Insights/ Threat Landscape · Post-Quantum Cryptography, Part 3 of 4
The new standards replacing your encryption
NIST finalised three post-quantum standards in 2024, with a fourth in draft. Where they're already deployed, what hybrid cryptography means, and the timeline.

Parts 1 and 2 established the threat and explained the mechanism. Part 3 covers the solution the global standards community spent eight years developing and finalised in August 2024. NIST's post-quantum standards aren't theoretical proposals. They're published, implemented in major cryptographic libraries, and already deployed in products your organisation likely uses today. The question for most organisations isn't whether the standards are ready. They are. It's whether your own systems are.
Data sourcesNIST FIPS 203, 204, 205 (August 2024) · NIST FIPS 206 (draft) · NIST IR 8413 (2022) · NSA CNSA 2.0 (2022) · ASD Information Security Manual (2024) · Apple PQ3 protocol announcement (February 2024) · ASD, Planning for post-quantum cryptography (2025)Figures reflect publicly reported research as of October 2026. Findings evolve. Verify current studies before citing externally.
01A different foundation entirely
Post-quantum algorithms aren't stronger RSA. They're different maths
RSA and ECC derive their security from problems that are hard for classical computers and easy for quantum ones: integer factorisation and discrete logarithms. Post-quantum algorithms are built on problems believed to be hard for both. The emphasis on "believed" is deliberate: RSA's security was also believed sound until Shor demonstrated otherwise in 1994.
NIST's evaluation process ran from 2016 to 2024. Cryptographers worldwide submitted candidates and attempted to break each other's submissions. Several were broken during the process and eliminated. The four selected in 2022 (Kyber, Dilithium, SPHINCS+ and Falcon) withstood years of sustained cryptanalytic attack from the global research community, the strongest evidence available, and the basis governments and major technology companies are now acting on.
Lattice problems: no polynomial-time quantum algorithm is known to break them. Hash-based schemes: security rests only on well-understood properties of the underlying hash function, which quantum algorithms don't meaningfully threaten. In March 2025 NIST also selected HQC, a code-based scheme, as a backup to ML-KEM built on different mathematics.
02Finalised, not proposed
Three standards finalised, a fourth on the way
In August 2024 NIST finalised three post-quantum standards, and a fourth, FN-DSA, is in draft. Organisations don't need to wait for further guidance. The algorithms are in major cryptographic libraries, and the question is now implementation, not readiness.
ML-KEM
Key encapsulation: replaces RSA and ECDH in TLS key exchange. Adds roughly 1 KB of extra data each way in the handshake, which is negligible on most connections.
✓ Chrome, Firefox, Cloudflare, AWSML-DSA
Digital signatures: replaces RSA and ECDSA for certificates, code signing, DKIM. Larger signatures (~3.3KB), fast verification.
✓ OpenSSL 3.5, Bouncy CastleSLH-DSA
Hash-based signatures: the most conservative option. Larger and slower, but the strongest long-term trust argument available.
✓ Root CA & firmware signingFN-DSA
Compact lattice signatures for bandwidth-constrained environments: IoT, embedded systems, strict size constraints.
⟳ Still in draft (October 2026)03Already running
The transition is already underway in infrastructure you depend on
Some of the most exposed data, data in transit over TLS, is already being protected post-quantum by infrastructure providers, without any action from your organisation. The gap is in the systems you own and operate directly: internal PKI, VPN infrastructure, authentication systems, data-at-rest encryption, and custom application cryptography.
The implication: the standards aren't waiting for vendor readiness. Vendors are already shipping. The gap is organisational readiness: specifically, the cryptographic inventory, vendor assessment, and migration planning most organisations haven't started yet.
04The pragmatic first step
Hybrid cryptography is the transition state, not the destination
Most cryptographic transitions don't happen as a hard cutover. RSA-1024 to RSA-2048 took around a decade. The retirement of SHA-1 has taken longer still. Post-quantum will be no different. The recommended approach during transition is hybrid cryptography: combining a classical algorithm with a post-quantum one, so the combined scheme's security depends on both.
Hybrid TLS uses both ECDH and ML-KEM simultaneously. The session key stays secure unless both algorithms are broken. A quantum computer could break the ECDH half, but the attacker would still need to break ML-KEM, which no known attack does. If a flaw is ever found in ML-KEM, ECDH still holds against today's attackers. This is what Chrome and Cloudflare now run by default.
Begin hybrid deployment: ML-KEM alongside ECDH for key exchange, ML-DSA alongside ECDSA for signatures where supported. Immediate protection, full compatibility.
NSA CNSA 2.0: from 1 January 2027, new equipment acquired for US National Security Systems should be CNSA 2.0 compliant. Browsers, servers and cloud services were expected to support and prefer CNSA 2.0 algorithms from 2025.
ASD: transition to post-quantum cryptography under way by the end of 2028, starting with critical systems and data.
ASD ISM: stop using traditional asymmetric cryptography (RSA, Diffie-Hellman, ECDH, ECDSA) by the end of 2030. The ISM is mandatory for many federal agencies and the reference point for most Australian organisations. CNSA 2.0 also requires exclusive use for software signing and networking equipment by 2030.
NSA CNSA 2.0: exclusive use for browsers, servers, cloud services and operating systems by 2033, with the NSA intending all National Security Systems to be quantum-resistant by 2035. The NSA does not require hybrid; ASD neither recommends nor prohibits it.
·Key takeaways
What to take into your next risk conversation
NIST published ML-KEM, ML-DSA and SLH-DSA in August 2024, and all three are in major libraries. FN-DSA is still in draft. The question isn't readiness. It's whether your systems are.
Chrome, Firefox, Cloudflare, Signal, Apple iMessage and AWS have all deployed it in production. The gap is in systems your organisation owns directly.
Combining classical and post-quantum algorithms provides immediate protection while keeping compatibility: the transition state, not the destination.
Inventory, vendor assessment, prioritisation, migration, and ongoing monitoring are distinct phases, each taking time. Part 4 covers the practical roadmap.


