Research
DAO to (Anonymous) DAO Transactions
Blockchain assets are increasingly held by DAOs, consortium wallets, and custodial treasuries that authorize spending through threshold or multi-party key management, yet existing payment mechanisms…

- arXiv
- 2604.04369
- Published
- 2026-04-06
- Authors
- Minfeng Qi, Lin Zhong, Qin Wang
AI summary
DAO to (Anonymous) DAO Transactions — Paper SummaryOverview
Research area: Blockchain cryptography and security — specifically privacy-preserving payment protocols, threshold signatures, stealth addresses, and hierarchical key derivation for organization-controlled wallets.
Technical level: Advanced. The paper assumes familiarity with elliptic-curve groups, Shamir secret sharing, Lagrange interpolation, ECDSA, BIP32 derivation, and standard cryptographic security experiments.
Scope (one sentence): The paper formalizes transfers between threshold-controlled organizations, presents a framework called Dao² that combines distributed key derivation, distributed stealth-address generation, and threshold ECDSA, and evaluates a secp256k1 prototype.
What This Paper Is About
Blockchain assets are increasingly held by DAOs, consortium wallets, and custodial treasuries that authorize spending through threshold or multi-party key management, yet existing payment mechanisms and privacy tools are built around single-user wallets. The paper asks how one threshold-controlled organization can pay another — optionally with recipient anonymity — while ensuring the received funds land under the receiver's own distributed control rather than becoming a single private key that breaks the DAO's policy.
The authors address this by designing Dao², a unified transaction framework with an ordinary "DAO-to-DAO" mode and an anonymous "DAO-to-anonymous-DAO" mode, the latter being the main technical focus because it requires recipient privacy, distributed derivation, and threshold-controlled redemption to work together in one protocol.
Key Contributions
-
Problem formulation and unified framework. The authors formalize DAO-to-anonymous-DAO transactions and present Dao², a payment framework that extends blockchain payments from single-user wallets to threshold-controlled organizations, noting that classical user-level payments arise as a special case when a DAO degenerates to one member.
-
End-to-end protocol design (§ V). They specify a complete transaction flow built from two core modules — distributed key derivation (DKD) for ordinary DAO-to-DAO receiving and distributed stealth-address generation (DSAG) for anonymous receiving — plus a compatible threshold-signature authorization layer used for both outgoing payments and redemption.
-
Formal security analysis. They define and prove transaction correctness, spending unforgeability, recipient indistinguishability, and Byzantine robustness against a probabilistic polynomial-time adversary with fixed sub-threshold corrupted sets in each DAO.
-
Prototype evaluation on secp256k1. They report that the core of an anonymous DAO-to-DAO transaction for a typical-sized (e.g., 7-member) DAO finishes in under 28 ms with about 2.8 KB of communication, and scales linearly with DAO size.
Main Findings
-
Core timing and bandwidth: For a 7-member DAO, the core anonymous path takes under 28 ms with about 2.8 KB of counted public data, scaling linearly with DAO size.
-
Two additive updates carry the whole design: DKD adds the derivation offset ω^(k) to every receiver share, and DSAG adds the stealth offset ρ^(k). Lagrange reconstruction then yields shares of b^(k) + ρ^(k), whose public key is the one-time destination D^(k). Neither the child secret nor the one-time spending key is ever reconstructed, and these shares feed threshold ECDSA directly.
-
Received funds stay under distributed control: Unlike conventional stealth-address schemes that place scanning and spending keys with one user, the one-time output here is represented from the start by threshold shares, so redemption happens under the receiver DAO's threshold policy.
-
Only scheme combining all four capabilities: A comparison table positions Dao² as the only approach with threshold signing (TS), stealth addresses (SA), distributed key derivation (DKD), and end-to-end integration (E2E). Threshold ECDSA schemes have TS only; CryptoNote/RingCT has SA only; DSAG has TS and SA; BIP32/distributed KD has DKD only.
-
Privacy is explicitly scoped to external chain observers: The recipient-indistinguishability definition uses honestly generated receiver keys and honest qualified subsets on both sides. A sender-side participant who took part in the actual DSAG computation already knows the receiver's child public key B^(k) and the session descriptor, so privacy cannot hold against that party — the authors call this inherent to stealth-address constructions.
-
Known DH values leak future destinations: The paper notes that an earlier participant who retained a descriptor and DH value can recognize later destinations on the same lineage under the same sender key, because Ω^(k+1) = Ω^(k) + ω^(k+1)A. Fresh labels do not prevent this.
-
Threshold configuration used: The core experiments use a 2-out-of-n policy with t₁ = t₂ = 2, motivated by real DAO deployments where settings such as 2-out-of-3 balance security and operational simplicity. For a 2-out-of-n policy, at most one member of each DAO is corrupted under the sub-threshold assumption.
-
Conditional security for the signing layer: Spending security and signing robustness are conditional on an "interface requirement" for a threshold-ECDSA adapter that updates all public-key-dependent auxiliary state when applying a tweak. The paper states that it does not implement or prove a concrete auxiliary-state adapter; the prototype only checks standard-ECDSA output compatibility with threshold parameter t = 2.
-
Single-member degeneration: If each side has one member, the threshold operations reduce to ordinary single-key operations and the construction yields a user-level payment flow.
-
Not reported: The paper reports no gas costs, no mainnet or testnet deployment results, and no side-by-side timing comparison against the cited threshold-ECDSA, ring-signature, or HD-wallet systems.
Methodology in Plain English
The authors build the protocol out of three standard cryptographic tools and then glue them together with simple additive arithmetic.
-
Starting point — BIP32. In an ordinary hierarchical deterministic wallet, a child key is the parent key plus an offset derived deterministically from the parent's chain code and a tag: x_child = x_par + ω (mod q), with the matching public key X_child = X_par + ωG. The offset is computed with HMAC-SHA512.
-
Step one — make derivation distributed (DKD). Instead of one party holding the parent secret, every DAO member holds a Shamir share. Each member computes the same offset ω^(k) locally from the public parent key and chain code, then adds it to their own share. Every share moves by the same amount, so the aggregate public key moves to B^(k) = B^(k−1) + ω^(k)G, and nobody ever reconstructs the parent secret.
-
Step two — make the address unlinkable (DSAG). The sender DAO's members each compute a partial Diffie–Hellman term Ω_i = a_i·B^(k) using their shares of the sender's signing key a_i, prove each term is correct with a DLEQ proof (a proof that the same scalar was used in two group equations), and combine the verified terms with Lagrange coefficients to get Ω^(k) = a·B^(k). A hash of that shared secret plus a fresh public label gives the stealth offset ρ^(k) = H(Ω^(k) ‖ ξ^(k)), and the one-time destination is D^(k) = B^(k) + ρ^(k)G. The receiver DAO computes the mirror-image terms Ω'_j = b_j^(k)·A and, by ECDH, arrives at the same Ω, the same ρ, and therefore matching one-time shares d_j^(k) = b_j^(k) + ρ^(k).
-
Step three — authorize with threshold ECDSA. Both the outgoing payment (under the sender's key A) and the later redemption (under the one-time key D^(k)) are signed by a qualified subset using standard-output threshold ECDSA with an additive tweak. On chain, D^(k) looks like an ordinary ECDSA key; the threshold machinery stays off chain.
-
Security argument. The authors define a threat model with an adversary who picks which participants to corrupt before setup, holds those sets fixed, and stays below each DAO's threshold across the entire key lineage. They assume hardness of DL, CDH, and DDH, treat the stealth hash and the ECDSA message hash as independent random oracles, and require DLEQ proofs to be knowledge-sound and zero-knowledge. Against this model they set up four formal experiments (correctness, spending unforgeability, recipient indistinguishability, Byzantine robustness) and prove the corresponding bounds are negligible.
-
Evaluation. They implement a secp256k1 prototype, measure the core anonymous transaction path, and check that the resulting signature is compatible with standard ECDSA output at t = 2.
Why This Matters
Impact on research. The paper targets what it identifies as a genuine gap: prior work studies threshold signatures, privacy-preserving payment addresses, and hierarchical key derivation mostly as separate primitives, and no existing system lets an organization receive through unlinkable one-time addresses, keep the funds under threshold control, and continue managing them inside a derived wallet. It also contributes a privacy definition for recipients in a threshold setting and is explicit about where that definition stops applying — a level of precision that is useful for later work building on it.
Real-world applications:
- DAO treasuries: payroll, grants, and vendor payments between organizations without exposing which treasury received a given transfer.
- Custodial exchanges and custodians: institutional transfers where the receiving side must remain under multi-party key management rather than collapsing to a hot-wallet key.
- Consortium wallets and project treasuries: shared funds that need audit separation and organized, policy-based asset management across many receipts.
- Privacy-conscious institutional payments on public chains: the paper notes anonymous receiving is "also achievable on Bitcoin/Ethereum via wallet-level support," and points to Monero-style stealth payments as a natural fit.
Industry relevance. The design deliberately keeps on-chain artifacts ordinary: the one-time destination D^(k) is a standard ECDSA key, and non-anonymous mode works with standard addresses on any blockchain such as Bitcoin or Ethereum. That means compatibility with existing chain infrastructure rather than a new chain or consensus change — attractive for wallet vendors, custody providers, and DAO tooling teams.
Future Directions
-
Build and prove the threshold-ECDSA adapter. The paper's spending-security and signing-robustness results are explicitly conditional on an interface requirement covering public-key-dependent auxiliary state under additive tweaks. It states it does not implement or prove a concrete adapter, so this is the clearest open item before deployment.
-
Close the retained-DH-value linkability gap. An earlier sender-side participant who keeps a descriptor and DH value can recognize later destinations on the same receiver lineage. Designing forward-secure handling of Ω values, or a key-refresh discipline tied to each transfer, would address this.
-
Evaluate beyond a 2-out-of-n prototype. Core experiments use t₁ = t₂ = 2 and a secp256k1 prototype; the paper reports no on-chain costs, no deployment results, and no comparison timings against the cited systems. Measuring larger thresholds, other curves, and real chain environments is a natural next step.
-
Support parallel receipts and proactive refresh. The paper notes derivations on one receiver lineage are serialized (index k+1 only after transaction k is committed or abandoned, with parallel receipts on separate branches) and that proactive share refresh is orthogonal to the core construction. Turning both into concrete, tested procedures is open work.
Target Audience
Cryptography and blockchain protocol researchers working on threshold signatures, stealth addresses, or privacy-preserving payments will get the most from this paper, since its core content is security definitions and a compositional construction. Wallet and custody engineers evaluating how to keep institutionally received funds under multi-party control will find the protocol flow and interface requirements directly relevant. DAO infrastructure and treasury-tooling teams will benefit from understanding what the framework does and does not protect against. Graduate students in applied cryptography or distributed systems looking for a well-scoped example of combining ECDH, Shamir sharing, and threshold ECDSA into one protocol will also find it a useful case study. Beginners should expect to need background reading on the underlying primitives first.
Authors’ abstract
Blockchain assets are increasingly controlled by organizations rather than individuals. DAO treasuries, consortium wallets, and custodial exchanges rely on threshold authorization and multi-party key management, yet existing payment mechanisms still target single-user wallets, leaving no unified solution for organizational transfers. We formalize the problem of \emph{DAO-to-(anonymous)-DAO} transactions and present \textsc{Dao$^2$}, a framework that enables one…
The full abstract on arXiv