Post-Quantum Encryption

Quantum computing threatens specific cryptography, standards bodies have responded, and the migration has begun. The subject is surrounded by both dismissal and exaggeration.

What is actually threatened

Public key cryptography — the mathematics used to establish keys and verify signatures. RSA and elliptic curve algorithms rest on problems that a sufficiently large quantum computer could solve efficiently, using an algorithm published in the 1990s.

This is the part used to set up secure connections and to sign software and certificates.

Symmetric encryption is much less affected. The algorithms that encrypt the bulk of data are weakened, not broken, and doubling key length restores the margin. Common current key sizes are adequate.

Hashing is similarly weakened rather than broken.

So the threat is specific: key exchange and digital signatures, not encryption in general.

Whether the machine exists

No. Current quantum computers are far below the scale required, and the engineering distance is substantial.

Estimates of when vary enormously among people qualified to make them, and anyone giving a confident date is overreaching.

The uncertainty is the reason to act now, not a reason to wait.

Harvest now, decrypt later

The argument that makes this current rather than future.

An adversary can record encrypted traffic today and store it, decrypting when capability arrives. Nothing prevents this and it is assumed to be happening at state level.

Whether it matters to you depends on how long your data must stay confidential.

Decades: state secrets, intelligence, some medical records, genetic data, long-term legal material. For these, the threat is present.

Years: most business confidentiality. Worth considering.

Days or months: most personal communication and commerce. The message you send today has no confidentiality requirement in twenty years.

For most individuals, harvest-now-decrypt-later is not a personal concern.

What has been done

Standards bodies ran a multi-year public competition and published standardised post-quantum algorithms, covering key encapsulation and digital signatures. These are now real standards with real implementations.

Major browsers and services have deployed hybrid key exchange — combining a traditional algorithm with a post-quantum one, so the connection is secure if either holds. This is already active in a substantial share of web traffic, without anyone noticing.

Several messaging services have deployed post-quantum key exchange.

Signatures are moving more slowly, because certificate infrastructure is harder to change than connection setup.

What it means for you

Almost nothing to do personally. The transition happens in software you use. Keeping browsers and applications updated is the entire user-facing action.

Do not buy consumer products advertising quantum resistance. The genuine work is in protocols and libraries, not in an app. Products marketing quantum security to individuals are marketing.

For organisations, there is real work: inventory where public key cryptography is used, identify data with long confidentiality requirements, ask vendors about their roadmap, and plan for signature migration, which is the harder half.

Crypto-agility is the durable lesson. Systems that can change algorithms without redesign will handle this and the next transition. Systems with algorithms hardcoded will be expensive.

What to ignore

Claims that current encryption is broken. It is not.

Products claiming unbreakable quantum encryption. Quantum key distribution is a real and separate technology with narrow applicability and significant practical limitations.

Predictions of a specific date. Nobody knows.

Urgency aimed at individuals. The transition is happening in infrastructure and it is being handled.

The honest position: a genuine long-term problem, being addressed competently, on a timescale that requires planning from organisations holding long-lived secrets and requires nothing from most people beyond keeping software updated.

What an organisation should actually do now

For anyone with responsibility for systems rather than personal use, the work is real and it is inventory before it is technology.

Find where public key cryptography is used. Connections, code signing, document signing, VPNs, embedded devices, anything with a certificate. Most organisations do not have this list.

Identify data with long confidentiality requirements. Anything that must remain secret for a decade or more is the priority, because of harvest-now-decrypt-later.

Ask vendors for a roadmap, in writing. The answer tells you whether they have thought about it.

Prioritise long-lived embedded devices, which cannot be updated easily and will still be deployed when this matters.

Plan for signature migration separately. It is the harder half, because certificate infrastructure and code signing chains change slowly.

Build crypto-agility into anything new — the ability to change algorithm without redesign. That is the durable lesson regardless of timing.