Defensive Security

Post-Quantum Cryptography: What to Do Now, Not Later

Post-quantum cryptography attracts either dismissal as a distant concern or breathless urgency, and neither serves planning well. The realistic position: no quantum computer currently threatens widely used public-key cryptography, standards for replacement algorithms now exist, and there is one specific scenario that makes preparation genuinely time-sensitive today.

Post-Quantum Cryptography: What to Do Now, Not Later

What the threat actually is

Widely deployed public-key cryptography — the algorithms securing key exchange and digital signatures — relies on mathematical problems that a sufficiently capable quantum computer could solve efficiently. Such a machine does not currently exist, and credible timelines vary widely.

Symmetric encryption is affected far less. Current guidance is that adequate symmetric key sizes remain sound, so the migration concern centres on public-key algorithms rather than on encryption at rest with strong symmetric keys.

Why some of it is urgent now

The reason preparation cannot simply wait is the harvest-now-decrypt-later scenario. An adversary capable of capturing encrypted traffic today can store it and decrypt it once capability exists.

That converts a future capability into a present risk for any data whose confidentiality must survive a long time — state secrets, health records, source code, long-term contracts, intellectual property, and personal data with lasting sensitivity. If your data stops mattering in two years, this is a manageable future project. If it must remain confidential for fifteen, traffic captured today is already exposed.

Crypto agility matters more than any single algorithm
Crypto agility matters more than any single algorithm

Where the standards are

NIST has completed its selection process and published standards for post-quantum algorithms covering key encapsulation and digital signatures. That removes the main reason organisations previously deferred planning, since the target is now defined rather than speculative.

Deployment is progressing in stages across the ecosystem. Hybrid approaches — combining a classical algorithm with a post-quantum one so security holds if either remains sound — have appeared in browsers and major services, and represent the sensible transition path rather than an abrupt switch.

What to actually do now

  1. Build a cryptographic inventory. Where is public-key cryptography used — TLS termination, VPNs, code signing, document signing, embedded devices, third-party services? Most organisations cannot currently answer this, and the inventory is the prerequisite for everything else.
  2. Identify long-lived confidential data and the channels carrying it. This is where harvest-now-decrypt-later actually applies and where migration should be prioritised.
  3. Assess crypto agility. How hard would it be to change algorithms in each system? Hard-coded algorithms, embedded devices and long-lifecycle hardware are the genuine difficulty, not the mathematics.
  4. Raise it with suppliers. Much of your cryptography is implemented by vendors, so their roadmaps determine much of yours. Ask now, and consider adding it to procurement criteria.
  5. Track certificate and key lifetimes that extend beyond your planning horizon, particularly for long-lived signing keys and embedded systems.
  6. Plan for hybrid deployment rather than an abrupt cutover.
  7. Avoid rushing bespoke implementations. Cryptographic implementation errors are far more likely to harm you than a quantum computer in the near term.
Inventory first — most organisations cannot answer where crypto lives
Inventory first — most organisations cannot answer where crypto lives

Keeping it in proportion

For most organisations this is a planning and inventory exercise now, and a migration project over the coming years — not an emergency.

It is worth being direct: an organisation with unpatched internet-facing systems and no multi-factor authentication should address those first. Post-quantum readiness is a genuine long-term requirement, but it should not displace controls addressing the attacks that are actually happening.

Related from TechBiz Security

Sources & further reading

Frequently asked questions

Can quantum computers break encryption today?
No. No existing quantum computer can break the public-key cryptography in widespread use, and credible timelines for such capability vary considerably. The planning concern is future capability applied to data captured today.
What is harvest-now-decrypt-later?
An adversary captures encrypted traffic now and stores it until quantum capability allows decryption. It makes preparation urgent for data whose confidentiality must persist for many years, even though the decryption capability does not yet exist.
Is symmetric encryption affected?
Far less than public-key cryptography. Current guidance holds that adequate symmetric key sizes remain sound, so migration effort focuses on public-key algorithms used for key exchange and digital signatures.
What is the most useful first step?
Building a cryptographic inventory — identifying where public-key cryptography is used across your systems, suppliers and devices. Most organisations cannot currently answer that question, and every subsequent decision depends on it.

Related reading

0 comments

Leave a comment

Comments are moderated before appearing.