Defunding CosmWasm is a Threat to the Cosmos Hub and Cosmos Ecosystem

Conversation
1

Today, the Interchain Foundation (ICF) announced a decision that will significantly impact the Cosmos ecosystem: they will be defunding CosmWasm. This is a huge blow to the infrastructure that powers almost the entire ecosystem and facilitates seamless cross-chain interoperability.

At Atom Accelerator DAO (AADAO), we firmly believe that CosmWasm is a critical resource for the interchain, and its development must continue. We refuse to let this vital project falter. That’s why we’re stepping up – proposing to contribute $250K to kickstart the funding effort and motivate other chains to participate.

The Impact of CosmWasm

CosmWasm has become a cornerstone of the Cosmos ecosystem, powering chains like Osmosis, Injective, Mantra, Neutron, and most of the 100+ chains in the ecosystem. With over 1.79 million downloads of the CosmWasm standard library, the framework’s reach and utility are undeniable.

Analyzing data from just 15 chains over the past 12 months, 15.7k CosmWasm contracts were deployed, with 5.2m accounts interacting with them, and $5.7 million generated in fees from those contracts!

CosmWasm is a game-changer for the Cosmos ecosystem as it empowers developers to build and deploy innovative dApps across interconnected blockchains. Its ability to connect developers to a thriving interchain ecosystem via IBC and its modular, flexible, and secure framework makes it a key player in the blockchain world.

It’s no exaggeration to say that CosmWasm enables the complex, cross-chain applications - from DAO DAO to CalcFinance - that define the future of decentralized finance and beyond.

Why CosmWasm Deserves Continued Funding

CosmWasm isn’t just another smart contract platform — it’s a solution that addresses the unique needs of the Cosmos ecosystem. Focusing on security, scalability, and flexibility eliminates common vulnerabilities while enabling high-performance development.

And let’s face it - we need more apps, not appchains, and it is much simpler to build apps on CosmWasm, than a full-on Cosmos SDK appchain. The ecosystem’s growth is tightly tied to CosmWasm’s continued success.

Key to its design is cross-chain interoperability, which enables the Cosmos ecosystem to be a leader in decentralized, multichain applications. With tight integration with the Cosmos SDK and IBC, CosmWasm’s modular design supports complex, cross-chain applications across ecosystems.

A Coalition of Chains to Fund CosmWasm

The future of the Cosmos ecosystem is at stake.

CosmWasm must be funded — and we can’t rely on a single entity to shoulder that burden. AADAO is committed to supporting a coalition of chains that will step in and provide the funding needed to keep CosmWasm alive and thriving.

On behalf of the Cosmos Hub, AADAO is contributing $250K to help secure CosmWasm’s future. We are joining Neutron, who kickstarted this initiative with an initial commitment of $250k.

This $500k is a good start but not enough to fund CosmWasm for the year as the framework consists of modules, libraries, tools, and documentation across 20 different repositories.

We are hoping this is just the beginning. We are calling on other chains, developers, and ecosystem stakeholders to join us in this effort.

A platform that allowed 15.7k apps to be built in a year and generated $5.7 million in fees for those apps should NOT be allowed to die by the wayside.

Now is the time to act. Join us in keeping a core pillar of the interchain alive.

6 Likes
2

Please keep us in touch about which chains do support this coalition

2 Likes
3

Hi folks, I think its worth clarifying the position of Interchain as laid out in the blog post today! The ICF is reducing its spending on CosmWasm, but it’s deeply committed to supporting long-term maintenance, improved security, and continued development of the project.

The popularity of CosmWasm within the Cosmos ecosystem is undeniable (rust smart contracts are awesome!), and its used by many chains today (as OP writes). But we don’t consider it a critical driver for growth of the ecosystem, or on the Cosmos Hub, over the next year, based on conversations with teams.

  • The tooling and developer ecosystem around it is still very nascent compared to other VMs, especially the EVM. And the # of developers who are familiar with it is still very low compared to other VMs.
  • Meanwhile, we’ve heard that simply adding support for other popular VMs (the EVM in particular but also SVM and Move) will drive integrations, liquidity, and devs for the ecosystem almost immediately (Many teams want to deploy liquidity into the ecosystem but literally can’t justify the expense of re-implementing their infrastructure / contracts for Cosmwasm with the current opportunity size).

So improving support for multiple, popular VMs will immediately enable market makers, custodians, interop projects, defi teams, and others to start experimenting and deploying in the ecosystem – and that liquidity will flow into CosmWasm over IBC. We need to make that happen.

For years, the ICF has been the sole funder of CosmWasm – spending multiple millions of dollars per year to fund its development and maintenance. But with our new focus on growth and the evidence that 10x or 100x growth requires a multi-VM approach, our recommendation to the ICF was that spending shouldn’t continue at that level. Cosmos needs many VMs.

Having said all that, the ICF is still committing $700k+ in funding to Confio for CosmWasm for the first half of 2025 – primarily for maintenance, security issue triaging, and support for IBC Eureka. In addition, we’re recommending the ICF fund comprehensive audits of Cosmwasm, wasmd, and wasmvm in 2025 – since we’ve heard from dozens of teams that the primary reason they make CosmWasm permissioned or don’t use it is concerns about its security.

We agree completely with the AADAO that a single entity shouldn’t fund CosmWasm at this time. It’s our hope that a collection of chains and entities – including Interchain – will still be able to come together to provide funding for maintenance and security of CosmWasm in the long term. We’re actively working with AADAO, Neutron, Confio and others to source that required funding collectively and contribute to it ourselves.

Of course, it’s still possible that, as an ecosystem, we won’t be able to raise the required funds for a dedicated team (e.g. Confio) to support CosmWasm – even with participation from ICF, AADAO, and Neutron. In this case, Interchain will make plans to put together a small internal team to do required security/maintenance work. We prefer to support and participate in ecosystem funding of an external team for now, rather than attempt to internalize development, since it would take us time to develop expertise internally (compared to a team like Confio) and our focus would very narrow (just be security & maintenance, rather than feature development & performance).

16 Likes
4

I imagine that if the cosmos hub community agrees to fund cosmwasm (something that the hub does not use and does not really benefit from) it would mean that there would be a double spending of atoms since Interchain Inc. would fund this at its level and that it essentially holds atoms. For me it is an assumed no, no return for the hub, then no expenses for the others if it is only one way as in the past. The day when the other cosmos chains will pay maintenance fees to the hub if it is intended to fund the cosmos technical stack then it will be a big yes. for the moment I call it theft.

5

triple spending because AADAO will pay with their atoms. lol

6

CosmWasm is undoubtedly a cornerstone of the Cosmos ecosystem today, but what will happen when we open the doors to developers using EVM, SVM, or even Move VM? Two outcomes are likely. On the one hand, CosmWasm’s relative share will decrease. However, these new developers will discover the cosmos ecosystem leading to a likely increase in the overall number of developers. On the other hand, it is highly probable that integrating modules like EVM or SVM will be less suited to CometBFT and the Cosmos SDK compared to CosmWasm today. In other words, new developers who are genuinely drawn to the cosmos ecosystem may quickly learn to use cosmwasm to fully optimize their projects which would tend towards an increase in the number of developers on cosmwasm and which would justify an investment.

From the Hub’s perspective, even though it doesn’t use CosmWasm directly except when enabled by governance, Neutron does rely on it and needs the necessary support to remain the leading permissionless CosmWasm platform. If the Hub considers Neutron an economic partner, it will need to support it in one way or another. In this context, investing more in CosmWasm could be a good strategy, even if I understand the current reservations.

3 Likes
7

Wasmvm isn’t safe as is, in any form, permissioned or not.

Note that I could have made a private report of this issue, except:

Fortunately, Simon is speaking to me again and if there were a security issue, I could report it privately.

Unfortunately, the intention is still to go with a cgo integration. Cgo is a plague, as is the extensive use of unsafe in both go and rust present in wasmvm.

So if the hub were to choose to fund confio, I think that the only right move is to fund a drop in replacement for the plagued code in the internal/API folder of wasmvm. It should just be go.

Here’s an article by cockroachdb, which makes distributed databases about their experience with cgo

They were never able to make things reliable, without the rewrite that they did. It was difficult and costly but it got done.

I’m not sure that replacing wasmer with wazero will even be particularly difficult or costly, but I have hit a few bitches in doing it.

I am doing that currently for $0 but of course my time is limited due to limited funding.

1 Like
8

Fyi, the hub does use cosmwasm. It has a live contract, daodao and is therefore at risk.

1 Like
9

only that, to low for any founding imo

1 Like
10

I definitely understand your point of view.

11

Hey Barry,

Montagu from AADAO here.
Thanks for providing further clarification and the rationale behind Interchain’s position!

Exactly, this aligns with our primary concern, especially when considering that the Hub itself and one of its leading Consumer Chains both leverage CosmWasm. However, it is encouraging to hear that Interchain has developed a contingency plan should funding challenges arise.

It’s also worth mentioning that we are fully aligned here. In fact, we have been lately working on similar initiatives and look forward to sharing more informations with the community later this week.

3 Likes
12

:fire: :fire: :fire: Agree with this completely.

But CW needs to be maintained until such a time that most of its current users have migrated away from it (assuming at least a 2-year timeline for the likes of Osmo, Injective, and Stargaze to reimplement everything). Someone should also hit up Thorchain, who I think just recently began working with CW to bring them into “Cosmos-proper”.

While the ICF has been the primary (not sole) funder of CW, given how slowly the ecosystem seems to be dragging its feet, IMO, CW should just switch to BSL (to force action) or we will likely need the in-house team at Interchain Inc to take over before the end of 2025.

2 Likes
13

To be clear, there is an underlying grant application for the Confio disbursement.

2 Likes
14

Thank you @ATOMAcceleratorDAO for the support and bring up the discussion. Thank you @bpiv400 for the detailed response in public explaining motivations and strategies. I respect all of those opinions and decisions.

Just one correction. Credit where credit is due. Between 2020 and 2024 significant (>=100k$) contributions to the development of the FOSS stack of CosmWasm were made by (in alphabetic order):

  • Cosmos Hub
  • Juno
  • Neutron
  • NYM
  • Osmosis
  • Terra / TFL
  • Tgrade

Smaller contributions came in through Comdex, cyber~Congress, Crypto.com, Fetch.ai, Kujira, Persistence, Provenance and Stargaze. Those include a variety of different models such as audit sponsorships, donations and Subscriptions. That being said, it is correct that the ICF is historically the primary funder of CosmWasm development and the main funder by far right now.

However, that is not supposed to affect the big lines in the discussion which I appreciate.

2 Likes
15

I strongly support the Skip team taking bold steps to improve Cosmos without being beholden to inertia and existing power structures, and this looks like just the kind of bold move I was hoping to see. Personally, I’ve written contracts in both EVM and CosmWasm (Gravity Bridge and Hydro), and I don’t have a strong preference. I actually really enjoy both of them. However, I do have some questions for @bpiv400 and the rest of the team:

  • You’ve presented the concept of Interchain putting together a small team in house for maintenance as sort of a worst case scenario- to me this sounds very attractive, and like it’s the right step. Why is this not the default option, with the community maybe funding advanced features or something?
  • CosmWasm is very tightly integrated into Cosmos. Tokens, IBC, accounts and keys (Keplr etc), frontend, calling into other SDK modules- it all works pretty seamlessly from CosmWasm. CosmWasm is so well integrated that teams such as Stargaze and Osmosis have been able to build a lot of their core chain features with it instead of Cosmos-SDK. On the contrary, the experience with EVM (not sure about SVM) has historically been a lot less smooth, with mismatches between address types, decimals, and other stuff in wallets, etc. Do you foresee the EVM and SVM getting a lot more integrated with Cosmos in the future (or is it already and I just don’t realize it yet), or do you think it doesn’t matter?
  • It doesn’t seem like having a different VM caused issues for Solana, or the Move community. Terra did pretty well with CosmWasm. On the other hand, I can’t think of an instance where a really iconic platform used the VM of another platform. Users of the EVM tend to be tightly associated with Ethereum itself, such as Polygon, and all the rollups. Do you think this is just an accident of history and that Cosmos and the Hub will be the first platform to break out with a VM agnostic approach, or do you think there’s a risk that dividing effort will result in Cosmos not knocking it out of the park with any of these VMs?
2 Likes
16

Hi - Adair here from Abstract, building interchain application infrastructure and tooling in CosmWasm for the last 3 years.

Thank you @ATOMAcceleratorDAO for the post highlighting CosmWasm’s strengths and thank you @bpiv400 for your responses. I’d like to respond to a few points you brought up:

Claiming the CosmWasm developer ecosystem is underdeveloped is incorrect. The tooling is on par or superior to that of other VMs, including the EVM.

(No links as they’re disallowed)

  • Scripting and Testing (like EVM Forge): cw-orchestrator: Environment-agnostic scripting (sync/async Rust), unit & integration testing, IBC abstractions, a CLI, live chain-state clone testing, and snapshot testing. Every script supports live environments.

  • Indexing: dao-dao-indexer, with incredibly fine-grained event handling

  • Tracing (like Tenderly): CosmWasm Tracing Tool

  • Frameworks: Sylvia, Abstract SDK, Fadroma

  • Interchain Infrastructure: Polytone, Abstract Module IBC

I worked in EVM-land for 2 years before finding CosmWasm and have done EVM development recently as well - the CosmWasm developer experience is significantly better when you know the tools.

Natively Supporting Other VMs Will Not Drive Liquidity

What benefit does in-housing alt-VM Cosmos SDK inclusion have for the Interchain as a whole? There has been demand for these other VMs, and this demand has been fulfilled by teams like Movement Labs, Evmos, Cantos (ha), Injective, and any number of teams that have alt-VM implementations on top of the Cosmos SDK. Builders that “would enter Cosmos if these were available” would already be here and probably are.

Supporting other VMs fragments the ICF’s focus and is something that would be far better done by external teams, and already is. Being “pretty good” at many VMs is not a strong sell.

And as jtremback points out:

No iconic platform has succeeded with a VM-agnostic approach. Prioritizing external ecosystems fragments the ICF’s focus, when CosmWasm already knocks it out of the park with its:

  • Seamless Cosmos SDK integration.
  • Native IBC capabilities, which no other VM has.

CosmWasm Is the Key to the Interchain Thesis

If the interchain vision is to succeed, CosmWasm is the key to making it happen:

• Native IBC Integration: CosmWasm isn’t just for smart contracts; it’s an orchestration layer. It maintains the ability to program logic locally, remotely, in synchronous and asynchronous contexts.

• EVM IBC: The upcoming IBC-Eureka upgrade and Union will connect CosmWasm with EVM, allowing chains to orchestrate liquidity and actions across ecosystems. THIS is what the Interchain and CosmWasm has been waiting for for years, and now is not the time to stop funding when developers will just start to be able to build remote EVM applications. Also we want to note that Abstract has fully programmable EVM Interchain Accounts end-to-end tested and shipping with Union.

The reason we haven’t seen many interchain applications yet is not a limitation of CosmWasm—it’s because:

  1. The tooling wasn’t ready (it is now).
  2. Developers haven’t yet embraced “interchain” thinking, because “interchain” up until Eureka + Union has only meant “Cosmos.”

If this decision is set in stone, CosmWasm must be maintained for at least 2 more years. Anything less is irresponsible.

CosmWasm is the most versatile, and only interchain-capable VM, and cutting it now would be a catastrophic misstep for Cosmos’ future.

7 Likes
17

I was reluctant to post here, but I feel like there’s an angle which could use a little more emphasis in the overall narrative and vision: large, innovative, real-world applications.

Some examples:

  • Simple but fun open gaming engines (Chess, Monopoly, Go, etc.)
  • More reliable recordkeeping (healthcare, business meetings, etc.)
  • eCommerce for tangible products with physical shipping
  • Supply chain verification all the way down to the end user
  • Voting systems and other tools needed for communities including but beyond the kinds of dao’s we’ve seen thus far.

These are just a few examples to illustrate a point - not meant to be exhaustive or perfect, just meaningfully illustrative.

My point is this: blockchain tech has uses far beyond the projects that have been successfully built in Solidity - and Solidity is simply a terrible choice for these particular requirements (even if it’s an excellent choice for many cryptocurrency projects).

I and many others see CosmWasm as a framework for building products of the future. It is uniquely positioned for this because Rust is a state-of-the-art general purpose language with a robust third-party package ecosystem. In addition to the language choice, CosmWasm has made some innovations that are purpose-built for this type of work (Rust-friendly storage API’s, helpers for on-chain randomness, etc.)

It’s not just a language or framework issue. The idea that EVM tooling is some ideal is… very far from how I see it and my first-hand experience. Game engines like Unity and Unreal should be more of an inspiration. Look to things used to build the kinds of applications I’m talking about today like Qt. Webdev frameworks like Next or newer Rust-centric things like Trunk are limited but get a lot of things right. There are many good examples out there. The EVM tooling does some things decently like quick iterative feedback cycles on code changes and integrated debugging tools, but the inherent language limitations make it impossible to share library code (not just types!) across the stack and so it becomes siloed in a fullstack project - you’re ultimately stuck writing your own tooling for full application development to tie it all together. It feels like a big step backwards to Makefiles, Gulp, and other solutions from 10+ years ago.

Those of us who have written a fair amount of CosmWasm code are very aware of the devrel and tooling problem. We know the tooling needs some love (notwithstanding the resources @adairrr shared)- but that’s exactly the point. CosmWasm is much better positioned than EVM to get over that hurdle and make more kinds of application development on blockchain tech easier. It takes resources to get there.

Maintaining the status quo may attract liquidity from pre-existing projects, but I do not see this as a strong way to inspire developers or business people who want to go further.

4 Likes
18

Outside the strengths and weaknesses of Cosmos as a blockchain versus other blockchains, which is something that only people like ourselves know or care about, Cosmos is an empty shell without CW. Without it, the only thing that remains is a bunch of starving networks fighting over the same scraps of cash while a few oligarchic teams and validators ride out the wave. My overall gut feeling is that the health and longevity of CW shouldn’t be compromised or left hanging on the promise of future funding.

Like a previous poster wrote, we haven’t even begun to explore applications of this technology to consumer markets, civics, science, generic “backend” engineering. Instead, the majority of dapps so far have been opportunistic land grabs, clones of existing products and trumped up “protocols.” At the same time, anything that’s not fully trustless or self-custodial is often dismissed or ignored because of premature ideological convictions. If CW isn’t sustained and strong going forward, there’s even less reason for anyone to try to use it in applications outside the domain of finance, which 90% of status quo blockchain teams think about.

As someone who has been writing CW exclusively and non-stop for almost three years now, writing dapps that are entirely powered by it, with no backends, no DBs, no indexers, etc., I know that it is far more capable than what I’d be able to do in EVM or SVM. As an engineer, CW is the principal value prop. If it’s neglected or downed out by other technologies, it would be a major blow, and all it would do is continue to centralize more wealth into hands that are already full, in my opinion.

5 Likes
19

Thank you for providing clarity on ICF’s position regarding CosmWasm funding. While I appreciate the strategic thinking behind supporting multiple VMs, I feel compelled to share some important perspectives as someone with extensive experience across CosmWasm, EVM, and SVM development.

First, I want to challenge the characterization of CosmWasm’s tooling as “nascent.” Having worked extensively with multiple VMs, I can attest that CosmWasm’s developer experience is exceptionally polished and productive. The tooling has consistently improved with each release, and in many ways offers advantages over other VMs. The comparison to EVM or SVM tooling perhaps deserves a more nuanced discussion.

While I understand the appeal of supporting multiple VMs to attract immediate liquidity and development activity, I worry we risk undervaluing CosmWasm’s unique strengths. It remains the only smart contract platform specifically designed for a multi-chain ecosystem - a fundamental differentiator that aligns perfectly with Cosmos’ vision.

My primary concern is that relegating CosmWasm to purely maintenance mode could have far-reaching consequences. Development teams need confidence in the long-term viability of their chosen platform. Without commitment to continued feature development (beyond just security patches), we risk eroding that confidence and potentially losing one of our ecosystem’s most innovative technologies.

I would strongly encourage ICF/Skip to take a more active role in facilitating a sustainable, shared funding model for CosmWasm’s ongoing development - not just maintenance. While the current funding commitment provides some stability, I believe we need a clearer path forward that ensures CosmWasm can continue to evolve and innovate.

The multi-VM strategy may well be the right approach for ecosystem growth, but it shouldn’t come at the cost of diminishing our support for a technology that embodies the core principles of Cosmos. CosmWasm’s design choices around inter-chain operability and security represent significant intellectual capital that we should carefully preserve and grow.

4 Likes
20

Hi, I’m the founder of Ark Protocol. I’m also a builder with +20 years of dev xp.

My “sole” skill is in CosmWasm. No Go and no Cosmos SDK.

What does it mean in case CosmWasm stays in “maintenance mode” and no more else is funding it?

This means it reaches its End-of-Lifecycle. Builders like me has either 2 options:

a) leave Cosmos
b) learn Go and Cosmos SDK

a) will hurt all of us and b) requires time and resources to invest for - simply: pain in the ass.

Ark, like many other protocols, build public goods and products. like ICS721 (aka the InterChain NFT bridge). There are 2 versions for the ICS721 spec: a module-based and a cw-based implementation.

cw-ics721 covers lots of more features than the module version, like:

  • additional security measures by providing outgoing and incoming proxy contracts
  • callbacks, so other users can provide even more interchain features

so again:

Defunding CosmWasm => Maintenance-only => End-of-Lifecycle => no future with builders either leaving or few turning into hardcore Go-Cosmos-SDK-nerds => no products, no value for Cosmos.

just my 2 cents on this!

One more things: thx Ethan, Simon and Confio team for building CosmWasm - in more than 2 decades as a dev I finally get back to the roots and enjoying building cool stuf. Thx a TON!!!

3 Likes