these are these are things that take a lot of, knowledge and proactive mental effort for, for a simple user to try to manage their money with.

And so most people end up making suboptimal decisions, on their investments. And what I'm hoping for, this is something I hope to see in the future, is that, the reasoning on how you think about like, you specify your goals. I want my I I have low risk tolerance, but I want to make sure that my savings continue to earn, as much as possible, check every single week, and rebalance my portfolio to make sure my savings are earning. That reasoning.

And then the the tools to make that shift happen or make the transaction happen now should it should become much easier to do both in the same place. So the stack that we are building with and you you can see the demos on our portal is where you can actually chat in your wallet and specify your goals, and you should have the agent surface the right application, right tool to be able to help you transact without a lot of complexity. So that that is how I hope.

I hope that, you know, builders build in the future unfolds, that your reasoning for your money becomes, like, so specific to your goals and you're able to fulfill your goals easily.

Drew, was hoping you were gonna bring up our portal here because I was, like, I Yep. You're not searching for it. I will be. So walk us through here. Yeah. I know we didn't really plan for this, but this is when things get really fun. Where somebody goes to the site. Where would you push them to first? Obviously, there's the main landing page. But, again, if we wanna start digging a little deeper, is it this agent tab here?

Yeah. So this is, like, this is what's publicly available. What we showed yesterday is a sneak peek. So what we envision over here is people to eventually create their agent wallet. Right? And when you create your agent wallet over here, you will get you will be able to set, like, some limits and policies with which you can delegate an agent to spend some money. And over time, as different developers build applications and link their agents in this, the user will have a chat window.

And that is something which, again, like, we showed we showed a sneak peek of yesterday. It's not live in the public site yet. That users will be able to say, like, hey. I want to predict, for instance. I'll take a fun prediction market example. I want to predict that Chelsea is gonna beat Brentford over the weekend, and I want to what are the odds? And they get the odds from market, and then they're able to say, okay. I want to, I want to spend, $10 predicting this.

And that prediction that is placed, like, right from the chat window.

So yeah. I was gonna say, is this, so this is giving this is giving your agents a wallet specifically to engage, interact with with with Arc. For a second there, I was I was almost reading this as, like, you're actually creating an agent out of the box that has a wallet, but it sounds like I I was gonna I'm off there, I'm glad I didn't go down that that that rabbit hole. What else? What else are we missing on the site? What should we travel down? It looks like there's some some test prompts here.

Sounds like I'll go play around with this this weekend. But let me open the aperture up to the the larger ARC ecosystem right now. What's available on day one? I see I there's a lot of people in the launch. You mentioned a few. Morpho, Arrow, LeFi, the list goes on. Who am I missing that we should we should call out here?

Yeah. You can, borrow so first, you can you can, if you're an enterprise, you can settling your payments on ARC. We have about 17 to 20 companies that are not listed here that are already making payments on ARC. These are the assets. I think if you go to the assets tab, you will see there are about 10 to 14 stablecoins that are there. So we have USDC, ERC, of course. But if you go further down, I think you should see our partner stablecoin list as well.

And so there's these partner local currency stablecoins are here to transact. And then you you can hold Bitcoin, so it is served BTC. And that you can hold Bitcoin the the ERC 20 version of the Bitcoin here as well. Trade against it. Morpho and Aave are the lending protocols that are supported. Extended eToro, social trading application eToro is there as well. So there's a there's a lot you can do. There's a lot to play around with. You can discover ARC also in, like, all the major wallets.

So in if you're using Phantom, MetaMask, Rainbow, any of the existing wallets, you should be able to use ARC from there itself.

That's awesome. Okay. So stepping back, big day yesterday. Sounds like you're taking a little bit of a breath today, duly duly deserved. When you think about the high level of where you go now so this is live. The the validators are there. The it seems like you have the right pieces in place structurally, technically.

Toward the end of the year, if we were to, like, you know, meet with you first day of twenty twenty seven, like, what would a successful q four look like for you personally on the product and technical side? Like, what's the what's the priority for you right now as you end the year?

Yeah. As we end the year, there are three different areas of focus that on the ARC team we are, like, working on. Number one is bringing more and more partners alive and giving users access to easier and easier ways to access, like, common use cases like pay, borrow, lend, trade. And in service of that, there are there's, like, liquidity provisioning. There is there's, like, setting up protocols, setting up pools.

So, basically, like, making ARC fully operational with with liquidity across different assets and different use cases. That's, like, number one. The second is progressing our agent stack. So there is there's there's a very interesting area of work where how do you ensure that you can trust agents with your personal data and with data that you don't want to be open on the Internet, but you want the intelligence. So having, like, auditable work for from agents and auditable proofs from agents published.

So that is something which we are working on as well. And then third is, like, progressing ARC towards a full proof of stake launch, onboarding more validators, launching the ARC token, making it publicly available, progressing our towards that. These are, like, the three main areas we are focused on over the next six to eight months.

It's awesome. Zach, any final thoughts from your side?

No. Because you kinda hit it. I was gonna do okay. I always love the the call to builders from, like, a product perspective because I think, you know, we talk to a lot of people who you're building the tooling for builders, but then you get ideas to build stuff that you it's very rare that you're gonna peel off from the new builders. Builders yourself.

Builders, I'm sure. I'm excited to see what builders do here. It seemed pretty pretty early, but just, like, a lot of different building blocks for

bringing these things to life. So Yeah. Keep us posted on this. Yeah. I can touch on this for a second. This is this is actually really cool. So one of the problems that this is actually live, you can you can use this today. Arc Studio, one of the problem builders have faced is they have had to, like, reinvent the same primitives on blockchains over and over again. And to go from an idea to an on chain application with smart contracts that have that are secure has been hard.

Art Studio is like our coding agent for building smart contracts from ideas, and it also has security scanning baked in. So it avoids, like, common pitfalls that developers face while authoring smart contracts. And it's something that you can try today. It's also available via CLI, you can link it up to your codex and cloud as well.

And so we are very excited to offer this because we want to have the have developers, like, build new ideas much faster and experiment with new ideas much faster and be able to do it, like, you know, without having to, like, learn the the deepest details of, like, how to author smart secure smart contracts. So we're to bring all the know how that we have into this this agentic coding framework.

Yeah. I I love the just this desire to push the boundaries, and this is kind of not a super arc related example, but I did see this on X the other day. I don't know if you've ever seen Megapod. They're on base. They are an on chain lottery, Raised 5,000,000 from, from Dragonfly and a few other people in the pre seed.

And they've gone more and more down the route of having an API, helping support teams, incentivize deposits, really just, you know, again, not what you typically think of when you think of the on chain economy. And somebody went out, launched a meme coin. The meme coin is they use the trading fees to buy lottery tickets, and then the lottery tickets, if they win, goes to buying back the token and message you burning it. And, they won, like, 200 thou they won the $200,000 lottery. And, it's novel.

It's not for anybody. But, again, for somebody to think to to think through and think some of these novel ideas, maybe they hit, maybe they flop. You kind of don't know until you build it. And, so I really admire that. Again, now we have the tools to build it with the studio for people that may not be super, you know, technology savvy, but have ideas for, really building mini mini economies. I think about the games that could be built, these experiences, the, you know, the personal finance apps.

Yeah. Absolutely. It should be, like, much easier to build apps, like financial applications, applications that have, like you said, an economy than it is today. And we are continuing to invest in abstractions. Again, going back to why we think of the problem as an operating system problem is because an operating system abstracts complexity, surfaces the right primitives, so you can build and iterate much faster. So that's exactly what we are focused on.

It's amazing. Alright. We're sending you off on some ARC fireworks.

Thank you for the time. I like this new send off that you've been doing, Drew, with Yeah. It's like This is super cool, guys. It's great. Gaggan, thanks for the time, man. Thank you so much again. Have a rest. Sleep well. Yep. We need it. Need it. Alright. Yeah, man. Bye. See you later.