Most security risks begin when the attacker strikes. The quantum threat is different. An adversary can intercept and store encrypted data today, at almost no cost, and wait until a cryptographically relevant quantum computer exists to decrypt it. This is harvest now, decrypt later (HNDL), sometimes called "store now, decrypt later". It means the quantum risk for some data has already begun.

What can be harvested

HNDL targets encrypted data in transit whose session keys were agreed using quantum-vulnerable public-key cryptography, such as RSA or elliptic-curve Diffie–Hellman. Examples:

  • TLS traffic to websites and APIs;
  • VPN tunnels between offices and to the cloud;
  • encrypted messaging that doesn't yet use post-quantum key exchange;
  • backups or files encrypted with keys wrapped by RSA and sent over networks.

Data at rest encrypted with AES-256 under keys that never travelled over a vulnerable channel is much less exposed. The weak point is usually how keys were exchanged or wrapped, not the bulk encryption.

Mosca's inequality

A simple model, proposed by Michele Mosca, helps decide what is at risk:

If X + Y > Z, you have a problem.

  • X, shelf life: how many years the data must stay confidential.
  • Y, migration time: how many years it will take you to move the relevant systems to quantum-safe cryptography.
  • Z, threat timeline: how many years until a cryptographically relevant quantum computer exists.

Nobody knows Z exactly. Expert surveys give a wide range, with meaningful probability within the next 10–15 years. You don't need a precise Z. You need to see which data has a long X and which systems have a long Y.

Worked example

A hospital's patient records must stay confidential for the patient's lifetime: X = 30+ years. Migrating its clinical systems, medical devices and partner links might take Y = 7 years. Even if a quantum computer is 20 years away (Z = 20), X + Y = 37 > 20. Records sent over vulnerable channels today could be readable within the patients' lifetimes.

By contrast, one-time passwords with a shelf life of seconds don't need protection against HNDL at all, though the systems that issue them still need migrating before quantum computers arrive.

Data with a long shelf life

Data type Typical confidentiality need
Health and genetic data Lifetime of the person and beyond
National ID, biometrics Lifetime (can't be changed)
Government and defence secrets Decades
Intellectual property, trade secrets, R&D 10–25 years
Legal and M&A documents Years to decades
Long-term financial records, customer KYC 10+ years
Authentication secrets reused for years Until changed

A quick prioritisation exercise

  1. List data categories you transmit and their required confidentiality period (X).
  2. List the channels each category travels over: TLS, VPN, file transfer, APIs, messaging.
  3. Estimate migration time for each channel (Y). Some are quick (a load balancer setting), others slow (embedded devices, partner integrations).
  4. Flag categories where X + Y is large, say over 10 years.
  5. Act first on flagged categories over internet-facing or third-party channels.

What to do about it

  • Enable hybrid post-quantum key exchange on internet-facing TLS where your platforms support it. Many CDNs, browsers and modern libraries already do. See hybrid key exchange in TLS.
  • Upgrade VPNs and site-to-site links carrying long-lived sensitive data as vendors add post-quantum or hybrid modes.
  • Minimise what travels. Data that is never sent can't be harvested.
  • Watch partner links. Ask key partners and vendors for their post-quantum plans for the connections you share.
  • Don't panic about signatures for HNDL. Signatures matter for authenticity once quantum computers exist, but an attacker can't retroactively forge a signature that was already verified. Signatures still need migrating, on a slightly different priority.

Learn more

Start with post-quantum cryptography explained and the migration roadmap. For cryptography fundamentals, see the free CISSP track.

Written and reviewed by the DigiTrust Asia editorial team. How we research, review and correct articles.

Strengthen your crypto fundamentals

Cryptography lessons and practice questions in the free CISSP track.

Create a free account