BTC exposure watch
STATUS PRE‑EVENT
Q‑Day — 2035
2028 2030 · NIST deprecated 2035 · disallowed 2055
Key‑recovery rate
How this control drives the cascade
Bitcoin exposed‑supply estimate

Parallel CRQCs — 1
Machines assumed independent, against separate targets. Grover, unlike Shor, would not benefit. See the Abelian column.
— SIMULATED TIME TO CLEAR ALL EXPOSED COHORTS
— TIME TO FIRST DISCRETE LOG — ZCASH SOUNDNESS GATE
No CRQC exists today. The best demonstrated error‑corrected logical qubit count is about 70 (IBM & University of Chicago, vendor‑reported with preprint, 30 Jul 2026); breaking secp256k1 needs 1,200–1,450 logical qubits and 70–90 million Toffoli gates. That is a gap of roughly 17–21× in logical qubits, plus many orders of magnitude in sustained gate count. Every timeline here is an assumption you control.

01  Pre‑state

Nothing is wrong. That is the point. Every byte below is being written to a permanent, public, replicated record right now, under cryptography that is expected to be readable later.

The premise everything else rests on

Harvest now, decrypt later

A future adversary does not need to be present at the moment of the transaction. Ciphertexts and public keys are archived by every full node, permanently. The only variable is when they become readable.

Shielded note ciphertexts on chain† 0 —
Bitcoin blocks added since you opened this page† 0 protocol target 10 min · median 12 min
Session elapsed 00:00:00 retention: indefinite

BITCOIN

NOMINAL
SECP256K1 · ECDSA + SCHNORR
Public key already visible on chain — —
Value at that exposure — —
SignatureECDSA over secp256k1. Taproot / P2TR uses Schnorr over the same curve. BIP‑340, BIP‑341
Key exposureP2PKH / P2WPKH hash the public key, exposing it only at spend. P2PK, bare multisig and P2TR publish the key directly.
TaprootA P2TR output publishes a 32‑byte x‑only public key in the scriptPubKey. A funded, never‑spent Taproot output is exposed from day one. This is the one exposure category that grows with adoption instead of decaying.
Spend windowPublic key sits in the mempool for the confirmation interval: protocol target 10 minutes, median 12 minutes measured.
Structural vs operational. Structural exposure is written into the script type. Operational exposure is behaviour: address reuse, partial spends, custody patterns. That is the part the published estimates disagree about.
Glassnode's split, and where Satoshi's coins sit
—

ZCASH

NOMINAL
GROTH16 / BLS12‑381 · HALO 2 / PALLAS‑VESTA
Shielded supply — —
Cumulative shielded transactions* — ciphertexts on chain, permanently
SaplingGroth16 proving system over the BLS12‑381 pairing. Keys, commitments and note encryption live on Jubjub. spec 5.4.9.2, 5.4.9.3, 5.4.10.2
OrchardHalo 2 proving system over Vesta; addresses, keys and note encryption on Pallas. spec 5.4.9.6, 5.4.10.3
Note encryptionSapling: Diffie‑Hellman with cofactor multiplication on Jubjub (5.4.5.3). Orchard: Diffie‑Hellman on Pallas (5.4.5.5). Sprout, legacy only: Curve25519 (5.4.5.1). Symmetric layer ChaCha20‑Poly1305, KDF BLAKE2b‑256.
On chainThe ephemeral public key epk is a plaintext field of every Output and Action description. spec 7.4
TransparentTransparent‑pool ZEC carries the same ECDSA exposure as Bitcoin.
Genesis 28 Oct 2016. Every note ciphertext written since then is still on chain and still public.

ABELIAN

NOMINAL
MODULE‑SIS / MODULE‑LWE · PQRINGCTX
Construction and assumption risk rather than a break on day one.
Implementationpqringct / pqringctx: the post‑quantum ring confidential transaction scheme and its enhanced successor. Written in Go.
Hard problemsModule‑SIS and Module‑LWE; LWE / Ring‑LWE. whitepaper 3.2.1
ConstructionLattice‑based signature with key derivation (3.2.2), lattice‑based additively homomorphic commitments (3.2.3), lattice‑based linkable ring signature (3.2.5).
PrivacyOne‑time (stealth) addresses hide the receiver; linkable ring signatures hide which input was spent among a larger set; commitments hide the amount. Design inspired by CryptoNote.
ConsensusDual PoW since the Aconcagua fork at block 464,000. ABEL‑ETHash (Ethash variant, GPU) and ABEL‑Nakamoto (Bitcoin PoW variant, ASIC/FPGA).
Cost of the defence
—
ECDSA / Ed25519
—
—
ML‑DSA‑44
Single control 00:00:00

This simulates the day a cryptographically relevant quantum computer is operational and pointed at a chain — not a discovery, a breach or a disclosure. Nothing here has happened, and nothing here is predicted for any particular date.

Machine assumed—
Work per key—
Wall clock per key—
Google Quantum AI et al., “Securing Elliptic Curve Cryptocurrencies against Quantum Vulnerabilities”, arXiv:2603.28846 (30 Mar 2026). The paper states these figures apply directly to ECDLP on secp256k1.
Time compression —
Rate —
non-linear clock — compression climbs across the run
Simulated time since Q‑Day T+00:00:00
Q‑DAY——
Live walkthrough space = play / pause  ·  ← → = step

BITCOIN

NOMINAL
SHOR → ECDLP ON SECP256K1
Exposed BTC recovered 0
$0 —
—
Exposed‑key cohorts
——
Second wave · spends in flight
Spends taken in flight* 0 —
Attacker margin — against the 10‑minute confirmation target
A spend publishes the public key, then sits unconfirmed. The attacker needs —; the network needs 10 minutes. The amber line is the attacker's finish.
Which rate the second wave uses
The paper's optimised on‑spend path, published for exactly this scenario, or the selected sweep rate if that is faster.
Per‑entity exposure
EntityBalance exposedState
Glassnode, “Measuring Bitcoin's Quantum‑Exposed Supply”, 20 May 2026. Percentages are share of each entity's own holdings with the public key already visible on chain, not share of total supply.

ZCASH

NOMINAL
TWO SEPARATE FAILURES
i · Transparent pool
Transparent ZEC — —
Status — same ECDSA mechanism as Bitcoin
* Modelled as tracking the Bitcoin key‑recovery front. A modelling choice, not a measurement.
Why it is modelled, not measured
The exposure share for transparent ZEC outputs is not separately published. The mechanism is not in doubt: transparent ZEC is ECDSA over secp256k1, exactly like Bitcoin.
ii · Soundness
iii · Retroactive confidentiality
Transactions retroactively decrypted* 0 —
Decryption front —
28 OCT 2016TODAY
DATESENDERRECIPIENTAMOUNT
⚠ Feed rows are illustrative placeholders, not chain data. The counter and the date range are from CONFIG.
“the note encryption for Ironwood, Orchard, Sapling, and Sprout pools is subject to ‘Harvest Now, Decrypt Later’ attacks: that is, a discrete‑logarithm‑breaking attack can be attempted as long as an adversary has both the ciphertext (which is published on the block chain) and the receiver address.” ZIP 2005 · zips.z.cash/zip-2005
Condition, stated plainly. The break is conditional on the adversary knowing or guessing the recipient's shielded address (pk_d), which is not published on chain. ZIP 2005 and ECC engineer Sean Bowe both state this explicitly. Ciphertexts are permanently public. Any shielded note sent to an address an adversary knows — or later learns — becomes decryptable retroactively once discrete logs fall. No future upgrade can undo that for notes already on chain. Published donation addresses, exchange addresses, and any address ever given to a counterparty fall inside that gate.
“the privacy of Zcash transactions may need to be maintained far into the future, and upgrading the encryption algorithm would not prevent a future adversary from attempting to decrypt ciphertexts encrypted before the upgrade.” Zcash protocol specification · section 8.7
iv · Ironwood — NU6.3
Migrated to quantum‑recoverable notes — —
Privacy of those same notes — unchanged by the upgrade
“It does not by itself make the protocol secure against quantum adversaries” — and addresses “Balance, Spend authentication, Spendability only — The situation with respect to Privacy is unchanged for any pool.” ZIP 2005, on NU6.3 “Ironwood” · activated 28 July 2026 at block 3,428,143 (ZIP 258)
Funds recovered. History still open. There is no ZIP addressing retroactive decryption of already‑published ciphertexts, and by construction there cannot be one; ZIP 2005 says only that changes are under consideration “for future transfers”.

ABELIAN

NOMINAL
LATTICE · NOT A PERIOD‑FINDING PROBLEM
Attack 1 · Shor
Attack 2 · Grover against proof‑of‑work
Effective PoW security — —
Verdict — degraded, not broken
“quantum searching cannot be parallelized better than by assigning different parts of the search space to independent quantum computers.” Zalka, Phys. Rev. A 60, 2746 (1999) · arXiv:quant-ph/9711070
Aggarwal et al. conclude proof‑of‑work is “relatively resistant to substantial speedup by quantum computers”: ASIC clock speeds dwarf projected quantum gate speeds. The quantum threat to Bitcoin is to signatures, not to mining.
NIST on the practicality of Grover
“Grover's algorithm requires a long‑running serial computation, which is difficult to implement in practice. In a realistic attack, one has to run many smaller instances of the algorithm in parallel, which makes the quantum speedup less dramatic.” NIST PQC Call for Proposals, Dec 2016, pp. 17–18 · the MAXDEPTH discussion · Aggarwal et al., Ledger vol. 3 (2018) · arXiv:1710.10377
Third‑party assessment, and standards language
A third‑party scorecard rates Abelian QRI 55/100, “Migration Stage 3 (Piloted)”. Treat that site as third‑party of unverified standing: one assessment, not an authority. layerqu.com/chains/abelian/

On standards language. Abelian's defensible claim is that it uses constructions inspired by NIST‑standardised algorithm families, not that it is NIST‑anything. NIST's own disclaimer: “Any mention of commercial products within NIST web pages is for information only; it does not imply recommendation or endorsement by NIST.” FIPS 206 / FN‑DSA is not a published standard; as of April 2026 it remains in NIST's administrative approval chain, not published even in draft.
Cost of the defence
—
ECDSA / Ed25519 signature
—
—
ML‑DSA‑44 signature
Post‑quantum security bought with bandwidth and state size. A real cost, paid in every block, forever.
* Modelling parameter with no published source, marked [VERIFY] in CONFIG: cohort quantisation, decryption throughput, mempool sample rate, transparent‑ZEC exposure share.   † Derived from an average historical rate, not live data. This page makes no network requests.   Cumulative shielded transaction counts are approximate, from a secondary source.

Q‑DAY — END STATE

What a cryptographically relevant quantum computer does to three base layers. Funds and privacy fail separately, and they do not recover the same way. Nothing here has happened; this is a model of published cryptographic properties under an assumed capability.

Q‑DAY ASSUMED  —
KEY RECOVERY  —
PARALLEL CRQCs  —
BTC EXPOSURE  —
BITCOINECDSA + Schnorr / secp256k1
ZCASHGroth16 / BLS12‑381 · Halo 2 / Pallas‑Vesta
ABELIANModule‑SIS / Module‑LWE lattice
Funds at riskCan an adversary move coins that are not theirs?
Privacy at riskCan an adversary read what was already written?
NOT A FORECAST. No CRQC exists today; the best demonstrated error‑corrected logical qubit count is about 70 against the 1,200–1,450 required. Q‑Day year, recovery rate and exposure estimate are user‑set assumptions. Chain‑layer cryptography only: what protects the coins and the privacy on each base layer. Nothing above the base layer is modelled.

Why the two rows differ. Funds can be repaired going forward. A protocol change and a migration bound the loss, and NU6.3 has already begun: — of shielded value has migrated. Published ciphertexts cannot be repaired. Conditional on the adversary knowing the recipient address, that history opens retroactively. No upgrade reaches backwards.

Model, stated. The cascade quantises the selected exposed supply into — equal cohorts, one recovered key per cohort, at the selected rate divided by the parallel‑CRQC multiplier. Cohort quantisation is presentational; real exposure is not uniformly distributed. Values marked * are modelling parameters flagged [VERIFY] in CONFIG. Values marked † derive from average historical rates, not live data.