TL;DR: On November 16, 2026, Ethereum co-founder Vitalik Buterin changed his mind about leanPCS, a design for making Ethereum's signatures quantum-safe, calling himself "a total bull" on it. The problem is real: Ethereum accounts sign with ECDSA, and a powerful enough quantum computer running Shor's algorithm could derive a private key from a public key already visible on-chain. No such computer exists today — the best machines have roughly 1,000 physical qubits, while breaking secp256k1 needs about 1,200 logical qubits — but 50–65% of all ETH sits in accounts whose public keys are already exposed, and the fix, EIP-8288, is still a draft with no activation date. For ordinary users, nothing needs to change today.
Key Takeaways:
- The threat is to signatures, not to mining or hashing. Shor's algorithm breaks ECDSA and BLS; Grover's algorithm only gives a quadratic speedup against SHA-256 and proof-of-work, which difficulty adjustment absorbs.
- Your public key becomes visible the first time your address sends a transaction. An address that has only ever received funds is still protected, because only a hash of its public key is on-chain.
- An estimated 50–65% of all ETH sits in accounts with revealed public keys, including the Beacon Deposit Contract's 86 million ETH and dormant wallets whose owners have lost their keys.
- The fix is a size problem, not a math problem. Quantum-safe signatures run 1,000–6,000 bytes against ECDSA's 65, and verifying one directly costs about 150,000 gas against about 3,000. EIP-8288 moves verification into the mempool and aggregates proofs so blocks don't carry the bulk.
- Timelines are ranges, not dates. A June 2026 Monte Carlo forecast puts a cryptographically relevant quantum computer at about a one-in-six chance by 2035, about 30% by 2040, and about 60% by 2050. The Ethereum Foundation targets full post-quantum protection by 2029.
Last updated: November 2026
Quantum computing has been crypto's background worry for a decade: always coming, never quite here. On November 16, 2026, it moved up the agenda, when Vitalik Buterin wrote on X that he had changed his mind about leanPCS, one of three routes to quantum-safe Ethereum signatures under draft proposal EIP-8288: "Over the past few weeks I've gone from thinking leanPCS is total bull to being a total bull."
It is a notable sentence from the person whose opinion carries the most weight in Ethereum's technical direction — and an easy one to misread. Buterin did not announce that Ethereum is quantum-safe, or that a quantum computer is about to break your wallet. He backed a design choice inside a draft with no activation date, no finished reference implementation, and a dependency on another draft. This article explains the threat, whose ETH is exposed, when it could arrive, what the fix looks like, and what you should — and should not — do about it.
What Happened on November 16
EIP-8288, "In-mempool signature and proof aggregation," was drafted by Buterin and Thomas Coratger in June 2026. It exists because quantum-safe signatures are far larger than ECDSA's 65 bytes. Buterin's summary: "This is unsustainable to put on chain, and 8288 fixes this." The proposal moves verification into the mempool, where nodes fold signatures and proofs into recursive STARKs so a block carries one aggregate proof; only the statements those proofs vouch for go on chain.
Buterin's thread compared three things a user could submit under that design: leanSPHINCS, a single hash-based scheme built on SPHINCS+ (NIST SLH-DSA); leanSTARK, a proof of any computation, once Ethereum agrees on an instruction set architecture; and leanPCS, one shared polynomial-commitment checker that different proof systems can use. Buterin now favors leanPCS — it only needs a mathematical field chosen, and would verify a STARK for roughly 20,000 to 40,000 gas — calling it a path to "the cryptographic proof singularity" arriving "much faster than any of us expected."
How ECDSA Works — and Why a Quantum Computer Breaks It
Every Ethereum account starts with a private key — a large secret number. Elliptic-curve math derives a public key from it, and your address is the last 20 bytes of the Keccak-256 hash of that public key. The hash is a one-way shield: the address reveals nothing about the public key behind it, or the private key.
When you send a transaction, you sign it with your private key. The signature proves you hold the key without revealing it — but it also lets anyone recover your public key from it, which wallets and nodes do with a function called ecrecover. So the shield has a hole: the first time an address sends anything, its public key becomes public, permanently, on a ledger that never forgets.
That is where Shor's algorithm comes in. Published in 1994, it solves the discrete logarithm problem — the mathematical foundation ECDSA rests on — in polynomial time on a quantum computer. Given a public key, a machine with enough error-corrected qubits could compute the private key behind it directly; a classical computer would need billions of years. The attack needs no seed phrase, password, or device access — only a public key already on-chain and a machine that does not exist yet.
Two confusions are worth clearing up. First, Grover's algorithm, the other famous quantum algorithm, is not the same threat: it gives only a quadratic speedup against hash functions and proof-of-work mining, which difficulty adjustment absorbs. As the June 2026 paper "Quantum Horizon" puts it, Shor's algorithm breaks the signatures; Grover's does not meaningfully threaten mining. Second, there is a threat that needs no quantum hardware at all: harvest now, decrypt later. Every exposed public key is already recorded on-chain, so an adversary can collect them today and attack them the day a capable machine exists — which is why the migration clock starts before the threat arrives.
Which ETH Is Actually at Risk?
Not all ETH is equally exposed. The dividing line is simple: has the address ever sent a transaction? If not, its public key is still hidden behind the address hash. If yes, the public key is on record — and the exposed group is larger.
| Address type | Status | Why |
|---|---|---|
| Never-spent address (only ever received) | Safe for now | Public key hidden behind the Keccak-256 address hash |
| Active externally owned account (EOA) that has sent transactions | Exposed | Public key recoverable from any past signature via ecrecover |
| Beacon Deposit Contract (86M+ ETH staked) | Exposed | Validator BLS signatures break under Shor's algorithm too |
| Smart contracts with exposed admin keys | Exposed | At least 70 major contracts, about 2.5M ETH; over $200B in stablecoins and tokenized assets sit behind exposed admin keys |
| Dormant or lost wallets | Exposed — and unfixable by the owner | Nobody holds the key to migrate them — e.g. 250,000 ETH from the 2014 pre-sale, wallet file and password lost |
How much ETH is that? The "Quantum Horizon" preprint estimates 50–65% of all ETH — about 61–79 million ETH — sits in accounts with revealed public keys, calling 55–60% the most defensible figure. Treat it as an estimate — a preprint, not peer-reviewed — though Deloitte in 2021 put it above 65%. The top 1,000 wallets alone hold about 20.5 million ETH with exposed public keys, and a March 2026 paper from Google Quantum AI, the Ethereum Foundation, and Stanford named at least five attack paths — wallets, smart contracts, staking, layer 2s, and data availability — with combined exposure above $100 billion.
The dormant wallets are the hardest case. The Foundation's roadmap FAQ, published March 24, 2026, ranks the exposed surfaces in order: user accounts first, then high-value operational keys at exchanges, bridges, and custody hot wallets, then multisigs and validator keys. For coins whose owners are gone, it names two outcomes when the risk window arrives: do nothing, or freeze the vulnerable coins. It estimates dormant exposure at roughly 0.1% of supply — small in percentage terms, but coins nobody can move to safety.
When Could This Actually Happen?
Short answer: nobody knows, and anyone who gives you a single year is selling certainty they do not have — the honest answer is a probability distribution, not a date.
| Source | Estimate |
|---|---|
| "Quantum Horizon" Monte Carlo forecast (June 2026) | About a one-in-six chance by 2035; about 30% by 2040; about 60% by 2050; 80% range roughly 2032–2060 |
| Global Risk Institute 2025 expert survey | 28–49% within 10 years; 51–70% within 15 years |
| Ethereum Foundation target (Lean Ethereum roadmap) | Full post-quantum protection by 2029; execution-layer migration takes additional years |
Breaking secp256k1 needs about 1,200–2,330 logical qubits, or roughly 0.5–320 million physical qubits. The best machines in 2026 have roughly 1,000–1,200 physical qubits and at most about a hundred logical ones: a gap of about 400–500 times even against the most aggressive estimate. But Google's March 2026 estimates cut the requirement roughly 20-fold, finding fewer than 500,000 physical qubits could break 256-bit elliptic-curve keys in a few minutes. That is why the Foundation's post-quantum team, while not believing a capable machine is imminent, is building as if the 2030s matter.
The "Quantum Horizon" authors draw one conclusion that matters more than any date: governance delay, not technical difficulty, is the main constraint. A migration started in 2026 finishes around 2029–2031, beating even an optimistic 2035 arrival. The cryptography is ready — NIST finalized four post-quantum algorithms in August 2024. The risk is that a decentralized network takes years to agree, ship, and get millions of users to move, and starts late.
What a Quantum-Safe Fix Looks Like
In August 2024, NIST finalized four post-quantum algorithms: ML-DSA (Dilithium), ML-KEM (Kyber, for encryption), SLH-DSA (SPHINCS+, hash-based), and FN-DSA (Falcon). For Ethereum, the relevant families are lattice-based ML-DSA and hash-based SLH-DSA; no known quantum algorithm breaks either. The hard part is not the math. It is the size.
An ECDSA signature is 65 bytes and costs about 3,000 gas to verify. A hash-based quantum-safe signature runs 1,000–6,000 bytes, and a STARK proof can reach about 128 kilobytes; verifying one costs about 150,000 gas — 50 times the ECDSA figure — before the data is stored. If every transaction carried a signature 15 to 92 times larger, blocks would choke on the data alone. That is the problem EIP-8288 exists to solve.
| Design | What it is | Readiness | Cost |
|---|---|---|---|
| leanSPHINCS | One hash-based scheme (SPHINCS+ / NIST SLH-DSA) | Closest to ready; a contract can already check a variant for about 150,000 gas | About 3,000 gas per claim plus 30,000 gas header share |
| leanSTARK | A proof of any computation, not just one signature type | Needs an instruction set architecture agreed first — "likely to take time" | General, but the most work to standardize |
| leanPCS | One shared polynomial-commitment checker for all proof systems | Not yet in the EIP text; only needs a field chosen — Buterin's pick | STARK verification for roughly 20,000–40,000 gas |
Aggregation is only half the migration. The other half is letting individual wallets switch signature schemes without waiting for the whole network to change at once — that is what account abstraction is for. EIP-8141 would introduce native account abstraction, so your account migrates when you migrate it. The rails already exist in part: EIP-4337 smart wallets number more than 26 million, with over 170 million UserOperations processed. The Foundation's phased roadmap covers the layers in order: consensus BLS signatures, data availability with KZG commitments, wallet ECDSA signatures, and application proofs such as Groth16.
None of this is shipped yet. EIP-8288 is a draft whose reference implementation is unfinished; leanPCS, the design Buterin now favors, is not in the written specification at all. The proposal depends on EIP-8141's frame transactions, slated for the Hegotá upgrade but themselves still a draft. There is no mainnet activation date and no assigned hard fork. What changed on November 16 is not the timeline — it is which route the most influential voice in Ethereum now thinks gets there fastest.
Bitcoin Has the Same Problem
Ethereum is not special here. Bitcoin uses the same curve, secp256k1, and the same signature scheme, ECDSA. A quantum computer that breaks Ethereum's signatures breaks Bitcoin's too, and Bitcoin has an additional exposure Ethereum does not: its earliest coins. Satoshi-era pay-to-public-key (P2PK) outputs expose their public keys directly, with no address hash in between at all, and the Foundation estimates Bitcoin's dormant exposure at roughly 5% of supply, against about 0.1% for Ethereum. Buterin has weighed in on such threats before: in September 2026 he dismissed the claim that AI could break Bitcoin's SHA-256 hashing, while noting quantum computing is a separate, longer-horizon question.
Bitcoin's answer takes a different shape. BIP-360 introduces Pay-to-Merkle-Root addresses, which strip out the quantum-vulnerable key-path spend Taproot currently supports, replacing it with a script-path structure compatible with post-quantum signatures. BIP-361 handles the legacy coins, and a co-author has estimated a full migration could take around seven years from rough consensus. The contrast with Ethereum is instructive: Ethereum's account-abstraction path lets wallets migrate one by one, while Bitcoin's migration needs users to move funds to new addresses — and both take years.
What Should You Do Today?
The short answer, from ethereum.org itself, is: nothing, yet. No quantum computer today can break wallet keys, and no standard Ethereum account is quantum-safe — but there is also no machine to be safe from. The correct posture is the boring one: keep up the security hygiene that already matters, and watch for the migration.
Do not panic-move your ETH to a fresh address. A new address is as protected as an address can be — until it spends, when its public key is exposed like the old one's. Moving funds adds no quantum protection; it adds transaction history. What protects you today is discipline against the threats that already exist. Wallet drainers, not quantum computers, are what actually empty wallets in 2026: learn how wallet drainers work and how to check and revoke token approvals, and clean up stale approvals with Etherscan's checker or Revoke.cash. A Permit signature granted to a dead project in 2024 is a live attack surface in a way a theoretical quantum computer is not.
Keep your seed phrase backed up offline, never typed into a website. If it is ever exposed, move funds to a new wallet immediately — a real emergency, unlike the quantum timeline. Moving wallets is ordinary: importing MetaMask into Rabby takes minutes, and the same pattern — new wallet, new keys, move funds — is what a future quantum-safe migration will look like.
Be skeptical of anyone selling urgency. A wallet, token, or service marketed as "quantum-proof" in 2026 is selling a label, not a standard — no finalized Ethereum post-quantum account standard exists yet. The credible signals are: EIP-8288 moving from draft to review, EIP-8141 landing in the Hegotá upgrade, wallets announcing post-quantum support, and the Foundation's roadmap updates. When those arrive, this article's advice changes. Until then, the quantum threat is a reason to follow Ethereum's roadmap, not a reason to move your coins.
This article is for education only, not financial advice. Quantum computing timelines are estimates, not predictions, and nothing here is a recommendation to buy, sell, or move any asset.
Frequently Asked Questions
Can a quantum computer steal my ETH today?
No. No cryptographically relevant quantum computer exists: the best machines in 2026 have roughly 1,000–1,200 physical qubits, while breaking secp256k1 needs about 1,200 logical qubits — roughly 0.5–320 million physical qubits after error correction. Your ETH is safe today; the risk is a future machine used against public keys already on-chain.
Is my ETH safe if my address has never sent a transaction?
Yes, for now. An address is a hash of its public key, which a hash does not reveal; the key becomes visible only the first time the address signs a transaction. An address that has only ever received funds has no exposed public key — but the protection ends the moment it spends.
What is EIP-8288, and when will it ship?
EIP-8288, "In-mempool signature and proof aggregation," is a draft by Vitalik Buterin and Thomas Coratger (June 3, 2026). It moves verification of large quantum-safe signatures into the mempool and aggregates them, so blocks carry one aggregate proof. There is no activation date, assigned hard fork, or finished reference implementation, and it depends on EIP-8141 frame transactions from the Hegotá upgrade, itself a draft.
Will quantum computers break Bitcoin too?
Yes. Bitcoin uses the same curve (secp256k1) and signature scheme (ECDSA), so the same attack applies. Its earliest coins are more exposed: Satoshi-era pay-to-public-key outputs reveal their public keys directly, and roughly 5% of supply sits in dormant addresses, against about 0.1% for Ethereum. Its migration runs through BIP-360 (Pay-to-Merkle-Root addresses) and BIP-361 (legacy coins), estimated at around seven years from consensus.
Should I move my ETH to a new wallet now?
No. Moving funds to a fresh address adds no quantum protection: the new address is safe only until it first spends, when its public key is exposed like any other. What helps today is ordinary wallet hygiene: keep your seed phrase offline and backed up, and revoke token approvals you no longer use. If a seed phrase is ever exposed, move funds immediately — because of the exposure, not quantum computing.
What happens to lost or dormant ETH if quantum computers arrive?
Nobody can migrate coins whose keys are lost — the hardest question in the roadmap. The Foundation's post-quantum FAQ names two outcomes when it arrives: do nothing, or freeze the vulnerable coins so they cannot be stolen even if their keys are derived. It estimates dormant exposure at roughly 0.1% of ETH supply. No decision has been made, and it is likely to be one of Ethereum's most contested governance questions.
Sources
- Unchained: Vitalik Buterin Says Ethereum's Quantum-Safe Plan Can Ship Sooner With This Design (November 16, 2026)
- AInvest: Vitalik's LeanPCS Flip Is a Bet on Governance Cost, Not a Quantum Timeline (November 2026)
- Bitcoin Foundation: Quantum-Safe Ethereum — Vitalik Buterin Backs leanPCS (November 2026)
- Gershteyn and Alber: Quantum Horizon — An Evaluation of Quantum Computing as a Threat to Bitcoin and Ethereum (arXiv:2606.14484, June 2026)
- Hadamard: Can Quantum Computers Break Ethereum? 2026 Update (October 2026)
- Lime Institutional: How Ethereum, Solana, Hedera, and Canton Are Preparing for the Quantum Threat to Blockchain Cryptography (2026)
- WATS Wallet: Is Your Crypto Wallet Quantum-Safe? Ethereum, Solana and TON Exposure Explained (October 2026)
- BitRss: As Quantum Q-Day Jumps to 2029, Ethereum Faces a New Fight Over What to Do With Coins Left in Old Wallets (November 2026)
- HackerNoon: Quantum Computing Breaks ECDSA — Your ZK-SNARKs Use the Same Math (May 28, 2026)
- The Tech Trends: Quantum-Safe Blockchains — Signature Migration Guide (October 2026)

