1 Signature scheme 0/10, weighted 30%
Band 0: Classical signatures only on mainnet, with nothing post-quantum protecting live funds. Short-lived devnets and one-off benchmarks sit here: they demonstrate research rather than something a user can hold keys on.
Cosmos Hub accounts sign with ECDSA on secp256k1 and validators sign consensus messages with Ed25519. Both are broken by Shor's algorithm, and nothing post-quantum protects ATOM on mainnet: Gaia runs Cosmos SDK v0.53.8, which has no ML-DSA key type.
source
2 Deployment stage 2/10, weighted 25%
Band 2: Tier 4: Debating.
Tier 4: Debating. Cosmos Labs, the Interchain Foundation subsidiary that maintains the Cosmos SDK and, by its own statement in June 2026, is bringing Hub technical maintenance in-house, has built and released ML-DSA-65 account and consensus keys in the Cosmos SDK, which is post-quantum work on exactly the signatures the Hub uses. It is not Tier 3 because nobody with authority over the Hub has published a plan to adopt it: no governance proposal, no Gaia release and no dated upgrade.
source
3 NIST alignment 5/10, weighted 15%
Band 3 to 5: NIST schemes are referenced as candidates without a committed selection, or the work is bespoke research.
The one post-quantum scheme on the table for the Hub's signatures is ML-DSA-65, a parameter set of the final FIPS 204 standard, implemented in the Cosmos SDK for both account and consensus keys. It has been built for the framework, not proposed for the Hub.
source
Placement in the band The top of the band rather than a 3 or a 4: the scheme is a single, specific parameter set of a final standard, already implemented for both roles the Hub needs. It cannot reach 6 because no one has proposed it for the Hub, so it is a framework option rather than the chain's selection.
4 Migration path 5/10, weighted 15%
Band 3 to 5: Migration is acknowledged as a problem with no agreed mechanism, or the mechanism is contested.
The mechanism exists in the framework. Under SDK v0.55 a validator rotates its consensus key to ML-DSA in place, and the chain counts as post-quantum once two thirds of voting power has rotated. Holders get no in-place migration: they create a new ML-DSA account with the command-line tool and send their funds to it, Ledger devices cannot sign with ML-DSA, and the Hub would need a governance-approved upgrade to v0.55 before any of it applies.
source
Placement in the band The top of the band rather than a 4: a working route exists and the Hub's ordinary software-upgrade governance could deliver it. It is not a 6 because there is no published plan for the Hub, and account holders would have to move funds to a new address without hardware-wallet support.
5 Exposure 2/10, weighted 10%
Band 0 to 2: Public keys are exposed for effectively all accounts, or a large, measured share of total supply sits in exposed addresses.
A Cosmos account address is a hash of the secp256k1 public key, so an address that has never signed keeps its key hidden. The key is revealed with the account's opening transaction, and staking, which is how ATOM earns, needs a signed delegation, so nearly all actively held ATOM sits behind a revealed key. Validator consensus keys are public by design.
source
Placement in the band The top of the band rather than a 3: the address structure is the same as on Ethereum-style key-hash chains, where staking or any send reveals the key, so effectively all ATOM in use is exposed. It is not a 0 or a 1 because never-used addresses do keep their keys hidden.
6 Verification 3/10, weighted 5%
Band 3 to 5: A post-quantum component exists, on mainnet, on a persistent testnet or as a merged implementation, but is unreviewed.
A post-quantum component exists as merged code: ML-DSA-65 validator keys and account keys were merged into the Cosmos SDK on 27 May and 2 June 2026. Both pull requests were approved by two repository maintainers, aljo242 and mattac21, with comments from an automated reviewer, and we found no review from outside the maintaining team and no audit. The code is not in Gaia and protects no Hub funds.
source
Placement in the band The floor of the band rather than a 4: the code is merged and released, but in a framework the Hub has not adopted, and it runs on no Hub network. It is not a 2 because a real, open source implementation of the Hub's replacement signatures exists.