Algorand ALGO

Claim: Algorand has published a comprehensive cryptographic inventory identifying Ed25519 signatures, ECVRF, and their exposure across spend authorization, consensus, and state layers. The three-part roadmap (secure the past, present, future) explicitly acknowledges quantum vulnerability of consensus and default accounts.

Coverage basis: Public documentation and official blog posts

Implementation score: 1 · Evidence confidence: High

Issue classification: none · Score treatment: not applicable

Algorand's cryptographic inventory is unusually thorough for the blockchain space. Google Quantum AI's March 2026 whitepaper cited Algorand 32 times as a leading example of PQC deployment. The Foundation's public communications are transparent about what is and is not protected.

• https://algorand.co/technology/post-quantum
• https://algorand.co/blog/algorand-post-quantum-ledger
• https://algorand.co/blog/technical-brief-quantum-resistant-transactions-on-algorand-with-falcon-signatures
• https://algorand.co/algorands-path-to-post-quantum-readiness

Claim: Algorand provides source code (go-algorand, falcon-signatures CLI), mainnet transaction proof (first Falcon tx Nov 3, 2025), State Proof transaction counts (140K+), developer documentation, and reproducible tooling.

Coverage basis: Open-source code, mainnet transactions, developer portal

Implementation score: 1 · Evidence confidence: High

Issue classification: none · Score treatment: not applicable

The falcon-signatures CLI (AGPL-3.0, 5 releases, latest v1.1.2) provides a reproducible end-to-end workflow. State Proof activity is publicly measurable (140K+ transactions over 3.5 years). Mainnet transaction proof for Falcon spend authorization is verifiable on-chain.

• https://github.com/algorand/go-algorand
• https://github.com/algorandfoundation/falcon-signatures
• https://developer.algorand.org/docs/get-details/stateproofs/

Claim: Optional mainnet Falcon-1024 spend authorization exists via LogicSig account abstraction and the falcon_verify AVM opcode (deployed September 2024 in v4.3.0 consensus upgrade). Default spend authorization remains Ed25519. Users can create new Falcon-controlled accounts or rekey existing Ed25519 accounts to a Falcon authorizer.

Coverage basis: Optional mainnet PQ path; default remains Ed25519

Implementation score: 0.75 · Evidence confidence: High

Issue classification: quantum-critical vulnerability · Score treatment: score-reducing

Quantum blocker: Default spend authorization remains Ed25519-only. Over 99% of ALGO value-at-risk is held in Ed25519 accounts with exposed public keys. New accounts default to Ed25519 with no warning or migration prompt.

Assurance: No independent audit of the falcon_verify opcode or LogicSig-based Falcon account abstraction.

The Falcon spend authorization path is technically operational but practically limited: requires CLI expertise, 4-transaction groups, and lacks feature parity with native Ed25519 (no multisig, no delegated LSig). The Algorand Foundation describes this as 'experimental' and 'not meant to be the ultimate solution.'

• https://algorand.co/blog/technical-brief-quantum-resistant-transactions-on-algorand-with-falcon-signatures
• https://algorand.co/technology/post-quantum
• https://hacken.io/discover/falcon-fn-dsa-post-quantum-signature/

Claim: Falcon LSig addresses are derived off-curve to prevent accidental Ed25519 control paths. Rekeying enables migration from Ed25519 to Falcon authorizer while preserving the original account address. However, default Ed25519 accounts expose public keys on first transaction with no automatic protection.

Coverage basis: PQ controls exist but not default; long-exposure paths remain dominant

Implementation score: 0.5 · Evidence confidence: High

Issue classification: quantum-critical vulnerability · Score treatment: score-reducing

Quantum blocker: Default Ed25519 accounts expose public keys on-chain, creating long-exposure quantum-vulnerable ownership paths for the vast majority of value.

Assurance: Off-curve derivation design appears sound but has not been independently audited.

The off-curve address derivation for Falcon LSig accounts is a well-designed mitigation against accidental Ed25519 control. The rekeying mechanism is a genuine innovation for migration. However, the design does not prevent users from creating new Ed25519 accounts by default.

• https://algorand.co/blog/algorand-post-quantum-ledger
• https://algorand.co/blog/technical-brief-quantum-resistant-transactions-on-algorand-with-falcon-signatures

Claim: Algorand consensus relies on Ed25519 signatures for block proposals and committee voting, and ECVRF for cryptographic sortition. Both are vulnerable to Shor's algorithm. The project acknowledges this as the 'hardest step' and states that PQ-VRF is 'active research.' No PQ consensus mechanism is deployed or in testnet.

Coverage basis: No PQ protection for consensus

Implementation score: 0 · Evidence confidence: High

Issue classification: quantum-critical vulnerability · Score treatment: cap-applying

Quantum blocker: Consensus authentication (Ed25519 block proposals/votes + ECVRF sortition) is entirely quantum-vulnerable with no deployed replacement. A quantum adversary could predict committee membership, forge proposals, and disrupt or manipulate consensus. This directly triggers the QRI Readiness & Risk Cap of 70.

Assurance: Multiple academic proposals for PQ-VRF exist (LB-VRF 2021, iVRF 2023, X-VRF, hash-based VRF 2026), but none have been adopted by Algorand.

The Algorand Foundation now employs Chris Peikert (co-author of GPV framework) as Chief Scientific Officer, which may accelerate PQ-VRF research. The iVRF proposal (2023) from ACM CCS shows particular promise with 32-byte overhead and faster-than-ECVRF evaluation, but remains academic.

• https://algorand.co/technology/post-quantum
• https://algorand.co/blog/technical-brief-quantum-resistant-transactions-on-algorand-with-falcon-signatures
• https://algorand.co/algorands-path-to-post-quantum-readiness
• https://github.com/algorand/go-algorand

Claim: State Proofs use Falcon-1024 signatures from stake-weighted participants every 256 rounds to create compact, PQ-secure attestations of chain state since 2022. 140K+ state proof transactions generated over 3.5 years.

Coverage basis: Mainnet PQ for history attestation; active state production depends on vulnerable consensus

Implementation score: 1 · Evidence confidence: High

Issue classification: none · Score treatment: not applicable

Assurance: No independent audit of the State Proof Falcon implementation identified. The SumHash512 hash function used in State Proofs is a bespoke construction (subset-sum family) with a narrower review base than SHA-2/SHA-3.

State Proofs represent genuine production PQ protection for chain history. This is a significant achievement: even if consensus is compromised by a quantum adversary, the historical ledger cannot be silently rewritten without detection. Scored at 1.0 because State Proofs are mandatory, default, and have been in production since 2022.

• https://developer.algorand.org/docs/get-details/stateproofs/
• https://algorand.co/technology/post-quantum
• https://algorand.co/blog/algorand-post-quantum-ledger

Claim: Algorand does not have a native shielded transaction pool, confidential asset layer, or ZK privacy protocol.

Coverage basis: N/A - no native privacy layer

Implementation score: 0 · Evidence confidence: High

Issue classification: none · Score treatment: not applicable

The absence of native privacy features is an architectural characteristic, not a security gap.

Claim: No evidence found regarding PQ protection for P2P transport or node identity in the evidence dossier.

Coverage basis: Unknown; likely classical

Implementation score: 0 · Evidence confidence: Low

Issue classification: assurance-only caveat · Score treatment: note-only

Assurance: P2P layer PQ status is not documented in the evidence dossier. Node identity is not consensus-critical for spend or finality on Algorand, so this may be satisfiable by design, but evidence is lacking.

P2P node identity is not consensus-critical, spend-critical, bridge-critical, or custody-critical in Algorand's architecture. While a quantum adversary could potentially impersonate nodes at the network layer, this would not directly enable theft, forgery, or consensus manipulation.

Claim: Lute Wallet has added experimental Falcon support. The Falcon Signatures CLI provides developer tooling. No major wallet (Pera), exchange, custody provider, or HSM vendor supports Falcon accounts in production workflows.

Coverage basis: Experimental/CLI only; no production wallet or custody support

Implementation score: 0.25 · Evidence confidence: Medium

Issue classification: operational/product caveat · Score treatment: note-only

Assurance: Lute Wallet's Falcon support is described as 'experimental' and has not been independently reviewed. No HSM or hardware-wallet vendor has announced Falcon integration for Algorand.

Wallet and custody support is a practical prerequisite for migration. Without Pera Wallet, exchange, or custody support, the Falcon spend authorization path is inaccessible to virtually all users.

• https://algorand.co/blog/january-2026-algo-insights-report
• https://github.com/algorandfoundation/falcon-signatures

Claim: Falcon accounts are opt-in since Nov 2025. Adoption rate and percentage of value migrated is not publicly reported. Based on the experimental nature of the tooling, lack of wallet/exchange support, and absence of any migration campaign, coverage is conservatively estimated at negligible (<1%).

Coverage basis: Negligible migration coverage; vast majority of value quantum-vulnerable

Implementation score: 0.05 · Evidence confidence: Low

Issue classification: quantum-critical uncertainty · Score treatment: score-reducing

Quantum blocker: Over 99% of value-at-risk remains in Ed25519 accounts with exposed public keys. No protocol enforcement, deprecation, or mandatory migration mechanism exists.

Assurance: Coverage estimates are inferred from the absence of data rather than measured directly. No public dashboard tracks Falcon-migrated value.

Algorand does not publish Falcon account adoption statistics. The Q1 2026 Transparency Report mentions State Proof transactions but provides no data on Falcon account creation or migration. Coverage is conservatively estimated at <1% based on the experimental nature of the tooling.

• https://algorand.co/technology/post-quantum

Claim: No evidence that any major treasury, exchange, foundation, bridge, or protocol-controlled wallet has been migrated to Falcon. The Algorand Foundation has not announced migration of its own wallets.

Coverage basis: No critical wallet migration identified

Implementation score: 0 · Evidence confidence: Medium

Issue classification: quantum-critical uncertainty · Score treatment: score-reducing

Quantum blocker: No critical treasury, foundation, exchange, bridge, or protocol-controlled wallet has been publicly migrated to Falcon.

Assurance: Confidence is Medium because the absence of migration cannot be definitively proven from public data alone. However, given the high profile of such migrations and the Foundation's transparent communication, the absence of any announcement is strong negative evidence.

The Algorand Foundation holds ~19.5% of stake (~390M ALGO) and has not announced migration. Major ecosystem protocols (Tinyman, Pact, Folks Finance, Lofty) have not announced Falcon migration.

Claim: Algorand has clearly identified Ed25519 accounts as the legacy vulnerable pool and documented the rekeying-based migration path. However, there is no protocol-level mechanism to deprecate, freeze, burn, or force migration of vulnerable accounts. No policy exists for handling dormant or abandoned Ed25519 accounts.

Coverage basis: Identified and measurable; no enforcement or deprecation policy

Implementation score: 0.25 · Evidence confidence: Medium

Issue classification: operational/product caveat · Score treatment: note-only

Assurance: The absence of a published policy for unmigratable accounts creates uncertainty about how the protocol would handle permanently vulnerable value.

Algorand's account model makes it possible to identify Ed25519 accounts on-chain. The rekeying mechanism provides a technical migration path, but no protocol rule prevents abandoned accounts from remaining vulnerable indefinitely.

• https://algorand.co/blog/algorand-post-quantum-ledger
• https://algorand.co/technology/post-quantum

Claim: Algorand has published a clear three-part roadmap: (1) secure the past via State Proofs (completed 2022), (2) secure the present via Falcon accounts (in progress, experimental mainnet), (3) secure the future via PQ-VRF for consensus (research phase).

Coverage basis: Public roadmap with clear phases; consensus phase still research

Implementation score: 0.75 · Evidence confidence: High

Issue classification: none · Score treatment: not applicable

The roadmap is unusually detailed for the blockchain space. Phases 1 and 2 have specific deliverables and evidence, while phase 3 remains at the research-plan stage.

• https://algorand.co/algorands-path-to-post-quantum-readiness
• https://algorand.co/blog/algorand-post-quantum-ledger
• https://algorand.co/technology/post-quantum

Claim: Falcon account creation and migration require CLI expertise (falcon-signatures tool), 4-transaction groups, and manual address derivation. Lute Wallet has experimental support. The primary wallet (Pera) does not support Falcon. Exchange and custody support is absent. Default account creation remains Ed25519 with no migration prompts or warnings.

Coverage basis: Developer-oriented experimental tooling; not accessible to average users

Implementation score: 0.25 · Evidence confidence: High

Issue classification: operational/product caveat · Score treatment: score-reducing

Migration accessibility is the critical bottleneck for Algorand's quantum readiness. The technical capability exists but the practical path is inaccessible to virtually all users.

• https://github.com/algorandfoundation/falcon-signatures
• https://algorand.co/blog/technical-brief-quantum-resistant-transactions-on-algorand-with-falcon-signatures

Claim: No enforcement mechanisms exist: no deprecation of Ed25519 signing, no freeze or burn policy for vulnerable accounts, no mandatory migration deadlines, no restrictions on creating new Ed25519 accounts, no exchange or custody coordination framework.

Coverage basis: No enforcement or coordination mechanisms deployed

Implementation score: 0 · Evidence confidence: High

Issue classification: quantum-critical vulnerability · Score treatment: score-reducing

Quantum blocker: No protocol-level enforcement, deprecation, or mandatory migration mechanism exists. Users can indefinitely create and use Ed25519 accounts with no restrictions, warnings, or migration deadlines.

The absence of enforcement mechanisms is a structural blocker to achieving meaningful migration coverage.

Claim: Algorand has general governance processes (xGov) and Foundation security contacts. No quantum-specific incident-response playbook has been published.

Coverage basis: General governance exists; no quantum-specific IR

Implementation score: 0.25 · Evidence confidence: Medium

Issue classification: assurance-only caveat · Score treatment: note-only

Assurance: The absence of a quantum-specific IR playbook is an assurance gap but does not directly create a quantum-vulnerable path.

Algorand's governance infrastructure (xGov, Foundation Board, Ecosystem Advisory Council) provides a framework for emergency response. The appointment of Chris Peikert as CSO strengthens cryptographic governance.

Claim: Algorand uses Falcon-1024, a NIST-selected lattice-based signature scheme (Round 3, 2022) targeting NIST Security Level 5 (equivalent to AES-256).

Coverage basis: NIST-standardized algorithm appropriate to the use case

Implementation score: 1 · Evidence confidence: High

Issue classification: none · Score treatment: not applicable

Assurance: Falcon's formal security proof was only recently established (Eurocrypt 2026, IBM Research). Ongoing cryptanalytic work continues to refine concrete security bounds.

Falcon's selection is well-justified for Algorand's constraints: compact signatures (~1.5KB), fast verification (<200 microseconds), deterministic mode available.

• https://algorand.co/technology/post-quantum
• https://algorand.co/blog/technical-brief-quantum-resistant-transactions-on-algorand-with-falcon-signatures
• https://csrc.nist.gov/projects/post-quantum-cryptography

Claim: No publicly documented independent cryptographic audit of Algorand's Falcon integration (falcon_verify opcode, LogicSig account abstraction, off-curve address derivation, State Proof Falcon usage) has been identified.

Coverage basis: Algorithm reviewed via NIST process; protocol-specific integration unaudited

Implementation score: 0.25 · Evidence confidence: Low

Issue classification: assurance-only caveat · Score treatment: confidence-only

Assurance: Per QRI v3.1, the absence of an independent audit does not by itself cap the QRI Score when the quantum-critical property is verifiable from public code and mainnet evidence. Confidence is capped at Medium.

The Falcon implementation credits Thomas Pornin and Chris Peikert/David Lazar. These are respected cryptographers, and deterministic signing with constant-time practices is documented. However, protocol-specific integration risks have not been independently assessed.

Claim: The Falcon Signatures CLI is open-source (AGPL-3.0) on GitHub. go-algorand is open-source. The implementation is publicly buildable and reproducible.

Coverage basis: Fully open-source and reproducible

Implementation score: 1 · Evidence confidence: High

Issue classification: none · Score treatment: not applicable

The falcon-signatures repository provides a complete, documented workflow from key generation (with BIP-39 mnemonics) to on-chain transaction submission.

• https://github.com/algorandfoundation/falcon-signatures
• https://github.com/algorand/go-algorand

Claim: Algorand's documentation mentions that Falcon can be 'tweaked or integrated with other algorithms as the cryptography field evolves.' The post-quantum strategy article discusses crypto agility in the context of account migration. No formal parameter agility specification exists.

Coverage basis: Acknowledged but not formally documented

Implementation score: 0.5 · Evidence confidence: Medium

Issue classification: operational/product caveat · Score treatment: note-only

Assurance: The absence of a formal parameter agility specification is an assurance gap but does not create a current quantum-vulnerable path.

The LogicSig-based Falcon account abstraction provides some inherent agility: the LSig program can be updated to use different PQ algorithms without changing account addresses (via rekeying to a new LSig).

• https://algorand.co/technology/post-quantum
• https://algorand.co/blog/algorand-post-quantum-ledger

Claim: Falcon uses deterministic signing mode (eliminating randomness-related side channels). The implementation uses constant-time coding practices. Falcon is not a stateful scheme (unlike XMSS/LMS), so state-management and anti-reuse concerns are minimal.

Coverage basis: Deterministic signing, constant-time, non-stateful scheme

Implementation score: 0.75 · Evidence confidence: Medium

Issue classification: assurance-only caveat · Score treatment: note-only

Assurance: Deterministic signing and constant-time claims are documented by the project but have not been independently verified through side-channel analysis or formal audit. Falcon's floating-point-based signing introduces potential for subtle implementation side channels.

Falcon's use of floating-point arithmetic in signing (Gaussian sampling over NTRU lattices) is a known implementation challenge. The deterministic mode mitigates some risks but does not eliminate all side-channel concerns.

• https://algorand.co/blog/technical-brief-quantum-resistant-transactions-on-algorand-with-falcon-signatures
• https://github.com/algorandfoundation/falcon-signatures

Claim: Algorand documents Falcon verification at under 200 microseconds. Signature sizes (~1.5KB), public key sizes (1,793 bytes), and LogicSig budget constraints are discussed. Academic research estimates 1.14x-3.4x throughput reduction for PQ-VRF.

Coverage basis: Partial analysis exists; no comprehensive formal benchmark

Implementation score: 0.75 · Evidence confidence: Medium

Issue classification: assurance-only caveat · Score treatment: note-only

Assurance: Performance metrics are provided by the project without independent reproduction.

The documented verification speed (<200us) is fast enough for block validation. The 4-transaction group requirement combined with ~1.5KB signatures creates practical throughput and cost implications that are acknowledged but not formally benchmarked.

• https://algorand.co/blog/technical-brief-quantum-resistant-transactions-on-algorand-with-falcon-signatures
• https://link.springer.com/chapter/10.1007/978-3-662-64331-0_29