Post-quantum readiness: what to actually do now, and what to ignore

Post-quantum cryptography attracts a great deal of noise and a small amount of genuinely useful near-term action. The useful part is not buying anything — it is knowing where you use cryptography and being able to change it. Most organisations cannot currently answer either question.

The risk that is real today

A quantum computer capable of breaking current public-key cryptography does not exist. But harvest-now-decrypt-later does: an adversary can capture encrypted traffic or exfiltrate encrypted data today and decrypt it once capability arrives.

So the question is not "when will quantum computers arrive?" It is: does any data I transmit or store today still need confidentiality in fifteen or twenty years? For most business data the answer is no. For state secrets, long-term financial records, health data, genomic data and certain intellectual property, the answer is clearly yes — and that data has a present-day problem.

Symmetric cryptography is far less affected — larger key sizes are a sufficient response. The exposure is concentrated in public-key algorithms used for key exchange and digital signatures.

What to do now

1. Build a cryptographic inventory. Where do you use cryptography, which algorithms, which key lengths, in which systems, with what certificate lifetimes, and which third parties are involved? This is the foundational task and it is genuinely difficult — cryptography is embedded in applications, libraries, appliances, protocols and vendor products.
2. Identify long-lived confidentiality requirements. Classify data by how long it must stay secret. This is what tells you which systems are actually urgent.
3. Assess crypto-agility. If an algorithm had to change, how hard would it be? Hard-coded algorithms, pinned certificates and appliances without firmware paths are where migration will hurt.
4. Ask suppliers for roadmaps, particularly for hardware with long deployment lifetimes — HSMs, network equipment, embedded devices.
5. Track the standards, which now exist for the first generation of post-quantum algorithms.

What not to do yet

• Do not rip out working cryptography in a panic. Current algorithms remain sound against current adversaries.
• Be cautious with proprietary "quantum-safe" products that do not implement standardised algorithms. Prefer standards-based implementations.
• Do not treat quantum key distribution as the answer — it is a niche physical-layer technology with significant deployment constraints, not a general replacement for public-key cryptography.
• Do not migrate before your inventory exists. You will migrate the systems you know about and miss the ones that matter.

Sequencing a migration

When you do migrate, order by exposure: long-lived confidential data first, then externally facing key exchange, then signatures and certificates, then internal systems. Hybrid approaches — combining a classical and a post-quantum algorithm so security holds if either survives — are the sensible transitional posture and are what most standards guidance recommends.

Expect practical friction: larger key and signature sizes affect protocol behaviour, performance and embedded devices with constrained memory. Test before committing.

The compliance angle

Regulators and large customers have begun asking about post-quantum planning, particularly in financial services and government supply chains. A credible answer today is a cryptographic inventory, a data-longevity assessment and a stated migration approach — not a completed migration. Being able to produce that is increasingly a procurement differentiator.

Key takeaways

• Harvest-now-decrypt-later makes long-lived confidential data a present-day issue.
• The cryptographic inventory is the foundational task and most organisations lack one.
• Crypto-agility matters more than any specific algorithm choice.
• A credible plan, not a completed migration, is what customers currently expect.