By Robert Bedington, CTO and Co-founder, SpeQtral
Most security leaders now accept that today’s public-key cryptography has an expiry date. A powerful enough quantum computer would break the RSA (Rivest–Shamir–Adleman) and elliptic-curve algorithms that protect most of the world’s encrypted traffic. Data stolen today can be stored and decrypted later, so sensitive information with a long shelf life is already at risk. Regulators have noticed. In July 2026, Singapore’s Cyber Security Agency published a Quantum-Safe Migration Handbook that sets expectations for critical infrastructure in sectors such as banking, energy, healthcare and transport.
The usual advice is to move to post-quantum cryptography (PQC) and, for the most sensitive links, to consider quantum key distribution (QKD). This is good advice, but it hides an awkward gap. For many organisations, the real choice is between an option that is cheap but unproven and one that is proven but expensive. There is plenty of room in between.
The Two Ends of the Spectrum
PQC is software. It replaces vulnerable algorithms with new mathematical problems that are believed to be hard for both classical and quantum computers. It works at any distance, needs no new infrastructure and is now standardised. Its weakness is the word “believed”. PQC algorithms are young. Their security rests on assumptions, not proofs, and several candidates in the standardisation process were broken by classical attacks along the way. A PQC algorithm could be found weak next year or in ten years, and anything protected by it would then be exposed.
QKD sits at the other end. It uses the physics of single photons to share keys, and any eavesdropping attempt leaves a detectable trace. Its security can be proven in a way that does not depend on an attacker’s computing power. The cost is practical: specialised hardware, dedicated optical fibre, distances limited to roughly 100 km per link without trusted relay nodes, and key rates that fall as distance grows. For a bank linking two data centres in the same city, it can make sense. For a regional network, or a team with a finite budget, it is often hard to justify today.
The industry discussion tends to treat these as the only two options. In practice there is a spectrum, and some of the most useful options sit in the middle.
Filling the Middle Ground: Layered Keys Over a Separate Physical Channel
One promising middle-ground approach applies defence in depth to the key itself.
A hardware quantum random number generator produces the key material. That key is then protected by several independent methods at once: a standardised PQC algorithm, a well-tested classical method such as elliptic-curve Diffie–Hellman, and pre-shared symmetric keys. The US National Security Agency recommends pre-shared keys for quantum resistance in its Commercial Solutions for Classified programme. The key also travels over a separate optical channel rather than the main data network.
To get the key, an attacker would need to break the PQC algorithm, break the classical algorithm, obtain the pre-shared keys and gain physical access to the optical path. No single breakthrough is enough. The approach uses standard optical parts rather than single-photon hardware, so it costs much less and has no QKD-style distance limit. It can also use the same key-management interfaces as QKD, such as the ETSI GS QKD 014 standard, so it plugs into the same encryptors and network design.
The same principle works just as well in space. Satellites are the most practical way to get keys between cities, across borders or between continents without long chains of trusted relay nodes. Satellite QKD is advancing, but it is still early and costly. A satellite optical link can deliver layered keys out of band to ground stations in the same way fibre does between data centres. Because the interfaces match QKD, an organisation can run its key infrastructure as if satellite QKD were already in place, then upgrade when it becomes available.
This approach doesn’t replace QKD where its proven security is truly needed. It is a step up from software-only PQC, adding protection against a different type of attack.
Why This Matters for Governance and Risk
For IT professionals, the real value of this middle ground is not technical. It is in how organisations can manage and stage the migration.
Defensible risk-based choices. Not every link deserves the same protection. A spectrum of options lets you match the strength of key distribution to the sensitivity and lifespan of the data, rather than defaulting to “PQC everywhere” or “QKD nowhere”.
Protection against unknown algorithm failure. A layered approach means that a future break in a PQC algorithm does not automatically become a breach. That supports crypto-agility, a principle regulators increasingly expect.
Future-proof architecture, affordable first step. If an interim solution uses the same interfaces and key-management standards as QKD, it can act as a placeholder. Organisations can build a quantum-safe architecture now and later swap in QKD on the most critical links. This might happen when hardware prices fall, at the next tech refresh, or if the quantum threat accelerates. None of it requires redesigning the network.
A path to long distances. On the ground and through satellite links, the same approach gives a realistic route to quantum-safe keys between cities and across continents, beyond the reach of fibre QKD.
A Practical Starting Point
Organisations do not need to bet everything on one technology. A sensible sequence is to:
The quantum threat rewards early, layered action more than waiting for a perfect solution. The good news is that the middle ground already exists, and it is far more affordable than many security leaders assume.
Latest News
