Injective USDC, IBCv2 expansion, and a community-focused turn
The following is a recap of the May 20, 2026 Cosmos Hub monthly community call, hosted as a Twitter Space and co-hosted by Zerk from Mad Scientists. This was the first community call, and they will run monthly going forward. A recap will also go up on the Cosmos Hub X account, and the space itself was recorded for anyone who would rather listen than read.
The intent of these calls is to share updates on what has happened in the Cosmos Hub and the surrounding ecosystem over the previous month, talk through one targeted topic per month, and surface questions from the community. The call sits inside a broader effort to be more proactive about communicating what is being worked on, so the community has more visibility into the Hub roadmap and a regular venue to weigh in on it.
Topic of the month: Injective USDC
The main topic this month was the announced transition of Cosmos’s canonical USDC from Noble to Injective.
Noble has been exploring an exit from the Cosmos ecosystem. While they have said they would maintain Noble USDC as a native primitive in Cosmos in perpetuity, the fact that this is no longer their core business model raises real concerns about ongoing security, maintenance, and the time and attention spent on such a critical primitive. That makes Noble USDC unlikely to be a sustainable long-term solution. Noble’s integration is custom, built on native Cosmos modules rather than EVM, which created enough complexity for Circle that Noble still does not have CCTP v2 support today. Eventually, when Circle winds down CCTP v1 in favor of v2, Noble will no longer be supported on CCTP, and the aim is to have everything migrated well before that point.
Once Noble’s exit became likely, Cosmos Labs immediately began exploring other options for USDC. Three were on the table: deploying USDC natively on the Cosmos Hub, cheaply importing USDC into the Hub via IBC and Eureka, and working with Injective, who were already preparing their own native USDC deployment. After weighing the trade-offs, the decision was to work with Injective and make Injective USDC the new canonical native USDC for Cosmos. The agreement that came out of that includes default support for Injective USDC via Skip Go, making it the primary supported USDC for IBC routing where possible, and direct support from Cosmos Labs for Cosmos chains that want to migrate from Noble USDC to Injective USDC as Noble USDC eventually phases out.
In exchange, Injective shares 50% of the USDC issuance incentives with ATOM, which will be used to buy back ATOM for the community pool. The exception is dYdX, where the split is 33% each between the Cosmos Hub, Injective, and dYdX.
It is worth being explicit about why USDC was not deployed natively on the Cosmos Hub. Circle’s quote for a native Hub deployment was in the range of three to five million dollars in year one, with an ongoing maintenance budget of a couple million dollars per year. Under those numbers, a native deployment would have been operating at a significant loss for the Hub every year. The Injective structure avoids that: Injective covers the deployment cost, and the Hub receives a share of the issuance incentives that makes the integration effectively immediately profitable.
For the rest of the Cosmos ecosystem, there is both a carrot and a stick for chains considering migrating to Injective USDC. The carrot is long-term support guaranteed by Injective for at least four years, improved routing through CCTP v2 (including auto-forwarding via Skip Go and the Skip API), and Hub alignment: Injective USDC is now Hub-aligned, so chains and ecosystem participants who want to support ATOM have a direct way to do so by routing through it. The stick is less something Cosmos Labs is wielding than a Circle timeline. Noble USDC will be discontinued by Circle on some timeline, and any chain that has not migrated by then will run into ongoing USDC support issues. The timeline is not imminent, but it is one-directional.
To make migration as smooth as possible, Cosmos Labs is spinning up a USDC migration working group to help individual teams with specific requirements.
Beyond the USDC mechanism itself, the Injective relationship is opening up further collaborations. Injective recently added stATOM support on Helix and Neptune (one of their lending protocols), which expands ATOM DeFi venues. Injective is also building new USDC-denominated products that are expected to be interesting for the ecosystem. The structural point is that incentive alignment between the Hub and Injective is now strong: growing USDC TVL on Injective in Cosmos directly grows the buyback flow to ATOM, so partnership work in that direction is mutually beneficial.
IBCv2 and Eureka expansion
A framing distinction first, because IBCv2 and Eureka often get conflated. IBCv2 is the IBC implementation developed by Cosmos Labs to connect Cosmos chains with non-Cosmos chains, including EVMs. IBC Eureka is a Cosmos Hub-specific product built on top of IBCv2, and is how the Hub today routes EVM assets such as wrapped BTC and other Bitcoin denominations into Cosmos.
The expansion roadmap moves IBCv2 to more chain types. EVM L2 support is the next step and is currently in audit (or audit remediation), expected in a few weeks. Solana follows shortly after. Both unlocks are major, and once they are live, the Hub’s own next move is to deploy Eureka connections to high-value chains as quickly as possible. Solana is a no-brainer to spin up given the current bridge surface. Osmosis has also requested a Eureka connection to Base, which is another strong opportunity. The medium-term aim is for the Cosmos Hub to become a competing provider for bridged assets between both Cosmos chains and non-Cosmos chains, and this is one of the largest growth areas in front of us right now.
On the technical side, there is still work ahead to make IBCv2 as permissionless to integrate as IBCv1, and that work is underway. The structural advantage of IBC is what makes this exciting at scale. CCIP is a useful comparison point: integrating CCIP typically means calling Chainlink, scheduling with their BD team, paying a meaningful integration fee, and waiting on their development timeline, usually three to four months for a connection. IBC does not require any of that. Any two chains can spin up a connection themselves within a couple of hours, without needing to involve Cosmos Labs at all. That difference is the core of the enterprise pitch.
Swift is the cleanest illustration of why this matters. Swift is currently setting up a blockchain ledger to pilot tokenized deposit transfers between member banks. Swift has 11,000 member banks. A model that requires bespoke point-to-point integrations negotiated through a single counterparty’s BD team is not viable at that scale. IBC’s permissionless and open standards are a much better fit for that shape of problem. The Hub’s role in that vision, longer term, is as a service-provider hub: a counterparty can deploy a single IBC connection to the Cosmos Hub and get instant connectivity to every chain the Hub is already connected to.
On reception, most enterprises in conversations with Cosmos Labs have never heard of IBC. They are usually more familiar with older blockchain infrastructure like Hyperledger Besu and are not familiar with the Cosmos stack. When the IBC model is explained: permissionless deployment, the ability to connect blockchains and off-chain servers, the limited integration cost, and the five-year track record without exploits the response is consistently positive.
The regulatory backdrop is doing real work here. Banks, fintechs, and other enterprises have always been aware of the cost savings blockchain rails could offer, but the regulatory and legal exposure had been too high to act on. With the current administration being more friendly to crypto-native rails, and with Clarity and other digital asset regulations starting to reflect that, the exposure has come down meaningfully, and enterprises are now more comfortable exploring the space.
A more community-focused turn
Since the Cosmos EVM on the Hub plans were scrapped in July, the strategy has been to aggressively pursue enterprise adoption, with the assumption that growth of the Hub would come downstream of that. The recognition now is that the enterprise sales side of the business takes longer to materialize than originally hoped. The community, the ecosystem, and builders cannot be asked to wait while institutional pipelines mature, so there are wins to pursue today that set up long-term institutional success while also delivering immediate value to the Hub and the people building on it. A lot of the IBCv2 work fits this frame, the native Hub DEX work fits it, and a number of community-facing programs being lined up fit it as well.
One near-term item is a hackathon, being put together with Mad Scientists, with details to be announced in the coming weeks. The theme leans heavily into vibe-coding. Tooling like Claude Code and Codex has changed enough in the last six months that the barrier to shipping a working product is meaningfully lower than it was even just 6 months ago. The encouragement to anyone who has not used these tools is to jump in now: a couple of weeks of hands-on time is enough to be competent by the time the hackathon opens.
A few other programs are taking shape alongside the hackathon. Cosmos Labs is considering a micro-grants style program for funding builder projects and subsidizing side projects on the Hub. Builder spotlights are on the list. x402 is a topic that is actively interesting and worth exploring further. Some community-led builder groups have started forming organically, and the direction is to engage with those rather than launch a parallel structure from the top down.
The DEX work is where the community-and-enterprise alignment shows up most clearly. The Hub will need an orderbook-style DEX to attract institutional order flow, since AMM-style venues are not a viable fit for those counterparties for both legal and complexity reasons. The structural question is how the community benefits from that order flow. One model under consideration is reminiscent of Neutron’s Duality DEX: the Hub’s orderbook would support plugins, and community-built DEXes could register against it as market makers. Any swap routing through the orderbook would pay a portion of fees back to the plugged-in DEXes. The design is intended to be open, so multiple DEXes on the Hub can plug in rather than only one. This is just one possible idea built on the principle that enterprise products built on the Hub should be designed with the community and open-source builders in mind.
One final point: Cosmos Labs does not want to operate as a voice from on high dictating how the developer ecosystem in Cosmos works. The direction is to listen to what builders want to build, what they need to build it, and how Cosmos Labs can help. The direct lines are open: @robomcgobo on both X and Telegram, and the Cosmos Hub forums are being checked regularly.
ATOM tokenomics update
A short update on the ATOM tokenomics work with Gauntlet came toward the end of the call. The work is still at an early stage but initial findings have been shared on the Cosmos Hub forums, and that weekly update series will continue as new findings come in. The broader framing is that the Hub is now growing its product suite as quickly as possible to bring in additional revenue and ATOM buybacks, and attention turns next to the supply side: how to make ATOM’s supply dynamics more efficient so that incoming revenue can be used as effectively as possible.
Depending on the timing of the June community call, the Gauntlet team may come up and present their phase-one work directly, since they are expected to be close to done by then. Gauntlet is also eager to share the work with the community themselves.
Next steps
The next monthly community call will take place in June. The hackathon details will be announced separately in the coming weeks. The USDC migration working group is being spun up to support Cosmos chains migrating from Noble USDC to Injective USDC. Mad Scientists is also working on a side project to transcribe these spaces and host them on a dedicated site for anyone who would rather read or scroll through timestamped sections than listen end-to-end.
Thanks to everyone who joined, and to Mad Scientists for co-hosting.