Post-Quantum Cryptography: A Calm, Practical Migration Plan
Nobody can tell you when a quantum computer capable of breaking today's public-key encryption will exist. Estimates vary widely, and anyone offering certainty in either direction deserves some scepticism. Fortunately, the planning question does not depend on the answer.
The risk has two parts. The first is "harvest now, decrypt later": an adversary can record encrypted traffic or copy encrypted archives today and decrypt them once the capability exists. Data that must stay confidential for ten or fifteen years is already exposed to that future attack. The second is that cryptographic migrations are slow. NIST observes in its October 2026 transition report that previous cryptographic transitions have taken 10 to 20 years, and RSA and elliptic curve cryptography sit inside almost everything: TLS, VPNs, code signing, identity systems, payment hardware, and firmware in devices you may run for a decade.
The standards are now settled enough to plan against, and the main timeline has moved from draft to final. This is a programme to start methodically, not a crisis.
What has actually been decided
- The algorithms. On 13 August 2024 NIST published its first three post-quantum standards. FIPS 203 specifies ML-KEM for key establishment, the part that protects encrypted sessions. FIPS 204 specifies ML-DSA, the primary digital signature standard. FIPS 205 specifies SLH-DSA, a hash-based signature scheme intended as a backup. A further signature standard derived from FALCON is in development, and NIST has selected HQC as an alternative key-establishment algorithm.
- The US timeline. NIST IR 8547, released as a draft in 2024 and published in final form in October 2026, states that NIST will deprecate quantum-vulnerable public-key cryptography after 2030 and disallow it under NIST standards and guidelines after 2035. Federal agencies face earlier deadlines for their most sensitive systems under Executive Order 14412: post-quantum key establishment by the end of 2030 and digital signatures by the end of 2031.
- The UK timeline. The National Cyber Security Centre's guidance, published in March 2025, sets three milestones: define goals, assess your estate and produce an initial plan by 2028; carry out the highest-priority migrations by 2031; complete migration of all systems by 2035.
Most private-sector organisations are not directly bound by the US federal deadlines. They will feel them anyway, because vendors build to NIST and government requirements, and sector regulators and large customers tend to follow. Which dates apply to you is a question for your own advisers and regulators; this article is a planning view, not legal advice.
Start with a cryptographic inventory
You cannot migrate what you cannot find, and most organisations do not know where their cryptography lives. NIST's report names cryptographic inventories as the first of the actions organisations should be taking now. For each system, the inventory should record which algorithms and key lengths it uses, and for what (key exchange, signatures, data at rest); which library or hardware module implements them; who owns the system; and whether the cryptography can be changed by configuration or only by replacing the product.
Expect it to take longer than planned. Cryptography hides in places nobody documented: libraries compiled into old applications, certificates issued years ago, embedded devices, partner connections, hardware security modules with their own upgrade cycles. Automated discovery tools that scan traffic, code and certificate stores help a great deal, but they will not find everything. Treat the inventory as a living register kept up to date by system owners, not a spreadsheet produced once for a steering committee. It also belongs in the wider security agenda we describe in cybersecurity for the AI era, not in a separate silo.
Prioritise by how long the data must stay secret
Not everything needs to move first. NIST's report frames urgency through what is known as Mosca's theorem: if data must stay secure for X years and migration takes Y years, you need to start before X plus Y exceeds the time until a capable quantum computer exists. You do not need a precise estimate of that last number to see that long-lived data comes first.
A workable order for most organisations:
- Long-lived confidential data in transit. Health records, legal and deal material, intellectual property, customer identity data. These are open to harvesting today, so the key establishment protecting them goes first.
- Long-lived trust anchors. Root certificates, code-signing keys and firmware signing for devices with long service lives. Signatures cannot be harvested in the same way, but NIST points out that the signature transition takes longer and involves more parties.
- Systems that cannot be upgraded. Where cryptography is fixed in hardware or in software nobody maintains, identify it now and schedule replacement within normal refresh cycles. That is far cheaper than an emergency later.
- Everything else. Short-lived data and systems due for retirement can follow vendor upgrade paths.
Build crypto-agility, not a one-off swap
The most valuable outcome of this programme is not ML-KEM in place of RSA. It is the ability to change algorithms again without rewriting applications. NIST's final report says plainly that a smooth transition requires building long-term cryptographic agility into infrastructure. Algorithms that look sound today may be weakened by new research, and the standards will keep evolving.
In practice that means centralising cryptography in a small number of well-maintained libraries and services instead of letting each application choose its own, managing certificates and keys through automated lifecycle tooling, and keeping algorithm choices in configuration and policy. If changing a cipher suite today needs a code release in forty applications, fix that first. It pays back whatever the quantum timetable turns out to be. A zero trust programme that already centralises identity and key management is a natural home for this work.
Treat this migration as rehearsal for the next one, because there will be a next one.
Ask your vendors the right questions
Much of your exposure sits in products you buy rather than build. Add post-quantum questions to procurement and renewals now, while you still have bargaining power:
- Which NIST post-quantum algorithms does the product support today, and what is the roadmap for the rest?
- Can algorithms be changed through configuration, or does that need a new version or new hardware?
- Does it support hybrid key establishment during the transition?
- For hardware and embedded devices, can the cryptography be updated in the field across the product's full service life?
- Will the post-quantum implementation be validated under FIPS 140, if your regulators or customers care about that?
Vague answers tell you something too.
Use hybrid approaches during the transition
Hybrid key establishment combines a classical algorithm with a post-quantum one, so a session stays protected as long as either holds. NIST's final report accommodates this explicitly as a hedge, while noting that hybrids add complexity and are generally expected to be temporary. You may already be using it: Google made hybrid ML-KEM key agreement the default in desktop Chrome from version 131 in late 2024, so many web connections already negotiate it where servers support it.
For most external-facing services, hybrid deployment is the sensible default for the next few years. It addresses harvesting today without betting everything on algorithms that have had less real-world scrutiny than RSA. The costs are larger handshakes and some compatibility testing, particularly with older network appliances that inspect traffic.
Common questions
What is "harvest now, decrypt later"?
It is a strategy where an adversary collects encrypted data today and stores it until a sufficiently powerful quantum computer can decrypt it. NIST describes it as one of the main reasons the post-quantum transition is urgent. It matters most for data that must stay confidential for many years.
When do organisations need to move to post-quantum cryptography?
NIST will deprecate quantum-vulnerable public-key algorithms after 2030 and disallow them after 2035, and the UK NCSC expects organisations to complete migration by 2035. Given how long cryptographic transitions take, the practical starting point is now, beginning with an inventory.
What are FIPS 203, 204 and 205?
They are NIST's first post-quantum cryptography standards, published in August 2024. FIPS 203 (ML-KEM) covers key establishment, FIPS 204 (ML-DSA) is the main digital signature standard, and FIPS 205 (SLH-DSA) is a hash-based signature scheme intended as a backup.
Where Ganexa can help
Ganexa helps security and technology leaders turn post-quantum guidance into a funded, sequenced programme: cryptographic inventory, prioritisation by data lifetime, crypto-agility architecture and vendor assessment. Our cybersecurity for the AI era practice runs this alongside existing identity and zero trust work. If you want a clear first step, book a free consultation.