Circle builds its own blockchain. It’s not a rollup, it’s layer 1.
In doing so, American stubcoin issuers bet that the shortest path between the dollar and smart contract is a full stack, sovereign, and custom-built pass for payments.
The new chain, called ARC, is Ethereum compatible, but uses USDC for gas and has sub-second deterministic finality.
The big question for Ethereum is whether ARC will shift the activity of stubcoin, act like a hub-and-spoke architecture, or route liquidity, flowing through CCTP and other bridges into the broader EVM ecosystem, while internalizing issuance and settlement.
The decision to build a new sovereign chain rather than deploying it to modular layer 2 has rekindled well-known debate on encrypted Twitter. Does this mark a return to the enterprise chain, or is it just a logical architecture for a dormantian payment platform? Will it weaken Ethereum’s long-standing role as the payment layer of Stablecoins Crypto, or will it further anchor through cross-chain interoperability and network effects?
The answer may be less ideological than practice. Also, whether the circle pursues the Ark as a walled garden, or simply as a better bank.
Vertical integration
The circle declined to comment on the technical design of the ARC, citing quiet times. But what we know from the published light paper is that L1 is built on Malachite, a new high-performance BFT consensus engine developed by the original informal system as a rewrite of Tendelint. Circle hired a 9 core team behind Malachite, but the engine remains open source and is already being used elsewhere, including Starknet’s decentralisation roadmap.
Unofficial co-founder Ethan Buchman told Blockworks that he views malachite as similar to rust language.
For Buchman, Circle’s decision to launch its own chain is not a rejection from Ethereum, but a standard business choice for a well-resourced team. “Every time people choose to build their own chain, it’s the same decision. They value vertical integration,” Buchman said.
“The kind of predictability you gave you, and how it gives you the uncertainty of hedging. You have a team, so you don’t rely on an external team of core components,” he said.
Crypto Twitter/X is split up with Net Impact. Bullish-Eth Read (Ryan Sean Adams, Armani Ferrante) says EVM-First Corporate Chains are still supplying Ethereum’s network effects. Bearish-Eth Camp (Jon Charbonneau) claims that when flows internalize elsewhere, marginal EVM developers have become closer to ETH.
Others worry about censorship and developer gravity. Eli Ben Sasson fired a warning shot against a more open system – “Corporate L1S/L2S will not succeed” – Adam Cochrane tried to disqualify ARC as a “consortium chain” rather than L1.
The anti-narrative from the operator is that deterministic finality and protocol-level control are important for payments and forex. The ARC paper tilts as follows: “ARC transactions are unconfirmed or 100% final and irreversible.”
This is the heart of L1 against L2 as a business decision. Rollups “rental” Ethereum finality and data availability. Sovereign L1S owns the basic layer policy surface, including gas sects, KYC hooks, disputes/refund logic, and transaction orders.
Circle starts with the equivalent of the Cosmos chain using an authority of-authority using a closed set of balloters. “It’s very cheap to operate,” Barry Plunkett, co-founder of Interchain Lab, told BlockWorks.
For example, a COSMOS VALIDATOR cost of 5-10 costs thousands of dollars a month. “The L2 is probably a little cheaper, so you don’t need that many boxes and don’t have the overhead of paying a third party to operate for you, so you have the DA cost, so it’s probably washing,” Plunkett said.
Sovereignty is attractive if your profit margins exist in the speed of capital working with settlement economics. If your priorities are broad combinations in time to market, centralized anterior roll-ups can be cheaper and easier to ship.
Stablecoin boom as a background
A Keyrock/Bitso report today claims that Stablecoins is already restructuring their payment status. By February 2025, monthly Stablecoin payments had tripled to $6.3 billion, with B2B volume jumping from $120 million to $2.7 billion over two years. Card-based flows exceeded $1 billion per month. On track, the annual silly payments could approach $1 trillion by 2030.
According to Stripe’s engineering head, formerly known as the CTO of Alchemy Guillaume Poncin, it helps explain why companies with distribution are focusing on sovereign rails.
“We’re seeing several companies compete to build a Stablecoin-First blockchain,” Poncin told BlockWorks. “The income opportunities from owning a settlement class would be a dwarf of traditional payment processing margins.” In comparison, the cost of infrastructure is small.
Because Poncin similarly boils L1 vs L2 analysis with control issues and time to market. “Choosing between the two is a subtle trade-off. L1 can be fully customized on all layers. He is not very interested in fragmentation due to ARC’s EVM compatibility, because it “fasters adoption by both developers and users.”
As Crypto-Native Builders chase complexity and yield, Circle is playing another game. This is an infrastructure that can compete with payment giants like Visa, Stripe, and even Fednow.
Ethereum doesn’t go anywhere
Critics have assembled the arc as a sign for the circle, where USDC makes up almost 30% of the stubcoin supply at Ethereum.
But according to Buchman, it exaggerates the incident.
“I think Ethereum is here for a long time, regardless of today’s decision circle. I think it has achieved a kind of breakout speed. It has entrenched itself as a fundamental import technology to human rights… its lifespan is related to its neutrality and conservatism.”
The circle continues to need to work on that. Past concerns about the USDC issuer’s impact on Ethereum’s instructions have been suppressed in Buchman’s view by the spread of competing Stablecoin publishers.
“This is part of Ethereum’s evolution, and the idea that Ethereum only works when all the calculations occur is obscene and I think we’re missing out on the fundamental role that Ethereum will have for the next 100 years or so,” Buchman said.
The centrality of Ethereum as a destination for Stablecoin flows can last despite issuance and forex trading moving upstream to ARC. It is still unclear how many Defi apps Circle will put in court to deploy on ARC, but in general, the application layer wants to live in the deepest liquidity.
One legitimate concern is whether ARC’s USDC native design will introduce new vectors for censorship or elimination, particularly in licensed jurisdictions. If the USDC is both an asset and a gas, what happens when the token itself is frozen?
“Technically, a fallback mechanism is possible,” Buchman said, but the baseline should be the current banking system, and he believes that ARC will ensure that its low bar will be cleared.
“Is that just as much aiding freedom and protecting human rights as Ethereum or Bitcoin? No. I don’t think anyone is trying to make themselves a child about it. But it feels like the circle is trying to go to the bat for the open economy.”
There are still questions left to be touched on by Circle about system-level privacy and regulatory constraints. “We pitched them vigorously into privacy,” Buchman added.
