The International Organization of Securities Commissions (IOSCO), the body whose members regulate more than 99% of the world’s financial markets, published on Thursday 19 November 2026 a 63-page final report on quantum computing and financial markets. For crypto readers the core is a four-page section on public blockchains. It says the move to quantum-safe signatures may be especially hard for a blockchain: the new signatures are roughly 35 to 40 times larger, the change needs a hard fork that thousands of independent validators have to adopt of their own accord, and coins behind already exposed public keys could stay at risk for as long as nobody moves them.
No machine can do this today. The report says so in its first paragraph, and IOSCO Board Chair Jean-Paul Servais repeated it in the media release issued from Madrid the same day (IOSCO/MR/26/2026), before adding: “Preparedness is a collective challenge for the whole ecosystem.”
The document is an analysis, not a rulebook. It sets no deadline and puts no obligation on any chain, exchange or wallet provider. What it gives is the global securities standard setter’s account of where blockchains stand in the post-quantum migration, with figures. One of those figures, a 90% loss of throughput on a Solana test network, is weaker than it looks once the footnote is followed.
What the IOSCO quantum report says about blockchains
The document (reference FR/07/26) was written by a Quantum Computing Working Group that IOSCO’s Fintech Task Force set up in 2026. It draws on a survey of task force members run in March 2026, on roundtables held in Tokyo, New York, Zurich and Mumbai in 2025 and 2026, and on a review of published research. Chapter 4 covers cryptographic weak points, first in banks and market infrastructures, then in public blockchains (pages 38 to 41).
Its starting point is in the executive summary: migration “may be particularly challenging for blockchain systems because cryptography is embedded in their core architecture” and any change needs coordinated action across a decentralised network. A layer-1 chain writes its cryptography into the transaction format, the account structure, the consensus process and the validation rules, the report says. Change the signature scheme and all four change.
Ethereum, Solana and Tron are named as examples of layer-1 networks that rely on elliptic-curve cryptography to identify users and validate transactions. The word Bitcoin does not appear anywhere in the text of the 63 pages, although the passages on address models and abandoned coins describe its situation closely. Many of the same cryptographic risks apply to permissioned and private ledgers, IOSCO adds; what differs is the governance and migration arrangements open to their participants.
A signature of 2,420 bytes where 64 are enough today
Elliptic-curve signatures are small, “typically 64 bytes per signature” in the report’s words. The signature scheme that the US National Institute of Standards and Technology (NIST) standardised for the post-quantum period, ML-DSA, is far heavier. IOSCO takes its lightest version, ML-DSA-44, at about 2,420 bytes per signature, and calls that “roughly 35 to 40 times” the size of common elliptic-curve signatures.
The figure matches the standard. Table 2 of NIST FIPS 204, published on 13 August 2024, gives the sizes below. The last column is arithmetic on those figures: each signature size divided by the 64 bytes IOSCO uses as its baseline, so 2,420 divided by 64 gives 37.8.
| ML-DSA parameter set | Private key (bytes) | Public key (bytes) | Signature (bytes) | Signature size against 64 bytes |
|---|---|---|---|---|
| ML-DSA-44 | 2,560 | 1,312 | 2,420 | 37.8 times |
| ML-DSA-65 | 4,032 | 1,952 | 3,309 | 51.7 times |
| ML-DSA-87 | 4,896 | 2,592 | 4,627 | 72.3 times |
The comparison is between one signature and another: a single ML-DSA-44 signature against a single 64-byte elliptic-curve signature. It does not measure a whole transaction, and it says nothing yet about speed. The 35 to 40 times also applies only to the smallest of the three parameter sets; the two stronger ones come to about 52 and 72 times the baseline. The public key grows as well, to 1,312 bytes in the lightest set.
Size matters on a blockchain because every node must receive, check and store each signature. The report lists the effects: throughput, bandwidth, storage and the time blocks take to spread across the network. With block size and node bandwidth capped, bigger signatures mean fewer transactions per block, which the report links to congestion, heavier storage for node operators and higher fees. A concrete case sits on the XRP Ledger’s roadmap: a draft amendment built on the same ML-DSA-44 scheme, under which quantum-proof XRP Ledger accounts would add about 3.6 KB to every payment that uses them.
The 90% Solana figure rests on one press report
Then comes the figure that CoinDesk and other crypto outlets repeated on Thursday. Evidence from “Solana’s quantum-resistant testnet”, the report says on page 39, showed “throughput degradation of approximately 90%” when elliptic-curve signatures were swapped for post-quantum ones.
The footnote for that sentence does not point to a benchmark or a technical paper. It cites a CoinDesk article of 4 April 2026 by Margaux Nijkerk. That article covered a test environment that the cryptography firm Project Eleven set up in work with the Solana Foundation. Its sentence on speed says a version of Solana using the new cryptography “ran about 90% slower than it does today, Pruden said”. Pruden is Alex Pruden, Project Eleven’s chief executive, so the figure is reported speech from one interview. The article names no signature scheme for the test and gives no hardware, workload or baseline in transactions per second. It puts the size of the new signatures at “roughly 20 to 40 times larger”, again on Pruden’s word, a wider range than IOSCO’s.
So the 90% describes one experimental version of Solana, as stated by the head of the company that ran the test. “Slower” in the article became “throughput degradation” in the report. It is not a measured property of ML-DSA, and the report does not say the test used ML-DSA-44. Neither IOSCO nor the April article gives fee data. The direction is not in doubt, since larger signatures cost capacity. The magnitude has not been published in a form anyone can reproduce.

A hard fork, thousands of validators and nobody to order the upgrade
The second obstacle is governance. A centrally run financial system can ship a security patch to every machine overnight, the report notes. A layer-1 network depends on “the independent, voluntary coordination of thousands of geographically dispersed validators and node operators” to move at the same time. Because the signature scheme sits in the transaction structure and the consensus rules, replacing it requires a hard fork: a change of rules that old software rejects.
No mechanism exists to force operators to install new software quickly. IOSCO expects any post-quantum migration of a public chain to be “slow and fragmented” for that reason. It makes the same point about crypto-agility, the ability to swap algorithms without rebuilding a system, which it treats as a main defence for banks: on a public blockchain it means changing protocol rules and getting decentralised participants to adopt them.
A further consequence matters after a fork succeeds, and IOSCO spells it out. Accounts, smart contracts and other arrangements still controlled by old keys remain exposed until their owners migrate them. An attacker with a capable machine could derive private keys from exposed public keys and forge withdrawal, transaction or governance signatures “even after other parts of the network have adopted PQC” (post-quantum cryptography). In IOSCO’s words, “the security of the network may continue to be constrained by the least-migrated components”.
Dormant wallets: exposed public keys that no owner will move
How exposed a coin is depends on when its public key becomes visible. Our guide to how private and public keys work covers the basics: the public key is derived from the private key, and a quantum computer running the right algorithm could, in theory, reverse that step for elliptic curves.
IOSCO separates two attacks. In some systems the public key stays hidden until an address is spent from. Once a transaction is broadcast, the key is visible, and an attacker would have to derive the private key and push a conflicting transaction before the original confirms. IOSCO calls this an “on-spend” attack, limited by the confirmation window. An “at-rest” attack targets public keys that stay exposed for long periods, with no clock running. Account-based chains, where public keys are easier to recover, “may present a broader population of exposed accounts”.
That is why the report says a chain’s quantum risk cannot be read from its signature scheme alone. Three other things count: whether and when public keys are revealed, whether addresses are reused, and how long a transaction takes to become final.
The sentence on abandoned coins is short and offers no remedy. Dormant or abandoned assets tied to exposed public keys “may present particular migration challenges” where holders cannot be identified or cannot transfer the assets to quantum-resistant addresses. The report does not estimate how many coins are in that position on any chain, and it takes no side on whether a network should freeze them, leave them or do something else.
Layer 2s inherit the weak points of the chain below
Rollups get their own subsection. A layer 2 may use different cryptography from its host chain, yet its security still runs through layer-1 settlement contracts, bridges, proof verifiers, data-availability arrangements and the keys that approve upgrades. A quantum weakness in any of them undermines the rollup, the report says, even if the rollup’s own components are quantum-resistant.
For optimistic rollups, the concern is the challenge period, typically seven days on Ethereum-based systems, during which an honest validator must spot and contest an invalid state. Forged signatures could be used to submit fraudulent transaction sequences or to censor honest validators during that window. For zero-knowledge rollups, the answer depends on the proof system. Many SNARK systems rely on elliptic curves and would need to be modified or replaced. STARK systems rely mainly on hash functions, which resist quantum attacks if the parameters are chosen properly. The report names zkSync and Starknet as examples of the category. Starknet’s token holders were asked in October whether the network should research becoming an independent post-quantum layer 1, in a non-binding poll that closed at 93.5% Yes on 22 October.
One more risk is specific to ledgers. Blockchain data is copied across every participant and kept permanently, so anything encrypted with vulnerable cryptography and stored on or around a chain can be collected now and opened later. A later migration cannot recall those copies. IOSCO is careful about the limit of this “harvest now, decrypt later” risk: it concerns confidentiality of historical data, not ownership or the validity of transactions already recorded. Rewriting past ledger records would also require beating the chain’s consensus and finality rules, the report notes in its market integrity chapter.
When could such a machine exist? The report gives two answers
The timeline is where headlines and text part ways. On page 32, the report says the timing “remains uncertain” and that conservative estimates assume relevant quantum computers could emerge in the 2030 to 2035 period. On page 60, under the heading “How Imminent and Serious is the Q-Day Threat?”, it reports that “several industry participants argued that organizations should prepare for the possibility that relevant capabilities emerge within the next five years”, possibly sped up by AI systems.
The second sentence is a planning assumption voiced by some of the people IOSCO consulted. It is conditional, and it is not a forecast by IOSCO that any blockchain’s cryptography will be broken by 2031. The sentence itself opens with the reminder that the exact timing remains uncertain. On the trend the page is blunter in its own voice, saying the timeline for quantum attacks “has compressed materially”, and it adds two warnings: an attacker would not need to own a quantum computer, only to get access to one, and institutions should plan for a cryptographic “zero-day” that arrives with little or no warning.
Hardware estimates are moving, which is part of why the dates are argued over. A recent example is the Alice & Bob preprint projecting that under 30,000 cat qubits could solve Bitcoin’s curve in about an hour, a calculation that has not been peer reviewed and rests on assumptions no existing machine meets.
What IOSCO asks of firms and regulators
The message to regulated firms is to start now. Migration to post-quantum cryptography is expected to take many years and amounts to a system-wide redesign “rather than a straightforward software update”, the introduction says. The practices the report sees emerging across jurisdictions are an inventory of where cryptography is used, a governance and transition plan, crypto-agility, testing, and engagement with outside suppliers such as software, cloud and infrastructure providers.
The survey shows how far there is to go. Most firms are still at the stage of awareness, training, inventories and first plans; migration testing and deployment “remain limited”. The report gives a name to the habit of waiting for clearer rules or more mature technology: “quantum-procrastination”. Regulators are not far ahead. About 55% of members told the survey they had set up cross-functional working groups or task forces on quantum inside their own authority, and about 25% a dedicated team (page 53). The members in question are those of the Fintech Task Force that answered, and the sample is small: the charts on pages 51 and 53 show between 11 and 20 answers per question.
For securities regulators there is a second instruction, and it touches crypto marketing directly. The report warns of “quantum washing”, where institutions overstate the role, maturity or performance of quantum computing in their products. The release compares it with earlier trends such as AI and blockchain, which bad actors used to mislead enthusiastic investors, and says regulators should review disclosures and act swiftly. The same question of proof hangs over tokens and wallets sold as quantum-safe. In October, nine crypto firms launched the Quantum Alliance to define and test what “post-quantum ready” means; most of its members sell post-quantum products themselves.

IOSCO was not alone on Thursday. The Bank of England, the Prudential Regulation Authority and the Financial Conduct Authority jointly published effective practices on quantum-related cyber risk for firms and financial market infrastructures. The paper states that no cryptographically relevant quantum computer is known to exist at present, describes boards agreeing a documented migration plan with milestones and dates and firms keeping a cryptographic inventory that covers third-party dependencies, and says the practices are not part of the regulators’ formal expectations. It sets no deadline of its own and does not mention blockchains, distributed ledgers or crypto-assets.
What the report leaves open for crypto
IOSCO describes the blockchain problem and stops there. Its chart of national and supranational migration timelines (Figure 3) comes with a note that those initiatives “do not specifically address migration timelines for public blockchain systems”. The report does not propose one. It does not say who, in a network without an operator, would be answerable for a late migration, or what a regulated custodian or exchange should do about assets held on a chain that has not moved. It contains no numbered recommendations at all, for chains or for anyone else; its closing sections are headed “Considerations for Further Engagement” and “Conclusion”.
It also contains no measured data of its own on blockchain performance. The size figures come from the NIST standard and an academic paper; the Solana figure comes from the press. Published benchmarks from the chains themselves, with the scheme, the hardware and the baseline stated, would settle how large the cost really is.
IOSCO says in its release that quantum stays a priority for 2027, with a focus on raising awareness across the financial sector together with market participants and other international bodies. The report presents itself as a basis for that further work, on post-quantum readiness, capacity-building and international cooperation. No follow-up publication has been dated.
