Explore our new Cybersecurity Terms Glossary, explained by our experts.

View the Glossary View the Glossary
PRIVIA

Cybersecurity Terms Glossary

What Is Post-Quantum Cryptography?

Post-quantum cryptography means algorithms that run on ordinary computers but resist attack by quantum ones. NIST standardised the first three in August 2024.

Illustration for the cybersecurity term Post-Quantum Cryptography
Home / Blog / Cybersecurity Terms / What Is Post-Quantum Cryptography?
17 November 2026

What Is Post-Quantum Cryptography?

What Is Post-Quantum Cryptography?

Post-quantum cryptography is the set of algorithms designed to withstand attack by a quantum computer while running perfectly well on the hardware you already own. The name causes confusion: nothing about it is quantum. These are classical algorithms built on mathematical problems that quantum computers are not known to solve efficiently, which is a different field entirely from quantum key distribution.

The need arises because Shor’s algorithm solves integer factorisation and discrete logarithms, the two problems underpinning RSA, Diffie-Hellman, and elliptic curves. Symmetric cryptography is far less affected; AES-256 remains comfortable. The public key layer is where everything must be replaced.

The Standards

NIST ran an open competition from 2016, received 82 submissions, and published the first three standards on 13 August 2024:

StandardAlgorithmPurposeBased on
FIPS 203ML-KEM (was Kyber)Key encapsulationModule lattices
FIPS 204ML-DSA (was Dilithium)Digital signaturesModule lattices
FIPS 205SLH-DSA (was SPHINCS+)Digital signaturesHash functions only

FN-DSA, derived from Falcon, is expected as FIPS 206 for cases needing compact signatures, and HQC was selected in March 2025 as a code-based backup key encapsulation mechanism, so that a future break of lattice assumptions would not leave the field without an alternative.

SLH-DSA deserves a note. It rests only on the security of hash functions, which are extremely well understood, making it the conservative choice for very long-lived trust anchors such as firmware signing roots. The price is signature size, measured in kilobytes, and slow signing.

The Practical Difference: Size

The algorithms are fast — ML-KEM operations often beat elliptic curve equivalents — but the values are much larger, and that is what breaks things.

Protocols with fixed field sizes, smart cards with limited storage, DNS responses constrained by packet size, and TLS handshakes that assumed a single initial packet all need revisiting. During the browser rollout of hybrid key exchange in 2024, the most common failure was not cryptographic at all: middleboxes and embedded stacks that could not cope with a larger ClientHello.

How It Is Being Deployed

The pattern is hybrid key exchange: run X25519 and ML-KEM together and combine both secrets, so the session holds if either algorithm survives. OpenSSH shipped a hybrid default in April 2022, Signal added PQXDH in 2023, Apple launched PQ3 for iMessage in February 2024, and Chrome enabled a hybrid group by default from version 124 in April 2024. By 2025 a large share of browser HTTPS traffic was already protected this way.

Signatures are moving more slowly, because certificate chains multiply the size problem and the public PKI has to move together. That is acceptable: a signature cannot be forged retrospectively, so confidentiality is the urgent half and authentication follows.

What UK Organisations Should Do

NCSC guidance published in 2025 sets three milestones: identify your cryptographic estate and build a plan by 2028, migrate the highest-priority systems by 2031, and complete the transition by 2035. NIST has indicated that classical public key algorithms should be deprecated around 2030 and disallowed by 2035.

  1. Inventory first. You cannot migrate what you have not found. Scan TLS endpoints, review code for cryptographic calls, pull certificate inventories, and ask suppliers.
  2. Prioritise by data lifetime. Anything that must stay confidential past 2035 is already exposed to harvest now, decrypt later collection.
  3. Enable hybrids where they exist. This is available today on web servers, CDNs, SSH, and several VPN products.
  4. Put it in procurement. Much of your cryptography sits in products you did not write, and contract terms are the only lever.
  5. Identify the hardware. HSMs, smart cards, TPMs, and industrial controllers may need replacement rather than reconfiguration, so they need the longest lead time.

Avoid two mistakes in the meantime: deploying non-standardised algorithms from the competition’s earlier rounds, and treating 2035 as the start date. The discovery phase is the long one, and we help organisations scope it within our cyber risk analysis services.

FAQ – Frequently Asked Questions

Is Post-Quantum Cryptography the Same as Quantum Cryptography?

No. Post-quantum cryptography is classical software, running on normal computers, using mathematics believed hard for quantum machines. Quantum cryptography, principally quantum key distribution, uses quantum physics over dedicated optical links. The first is deployable today across the internet; the second needs specialist hardware and solves a narrower problem.

Do I Need to Replace AES as Well?

No. Grover’s algorithm offers only a square-root speed-up against symmetric key search, so AES-256 retains a very large margin and AES-128 remains adequate for most purposes. Moving symmetric encryption to 256-bit keys is a cheap precaution, but the migration work concentrates on key exchange and signatures.

Which Algorithm Should I Choose for Signatures?

ML-DSA for general use, balancing size and speed. SLH-DSA where you need the most conservative security assumption and can tolerate large, slow signatures, such as firmware or root signing keys that must remain trustworthy for decades. FN-DSA will suit constrained contexts once it is finalised as a standard.

When Will Quantum Computers Actually Break RSA?

Nobody knows. Current machines are far from the scale required, but estimates of that scale have fallen sharply: 2025 research suggested under a million noisy qubits could factor RSA-2048 in under a week, down from twenty million in a 2019 estimate. Expert surveys place meaningful probability in the 2030s, which is why migration timelines target 2035.

Related Terms

You May Be Interested In These

What Is TruffleHog?

Cybersecurity Terms

17 November 2026

What Is TruffleHog?

Read More
What Skills Does an OT Security Engineer Need?

Cybersecurity Terms

17 November 2026

What Skills Does an OT Security Engineer Need?

Read More
What Is Chisel?

Cybersecurity Terms

17 November 2026

What Is Chisel?

Read More