NIST IR 8547 treats the move to post-quantum cryptography as a long, uneven engineering programme rather than a single switch-over, and its treatment of digital signatures shows why: the old algorithms are all vulnerable to Shor's Algorithm, the new ones come in several different shapes, and the history of cryptographic migration says the work takes a decade or two.
Executive Summary
- NIST IR 8547 describes NIST's approach to moving from quantum-vulnerable algorithms to post-quantum digital signature algorithms and key-establishment schemes.
- The report supports Executive Order 14412 and OMB Memorandum M-26-15, which direct federal agencies to prioritise migrating their most sensitive systems.
- Historically, standardisation to full integration can take 10 to 20 years, and the harvest now, decrypt later threat model is a main reason for urgency.
- The current signature standards under FIPS 186 (ECDSA, RSA, EdDSA) are vulnerable to Shor's Algorithm; FIPS 204 and 205 specify quantum-resistant replacements.
- Stateful hash-based schemes are quantum-resistant but need careful state handling, so they aren't intended for general use.
- NIST is developing a FALCON-derived FIPS and evaluating further signature candidates.
What the report is, and who it's for #
NIST IR 8547, "Transition to Post-Quantum Cryptography Standards", is a NIST Internal Report published in October 2026 by Dustin Moody, Ray Perlner, Andrew Regenscheid, Angela Robinson and David Cooper of NIST's Computer Security Division. Its author list carries a note worth recording: David Cooper is a former NIST employee, and all work for the publication was done while at NIST. The report was approved by the NIST Editorial Review Board on 2026-10-21, which makes it a very recent document as these things go.
The stated purpose is plain. The report describes NIST's approach to transitioning from quantum-vulnerable cryptographic algorithms to post-quantum digital signature algorithms and key-establishment schemes. It identifies the existing quantum-vulnerable cryptographic standards and the quantum-resistant standards to which information technology products and services will need to transition. That pairing, a list of what's exposed alongside a list of what replaces it, is what makes the report practical rather than merely advisory.
It's also written for a wide readership. The document is intended for a broad audience, including federal agencies, technology providers, standards organisations and Cryptographic Module Validation Program (CMVP) laboratories. And it has an explicit job in policy: it supports the implementation of Executive Order 14412 and OMB Memorandum M-26-15, which direct federal agencies to prioritise the migration of their most sensitive systems to post-quantum cryptography.
If you run a startup that sells into government, or builds the products government buys, that policy link is the reason to read it. The report is intended to encourage engagement with industry, standards organisations and relevant agencies, with the aim of facilitating and accelerating adoption. Engagement with industry is written into the purpose, not bolted on afterwards.
The operative status of the document #
A transition report that told everyone to wait would be useless, so it's worth being precise about how the report positions itself. Federal agencies may use its concepts and methodologies even before the completion of companion publications that are still under development. At the same time, until each companion publication is completed, current requirements, guidelines and procedures, where they exist, remain operative.
Those two statements fit together without strain. You can plan with the new material now; you can't treat the existing rules as withdrawn. NIST also suggests that federal agencies may wish to closely follow the development of the new publications for planning and transition purposes. That's phrased as a suggestion rather than a requirement, and I'd read it as sensible advice for anyone outside government too, though the report itself directs it at agencies.
There are a few housekeeping points that matter for anyone commenting on or building from the document. All comments are subject to release under the Freedom of Information Act (FOIA), so don't submit anything you wouldn't want public. On patents, the position is limited but clear: as of the date of publication, following call(s) for the identification of patent claims, no such claims have been identified to ITL. ITL hasn't undertaken a patent search to identify which, if any, patents may apply, and no representation is made or implied that licences are not required to avoid patent infringement in the use of the publication. In other words, an absence of identified claims isn't a clearance. Anyone shipping an implementation should do their own diligence.
Why the clock matters #
The threat the report is guarding against is stated with appropriate care. Future quantum computing may be able to break current cryptographic algorithms, rendering data and information vulnerable. The verb is "may": the report frames this as a prospect, not a certainty with a date attached. What the report offers in response is post-quantum cryptography (PQC), the term for new cryptographic methods that can protect data from both current conventional computers and the quantum computers of tomorrow.
The urgency doesn't depend on a quantum computer existing today. It comes from two facts that sit side by side. The first is the length of historical migrations: the journey from algorithm standardisation to full integration into information systems can take 10 to 20 years. The second is the harvest now, decrypt later threat, in which adversaries collect encrypted data now with the goal of decrypting it once quantum technology matures. The report states that this threat model is one of the main reasons the transition to PQC is urgent.
Put those together and the logic is uncomfortable. If integration takes 10 to 20 years, and an adversary can collect today's ciphertext for later use, then waiting for the threat to become visible means waiting past the point where the work should have started. I'd argue this is the sharpest argument in the report's framing, and it's built entirely from the two facts above rather than from any forecast of quantum hardware.
It's also worth being clear about what the harvest threat describes. The report defines it as the collection of encrypted data now, for decryption once quantum technology matures. That is a statement about data confidentiality over time, and it's the reason the report treats the transition as something to start rather than something to schedule for later.
A transition with precedent, and without one #
Cryptographic updates aren't new. The report notes that updating cryptographic technology has occurred many times at different scales, such as increasing key sizes or phasing out insecure hash functions and block ciphers. Engineers who've lived through those changes will recognise the pattern.
But the report is careful not to oversell the precedent. While the transition to PQC is without precedent in scale, it benefits from a level of awareness and understanding that previous cryptographic changes did not have. Both halves of that sentence carry weight. The scale is larger than anything before it; the preparation is better than anything before it. NIST also acknowledges directly that the transition will demand substantial effort across diverse applications and infrastructures with specific requirements and constraints. There's no pretence that one migration recipe fits every system.
The shape of the work is also stated. The transition involves adopting new PQC algorithms, and it involves careful deprecation and eventual removal of quantum-vulnerable algorithms that are currently widespread. Adoption is the visible half. Removal is the half that tends to be neglected, because an algorithm that is still accepted anywhere is still a place where the old assumptions live. The word "careful" in the report is doing real work: widespread algorithms can't be switched off by decree.
Cryptographic agility as the organising principle #
If one idea in the report generalises beyond the specific algorithms, it's agility. The report says ensuring a smooth transition requires building long-term cryptographic agility into organisational infrastructure, to allow for seamless updates as post-quantum standards evolve.
That sentence is worth taking at face value. Standards are still evolving: a FALCON-derived FIPS is in development and other signature candidates are under evaluation, as covered below. An organisation that hard-codes one algorithm into its systems will pay again when the field of standards shifts. One that has built the ability to swap algorithms has turned a recurring emergency into a routine change. I'd go further and say agility is the one investment that stays useful whichever way the remaining standards decisions fall, though that judgement is mine rather than a claim from the report. What the report itself says is that agility is required.
NIST also commits to its own side of the bargain. It will continue to monitor developments closely and adjust transition efforts as appropriate based on new technological or operational developments. The plan is explicitly revisable. For organisations that want practical help, the NCCoE Migration to PQC project provides supporting resources and practical guidelines to assist organisations in managing their cryptographic transition.
The report positions itself as the technical framework necessary to meet the accelerated migration timelines for federal high value assets and high impact systems, and as a blueprint for critical infrastructure owners and operators and other organisations. For founders outside government, that second role matters most: the document is meant to be read by people who don't answer to a federal agency.
What digital signatures do, and why they're a distinct problem #
Digital signature algorithms are used to provide identity authentication, integrity authentication, source authentication and support for non-repudiation. They're used in conjunction with hash functions or eXtendable-Output Functions (XOFs) to sign messages of arbitrary length. That second detail is a reminder that a signature scheme is rarely a lone primitive; it's part of a pipeline in which the hash or XOF handles arbitrary-length input.
NIST-approved digital signature algorithms were historically specified in FIPS 186. The current revision of FIPS 186 specifies the Elliptic Curve Digital Signature Algorithm (ECDSA) and adopts the RSA algorithm specified in RFC 8017 (PKCS #1 Version 2.2) and the Edwards-Curve Digital Signature Algorithm (EdDSA) specified in RFC 8032.
And here is the problem in one sentence from the report: these algorithms are vulnerable to Shor's Algorithm on a cryptanalytically relevant quantum computer. Note the precise conditions. The vulnerability is stated as conditional: it applies on a quantum computer that is cryptanalytically relevant. The report doesn't claim such a machine exists today. It claims that these algorithms can't be relied on against one.
For anyone who signs software, issues certificates or authenticates services, that list is the inventory of what has to change. ECDSA, RSA and EdDSA between them cover the algorithms in the current FIPS 186 revision. Nothing in the report suggests any of them gets a pass.
The first replacements: ML-DSA and SLH-DSA #
Two published standards answer the problem. FIPS 204 specifies the Module-Lattice-Based Digital Signature Algorithm (ML-DSA), which is derived from the CRYSTALS-Dilithium submission. FIPS 205 specifies the Stateless Hash-Based Digital Signature Algorithm (SLH-DSA), which is derived from the SPHINCS+ submission. FIPS 204 and 205 each specify quantum-resistant digital signature schemes.
The naming deserves a moment, because it trips people up. The standards carry the ML-DSA and SLH-DSA names; the submissions they derive from were CRYSTALS-Dilithium and SPHINCS+. If you're reading older material or library documentation, you may meet either set of names, and the report's wording tells you how they map: ML-DSA from CRYSTALS-Dilithium, SLH-DSA from SPHINCS+.
Having two standards rather than one is itself informative. The report doesn't spell out the design rationale for the pair, so I won't invent one. What the report does state is that there are two published, quantum-resistant signature schemes, each with its own FIPS, and that they're derived from different submissions.
Stateful hash-based signatures: secure, but demanding #
There's a further family the report discusses: stateful hash-based signature (HBS) schemes. The report states that these schemes are also resistant to attacks by quantum computers. The catch is operational, and the report describes the mechanism clearly.
In a stateful HBS scheme, the private key consists of a large set of one-time signature (OTS) private keys, either stored directly or deterministically generated as needed from a seed. The security of these schemes relies on the signer ensuring that no individual OTS key is ever used to sign more than one message.
Read that carefully, because it moves part of the security burden from mathematics to operations. The scheme's guarantee holds only if the signer tracks which one-time keys have been spent and never reuses one. A signer who loses track of that state, perhaps through a restored backup or a duplicated system, is outside what the security claim covers. The report doesn't enumerate such failure scenarios, and I'm offering them only as an illustration of what "relies on the signer to ensure" means in practice.
Because of this need to maintain state, stateful HBS schemes are not intended for general use but may be useful for applications like software and firmware signing. The report hedges that last clause with "may", and it's right to. Software and firmware signing is named as a possible fit; the report doesn't promise it's the right choice for any given deployment. The sensible reading is that these schemes are a specialised tool, quantum-resistant and usable, but gated by the discipline of state management.
What's still to come on signatures #
The signature picture isn't finished. NIST is developing a FIPS that specifies a digital signature algorithm derived from FALCON as an additional alternative to these standards. The wording matters: it's an addition, an alternative alongside FIPS 204 and 205, not a replacement for them, and it's in development, not published. Anyone planning a migration should treat it as forthcoming rather than available.
Beyond that, NIST is also evaluating other proposed digital signature algorithms for possible standardisation through the Additional Digital Signature Schemes for the NIST Post-Quantum Cryptography Standardization Process. The report frames this with "possible standardisation", so none of those candidates should be treated as a settled outcome. They're under evaluation, and some may never become standards.
This is where the agility argument earns its place. The set of post-quantum signature standards is not settled: FIPS 204 and 205 are published, a FALCON-derived FIPS is in development, and other proposed algorithms are being evaluated for possible standardisation. A system designed around one fixed choice is being designed against a moving target. A system designed to swap algorithms is designed for the actual state of the standards process.
Reading the report as a planning document #
Taken as a whole, NIST IR 8547 is less a technical specification than a statement of intent with a technical spine. The spine is the inventory: FIPS 186's ECDSA, RSA and EdDSA on one side, FIPS 204's ML-DSA and FIPS 205's SLH-DSA on the other, with stateful HBS schemes available for narrower uses and a FALCON-derived standard on the way. The intent is the policy link to Executive Order 14412 and OMB Memorandum M-26-15, and the explicit hope that industry, standards organisations and agencies will engage to accelerate adoption.
Where I'd take a firm view is on the weight of the timing argument. The report gives two facts, a historical 10 to 20 year integration span and the harvest now, decrypt later threat, and doesn't need a predicted date for a quantum machine to make its case. That's a strong way to argue, because it rests on things that can be checked against past migrations rather than on a forecast that can be disputed. Founders who find the quantum timeline too speculative to plan around should notice that the report's urgency doesn't depend on that timeline.
I'd also take the report at its word on difficulty. Substantial effort across diverse applications and infrastructures, each with specific requirements and constraints, is not the language of a routine upgrade. The transition is without precedent in scale, even if better understood than earlier ones. Both statements are true at once, and planning should assume both.
The practical shape of a response follows from what the report actually says. Know which of the current FIPS 186 algorithms you depend on, since those are the ones vulnerable to Shor's Algorithm. Know that FIPS 204 and 205 are the published quantum-resistant signature options. Treat the FALCON-derived standard and the additional candidates as pending. And build in the agility that lets you absorb whatever the standards process settles on. None of that requires waiting for the companion publications, since agencies may use the report's concepts and methodologies now, while existing requirements, where they exist, stay operative until those publications are finished.
An open question worth watching #
The report leaves one question open in a way its own text makes visible: how the signature field will look once the FALCON-derived FIPS arrives and the additional-signature evaluation concludes. NIST has said it will monitor developments and adjust transition efforts as appropriate, which means the guidance itself may shift as those decisions land. For anyone building signing infrastructure today, the thing to watch isn't only whether a quantum computer appears, but whether the set of standardised signature algorithms changes underneath a deployment that was designed around a fixed list. That is the case for agility, and the answer to how much it will matter is still being written by the standards process.
Frequently Asked Questions #
Which digital signature algorithms does NIST IR 8547 identify as quantum-vulnerable? The report names the algorithms in the current revision of FIPS 186: ECDSA, RSA as specified in RFC 8017, and EdDSA as specified in RFC 8032. It states that these algorithms are vulnerable to Shor's Algorithm on a cryptanalytically relevant quantum computer.
What are the post-quantum replacements for these signatures? FIPS 204 specifies ML-DSA, derived from the CRYSTALS-Dilithium submission, and FIPS 205 specifies SLH-DSA, derived from the SPHINCS+ submission. NIST is also developing a FIPS for a FALCON-derived signature algorithm as an additional alternative, and is evaluating other proposed signature algorithms for possible standardisation.
Why is the transition urgent if no cryptanalytically relevant quantum computer exists yet? The report points to two things. Historically, moving from algorithm standardisation to full integration into information systems can take 10 to 20 years, and the "harvest now, decrypt later" threat means adversaries can collect encrypted data now to decrypt once quantum technology matures. The report calls that threat model one of the main reasons the transition is urgent.
Are stateful hash-based signatures suitable for general use? No. The report says stateful hash-based signature schemes are resistant to attacks by quantum computers, but their security relies on the signer never using a one-time signature key to sign more than one message. Because of this need to maintain state, they are not intended for general use, though they may be useful for applications like software and firmware signing.
Can organisations act on the report before NIST finishes its companion publications? Federal agencies may use the report's concepts and methodologies before those companion publications are completed. Until each one is completed, current requirements, guidelines and procedures, where they exist, remain operative.
References #
- Moody, D., Perlner, R., Regenscheid, A., Robinson, A. and Cooper, D., (2026). Transition to Post-Quantum Cryptography Standards ⧉. National Institute of Standards and Technology, NIST Internal Report 8547.
- National Institute of Standards and Technology, (2023). Digital Signature Standard (DSS) ⧉. Federal Information Processing Standards Publication FIPS 186-5.
- National Institute of Standards and Technology, (2024). Module-Lattice-Based Digital Signature Standard ⧉. Federal Information Processing Standards Publication FIPS 204.
- National Institute of Standards and Technology, (2024). Stateless Hash-Based Digital Signature Standard ⧉. Federal Information Processing Standards Publication FIPS 205.
- Cooper, D. A., Apon, D. C., Dang, Q. H., Davidson, M. S., Dworkin, M. J. and Miller, C. A., (2020). Recommendation for Stateful Hash-Based Signature Schemes ⧉. National Institute of Standards and Technology, NIST SP 800-208.
- Alagic, G. et al., (2024). Status Report on the First Round of the Additional Digital Signature Schemes for the NIST Post-Quantum Cryptography Standardization Process ⧉. National Institute of Standards and Technology, NIST Internal Report 8528.
Last reviewed .
