Zcash developers are preparing an upgrade that would permanently strand coins left in the network's oldest privacy pool unless their owners move them first. The NU7 upgrade would leave about 22,309 ZEC in the legacy Sprout pool unspendable, even for holders who still have their private keys, and a decision on mainnet activation is due on 20 October. Separately, another team has proposed a way to keep shielded swaps private from future quantum computers. Meanwhile, on-chain investigators say hackers are moving stolen funds into ZEC.
What NU7 changes for Sprout
The change is set out in ZIP 2003, a Zcash improvement proposal described by CryptoSlate. Its listed owner is Daira-Emma Hopwood, and it was created on 31 October 2024. Under the proposal, the network would reject version 4 transactions from NU7 onward. Sprout transfers only work in that format, and the newer version 5 and version 6 formats do not support them.
That is the core problem. A private key authorises a spend, but the network still has to accept the transaction that carries it. If no valid format remains, Sprout holders would have their keys but no way to use them.
The rule covers every version 4 transaction, including transparent and Sapling ones. Those can usually be rebuilt in a newer format, but Sprout transactions cannot. Re-signing an old transaction does not help either, because the format itself would be invalid.
The coins would not be destroyed. ZIP 2003 keeps Sprout's remaining balance in the issued supply. It also leaves open the theoretical option of re-enabling version 4 transactions or building another way to retrieve the funds, while warning that spending may never be restored. CryptoSlate cites the zkp.baby dashboard, which showed 22,309.44 ZEC in Sprout at block 3,521,260 on 17 October.
Why developers want to close Sprout
The main argument is maintenance and security. Sprout needs its own verification logic. It also needs its own handling of private notes and spent-note identifiers, the markers that stop a coin from being spent twice. Nodes have to do that extra work when they accept new transactions, even though few people still use the pool. Verification cost also affects how exposed the network is to denial-of-service attacks.
Some protections are already shared with newer systems. Sprout uses cryptographic components that Sapling also uses, and pool accounting checks limit how far any balance error could go. The proposal argues that Sprout-specific complexity still remains for every node implementation to maintain. This would be the second step in retiring the pool: an earlier proposal, ZIP 211, stopped new value entering Sprout but kept withdrawals open.
The costs and benefits fall unevenly. The whole network gets the security gain, but individual holders carry the migration burden. A September coordination announcement reported agreement among ecosystem organisations and engineering teams. CryptoSlate described the coinholder poll result as advisory.
What legacy holders should check
Anyone who held ZEC in the network's early years should check whether any of it is still in Sprout. The documentation highlights these points:
- Keep old wallet files. Zallet, the newer wallet software, does not import Sprout spending keys or funds. Its migration guide tells users to keep their original wallet.dat file.
- Allow time for recovery. The Argos recovery tool supports holders who have a legacy wallet file or standalone Sprout spending keys. A wallet file with usable note data and a cached witness (the proof data needed to spend a note) can skip a historical scan. With a key alone, Argos estimates roughly 280 GB of data transfer and possibly days of processing, while storing under 500 MB on disk.
- Know where the coins land. Argos sweeps Sprout funds into Sapling, even if the user supplies a Unified Address. Moving them anywhere else is a separate step.
- Expect some loss of privacy. Under ZIP 308, value moving from Sprout to Sapling passes through public pool accounting, so the amount becomes visible, although no transparent address is needed. ZIP 308 coordinates timing and randomises amounts to make transactions harder to link. ZIP 2003 warns that a short or poorly coordinated exit window makes unusual amounts and timing easier to tie to their holders.
Node operators have a separate issue. ZIP 259 waives the usual recommendation of about three months between the first supporting software release and activation. Unsupported software may stay online and keep following pre-NU7 rules, so operators need compatible software to validate the upgraded network.
A quantum fix for shielded swaps
The second change deals with a narrower risk that is harder to spot. According to CoinDesk's report on the Zakura proposal, the team behind Zakura, a newer version of the software that runs the Zcash network, wants every ZEC swap made through NEAR Intents to have its own viewing key.
NEAR Intents is a service for swapping tokens across blockchains. When a user swaps shielded ZEC for USDC, the service posts one of the user's private Zcash addresses on NEAR's public chain so it knows where to send any refund. The address stays there even if no refund is ever needed.
On today's computers, that address reveals nothing. But every Zcash address contains a public key derived from the wallet's viewing key, the secret a wallet uses to detect incoming payments. In theory, a large enough quantum computer running Shor's algorithm could work backwards from the address to that key. A wallet's many addresses normally share one viewing key, so cracking one address could expose every payment the account has received, including amounts and memos. The funds themselves would stay safe, because spending requires a separate key.
Under Zakura's design, each swap gets its own viewing key, so a cracked refund address would expose only that one swap. This creates a practical problem: after restoring from a backup, a wallet would have to scan the whole chain with hundreds of keys. Zakura's answer is private information retrieval, a technique that lets a wallet query a database for its refunds without the database learning which addresses it asked about.
No quantum computer capable of this attack has been demonstrated, and nobody has a firm timeline for when one could be built. Zakura says wallets built on its open-source libraries, including Vizor, could add the feature within weeks. Hardware wallets would follow once their firmware supports it. That would come ahead of the January target that Zcash developers set earlier this month for quantum-resistant payments.
Stolen funds and the privacy trade-off
These upgrades come as investigators report hackers using Zcash's shielded pools. According to crypto.news, ZachXBT reported on 30 September that wallets linked to the Bitget attackers had sent about 2,700 ZEC, worth roughly $3.8 million at the time, into Ironwood. Zcash launched Ironwood as a new shielded pool on 28 July 2026, after finding a vulnerability in its older Orchard pool. Bitget puts its total loss from the 24 September breach at about $387.5 million across several chains, but that figure covers the whole incident, not just the funds sent to Zcash.
On 17 October, analyst Specter said funds linked to the Cryptobills Ledger incident and the $12 million Triple-A exploit had also entered Zcash. Specter did not publish a full transaction trail or say how much was moved, and the claims have not been independently verified. Investigators can often identify deposits into a shielded pool, but the public record usually does not show where the funds go next.
None of the reports point to a flaw in Zcash itself. They do sharpen a long-running tension, though. Developers keep strengthening privacy, now including protection against future quantum attacks, while the same features attract attention when stolen funds use them. The sources do not report any regulatory response so far.
What to watch
- 20 October: the planned decision on NU7 mainnet activation. Testnet passed activation at height 4,465,026, which was scheduled for around 6 October using Zebra 7.0.0-rc.0.
- The mainnet activation height: once it is set, it becomes the deadline for Sprout exits. 5 November remains the target, depending on the 20 October decision.
- The Sprout balance: whether the roughly 22,309 ZEC shown on zkp.baby falls as holders migrate before the cutoff.
- Zakura's per-swap keys: whether Vizor and other software wallets ship them in the coming weeks, ahead of the January target for quantum-resistant payments.
- Further on-chain evidence: transaction-level data showing how much stolen value from the Cryptobills Ledger and Triple-A incidents actually reached ZEC.
