Quantum computers have not arrived yet, but shielded transactions stay on the chain forever. The Zcash community deployed quantum recoverability first, with the goal that “funds can be protected whenever quantum computers arrive,” and is preparing for a full post-quantum transition.
What do quantum computers threaten in Zcash?
A sufficiently large quantum computer can solve the elliptic curve discrete log using Shor’s algorithm. Bitcoin’s signatures and Zcash’s shielded protocols both rest on the assumption that this problem is hard. ZIP 2005 states that being able to compute even “a single” discrete log would allow arbitrary inflation or theft of funds.
Zcash has one more risk that Bitcoin does not. Because note commitments in the shielded pools are not binding against a quantum attacker, even if the proof system is replaced with a post-quantum one, forging and spending a note that is not in the tree would still be possible. So simply swapping the proof system does not solve it.
Risk area
What is at risk
Current response
Status
Balances (forgery, inflation)
Solving the discrete log just once would let an attacker forge ZEC in a shielded pool, because note commitments are not binding against quantum attacks.
Switch to ZIP 2005 quantum recoverability + a future recovery protocol
Being addressed
Spend authority (theft)
Signatures could be forged to spend other people's funds.
The recovery protocol additionally requires a proof of knowing a new secret value (qsk)
Being addressed
Privacy (harvest now, decrypt later)
An attacker who holds both the ciphertexts published on the chain and the recipient's address could decrypt past transactions once quantum computers exist.
Outside the scope of ZIP 2005. Post-quantum privacy is being researched in Tachyon and elsewhere
Research stage
Transparent addresses
Exposed to the same kind of attack as Bitcoin (attacking at the time of a spend, when the public key is revealed).
No change. Avoid reusing transparent addresses and keep funds in the shielded pool
Up to users
How close is the quantum threat?
Nobody knows exactly when. On the April 2025 Arborist call, forecasts ranged from “as soon as 3 years, or as late as 10 to 30 years,” and the call summarized that NIST and others treat 2030 to 2035 as the deadline for retiring elliptic curve cryptography. Meeting notes #99 →
Mar 31, 2026
Google Quantum AI’s resource estimate
Google Quantum AI, together with researchers from the Ethereum Foundation and Stanford, published an estimate that the secp256k1 discrete log (the signature curve of Bitcoin and Ethereum) could be solved in about 9 to 18 minutes with 1,200 to 1,450 logical qubits and fewer than 500,000 physical qubits. The resources required are much lower than earlier estimates.
Zcash’s reaction
“We need to move faster”
Citing the paper, Daira-Emma Hopwood said work on a fully post-quantum Zcash design and the deployment of quantum recoverability should be sped up, and said they would submit a correction to the paper’s claim that Halo 2 is better because it has no trusted setup. Meeting notes #122 →
Zcash uses other curves such as Pallas and Jubjub rather than secp256k1, but they all rely on the elliptic curve discrete log problem, so the threat is of the same kind.
Quantum recoverability: an escape route laid down now
ZIP 2005 “Ironwood Quantum Recoverability” is a proposal written by Daira-Emma Hopwood and Jack Grigg (str4d). A draft was created in March 2025, and it reached Proposed status in May 2026. The core idea is to slightly change only how notes are created.
Now: create notes in a ‘quantum-ready’ way
Every field of the note is fed into a hash to derive the random value (rcm). This makes it impossible to swap the note’s contents without breaking the hash, so the commitment becomes binding even against quantum attacks. The same technique is applied to how the incoming viewing key is derived.
Now: all Ironwood notes use this method
Since NU6.3 on July 28, 2026, every note in the Ironwood pool is created this way. That is why “recoverable note = Ironwood note” is an exact distinction.
Later: if the quantum threat becomes real
The elliptic-curve-based protocols are turned off, and a recovery protocol that uses a post-quantum proof system is turned on. A spender recovers funds by proving that the note was created as described above and that they know the new secret value (qsk).
What it solves
Balance preservation (preventing forgery), spend authority and spendability of Ironwood notes. Existing non-multisig keys and addresses do not need to be recreated, and it was designed with FROST multisig and hardware wallets in mind.
What it does not solve
Privacy stays as it is in every pool. It also does not make Zcash quantum-resistant right now. Funds left in Sprout, Sapling and Orchard cannot be recovered.
The Zcash community’s quantum preparation so far
Post-quantum privacy mentioned as an NU7 candidate
Daira-Emma Hopwood mentioned hybrid post-quantum note encryption to prevent 'harvest now, decrypt later' attacks. Read more →
ZIP 2005 drafted
A proposal began to change how notes are created so that funds could be recovered through a future post-quantum protocol. Read more →
Design discussed on the Arborist call
The call covered the problem that a quantum attacker can forge today's note commitments, the fix of deriving rcm from a hash, and even the FROST key structure. Read more →
Proposal to add a new key (qsk) and apply it to v6 transactions
A proposal was made to add a quantum-ready key that is useful for hardware wallets and to apply it to every shielded output of v6 transactions. Read more →
Overwhelming support in the NU7 sentiment poll
'Orchard quantum recoverability' was supported by 90.5% of ZCAP and 94.6% of coinholders. Read more →
Google's quantum resource estimate paper
An estimate was published that the secp256k1 discrete log could be solved with 1,200 to 1,450 logical qubits and fewer than 500,000 physical qubits. On the April Arborist call, Daira-Emma said Zcash should speed up its preparation. Read more →
ZIP 2005 moves to Proposed
The final changes to the specification were merged, and work continued on turning the informal security argument into a formal proof. Read more →
Ironwood (NU6.3) activated
Notes in the Ironwood pool are created in a quantum-recoverable way from the start. Wallets were advised to move existing funds to Ironwood. Read more →
Shielded Labs launches Epoch
A three-person team was formed to develop post-quantum cryptography to production quality and formally verify it. The goal is a first implementation by the end of 2027. Read more →
Toward a fully post-quantum Zcash
Quantum recoverability is only a bridge. Work is under way in parallel on actually becoming post-quantum once that bridge is crossed.
Shielded Labs
Epoch: developing post-quantum cryptography
It aims for ‘post-everything’ cryptography that withstands quantum computers, AI-assisted attacks and state-level adversaries, and guarantees even the implementation through formal verification. Ulrich Haböck, formerly of StarkWare, and two engineers from Aztec have joined, and the goal is a production implementation by the end of 2027. Read more →
Project Tachyon
The path to post-quantum privacy
It separates the wallet-to-wallet payment protocol from the on-chain shielded protocol, removing quantum-vulnerable privacy assumptions from the chain. The roadmap includes an approach using PIR and post-quantum key exchange. Official overview →
Recovery protocol
Choosing a post-quantum proof system
ZIP 2005 sets out only the outline of the recovery protocol and deliberately leaves open which post-quantum proof system and hash to use. It also considers that there is no need to put them into the consensus rules now.
Legacy cleanup
Closing the pre-quantum pools
ZIP 2003 proposes stopping Sprout spends by disallowing v4 transactions, and on the forum a discussion has begun about making Sapling withdraw-only. The post that started the Sapling discussion sets the goal of freezing all pools built the pre-quantum way before quantum computers arrive. Sapling discussion →
What users can do now
Move funds to Ironwood
Only Ironwood notes can be recovered through the recovery protocol. Shielded wallets that support Ironwood provide a way to move existing balances to Ironwood. Shielded wallet list →
Don't leave funds in Sprout, Sapling or Orchard
ZIP 2005 states that funds in these pools will become unspendable once those protocols are turned off. When that will happen has not been decided, but it is safer to move them in advance.
Don't store funds long-term in transparent addresses
Transparent addresses carry the same quantum risk as Bitcoin. In particular, reusing the same transparent address leaves the public key exposed for a long time.
Don't publish your address widely
A 'harvest now, decrypt later' attack needs the recipient's address. Using a new shielded address for each counterparty narrows the exposure.
Keep wallets and firmware updated
Quantum recoverability is a change to how wallets create notes. Hardware wallets also need firmware that supports the new key structure.
Frequently asked questions
Is Zcash quantum-resistant today?
No. All current shielded protocols rely on the hardness of the elliptic curve discrete log. Quantum recoverability lays down a path for moving to safety later. It is not a mechanism that stops quantum attacks right now.
Why 'quantum recoverability' and not 'quantum resistance'?
Because the proposal itself does not stop quantum attacks or recover funds, but makes it possible to recover them in the future. It was renamed because 'resistance' would also overstate things, as privacy is not covered.
If quantum computers arrive, will my past transactions be revealed?
It is possible if the attacker has both the transaction's ciphertext (published on the chain) and the recipient's address. ZIP 2005 explains that without the address, the zero-knowledge proofs and signatures alone leak no additional information.
Halo 2 has no trusted setup. Doesn't that make it safer?
It makes no difference against quantum attacks. Daira-Emma Hopwood countered that solving a single discrete log breaks the proof system, value commitments and note commitments alike, so whether there is a trusted setup does not matter.
Is it better than Bitcoin?
The difference is that Zcash has already deployed the recovery path through Ironwood and received more than 90% support in the NU7 sentiment poll. However, the risk to transparent addresses is the same as Bitcoin's, and the privacy-side threat is an additional challenge unique to Zcash.