Post-Quantum Encryption for Memories Meant to Last Decades
Post-quantum encryption, explained calmly: what harvest now, decrypt later means for family archives, which encryption is exposed, and how hybrid schemes help.
Key takeaways
- Post-quantum encryption uses math problems believed to resist both ordinary computers and future quantum computers.
- Harvest now, decrypt later means data copied today could be opened years from now if its key exchange relied on quantum-vulnerable math.
- Symmetric encryption such as AES-256 is expected to hold up well; the main exposure is public-key key exchange and signatures.
- Hybrid schemes pair a classical algorithm with a post-quantum one, so an attacker would need to break both to recover a key.
- NIST published the first post-quantum standards, FIPS 203, 204 and 205, in August 2024.
Post-quantum encryption is cryptography designed to resist attacks from future quantum computers as well as from today's machines, and it runs on the ordinary phones and laptops you already own. It matters most for data that must stay private for a long time, because of a threat known as "harvest now, decrypt later": someone could copy encrypted data today and wait for a computer able to read it. Family archives meant for grandchildren are a textbook case, and the reassuring news is that standards and practical defenses already exist.
This guide explains what post-quantum encryption is, why quantum computers matter, which parts of encryption are actually exposed, and what hybrid approaches do, without the hype.
What is post-quantum encryption?
Most secure connections and shared files today rely on public-key cryptography such as RSA or elliptic-curve methods, including X25519 for agreeing on keys and Ed25519 for signatures. Their security rests on math problems, like factoring very large numbers or computing discrete logarithms, that ordinary computers cannot solve in any practical amount of time.1
Post-quantum cryptography uses different math problems, such as problems built on mathematical lattices or on hash functions, for which no efficient attack is known on either an ordinary or a quantum computer.12 The name describes what it defends against, not how it runs. Post-quantum algorithms are software, and they work on everyday devices.
It is also different from quantum cryptography, which uses physics and special hardware to distribute keys. When people talk about protecting files and messages with post-quantum encryption, they mean new algorithms running on ordinary computers.
Why quantum computers matter for encryption
Quantum computers are not simply faster computers. For a few specific problems, they can use quantum effects to take shortcuts. Two algorithms explain most of the concern.
Shor's algorithm, published in 1994, could efficiently factor large numbers and compute discrete logarithms on a large enough quantum computer.3 Those are exactly the problems behind RSA and elliptic-curve cryptography. A sufficiently capable machine running it could, in principle, work out private keys from public ones, which would break today's common methods of exchanging keys and making digital signatures.4
Grover's algorithm offers a much smaller speedup for searching through possibilities, such as trying every key for a symmetric cipher.5 In theory it roughly halves the effective strength of a key, so a 256-bit key would offer security comparable to a 128-bit key against a quantum attacker, which is still considered very strong. In practice, NIST expects the gain to be smaller still, because Grover's algorithm is hard to split across many machines, and it describes AES-256 as safe for a very long time.6
No publicly known quantum computer can break RSA or elliptic-curve cryptography today.7 Doing so would require far more stable, error-corrected qubits than current machines have, and estimates of when, or whether, that will happen vary widely. That uncertainty is the reason to plan early for data that must last.
Harvest now, decrypt later: why time matters
Imagine someone copies a sealed envelope today and files it away, waiting for a letter opener that has not been invented yet. If the envelope's seal is one that future tools can open, the letter inside becomes readable on that day, no matter how carefully it was sealed in the past.
That is harvest now, decrypt later, a threat model recognized by governments and standards bodies.47 Encrypted data can be collected today, through a breach, an intercepted connection or a leaked backup, and stored until the tools to read it exist. Once a copy is taken, re-encrypting your own files later does nothing for the copy someone else already holds.
A rule of thumb known as Mosca's theorem turns this into a simple comparison. Add how long your data must stay private to how long it takes to move to new cryptography. If that total is longer than the time until a capable quantum computer exists, the move should start now.7
Why long-lived family archives are a textbook case
Most data loses its sensitivity quickly. A credit card expires, a one-time code is useless within minutes, and last year's shopping list never mattered much. Family archives work the other way.
- They are meant to last generations. A letter for a daughter's wedding day, or recordings of a grandmother's stories, may be opened thirty or sixty years from now.
- They stay personal. Health histories, adoption records, private letters and childhood photos do not become less personal with time. A toddler's bath photo becomes an adult's photo.
- They get copied. Archives move between devices, backups, services and companies over the decades, and every copy is another chance for someone to harvest it.
- They cannot be resealed after the fact. Protection has to be in place before a copy is taken, not added later.
That combination of a long shelf life and lasting privacy is exactly the scenario harvest now, decrypt later describes.
Symmetric vs public-key: which parts of encryption are exposed
Not all encryption is equally affected, and this is the most reassuring part of the story.
Symmetric encryption, such as AES-256, uses the same secret key to encrypt and decrypt data. In well-designed systems, it is what encrypts the actual contents of your photos and files. Because Grover's algorithm only halves its effective strength in theory, AES-256 is generally regarded as a sound choice for long-term data.6
Password-based keys depend on the password more than on quantum computing. A memory-hard function such as Argon2id makes each guess expensive, but a weak password stays weak. A long, unique passphrase is still the best protection you control.
Public-key cryptography is where the exposure lies. It is used whenever two people who have never shared a secret need to agree on one, for example when you share an encrypted album with your brother or a cousin sends you a recording. The file key travels inside that exchange. If the exchange relies only on quantum-vulnerable math and someone records it, a future quantum computer could recover the file key and, with it, the file.
Digital signatures prove who created or approved something and that it has not changed since. Forged signatures are a future risk rather than a retroactive one, because signatures are not vulnerable to harvest now, decrypt later.7 Still, archives that must prove their authenticity decades from now benefit from post-quantum signatures as well.
In short, the contents of a well-encrypted file are in good shape. The keys that travel between people, and the signatures that vouch for an archive, are what need upgrading.
What hybrid post-quantum schemes are and why they are used
A hybrid scheme combines a classical algorithm with a post-quantum one and derives the final key from both. For key exchange, a common pairing is X25519 with ML-KEM-768.8 An attacker would need to break both to recover the key.
Why not switch entirely to the new algorithms? Because they are young. Classical algorithms like X25519 have been studied intensely for many years, while post-quantum algorithms have had less time under public scrutiny. A hybrid design keeps the proven classical protection in place today and adds protection designed to resist tomorrow's computers. If a weakness were ever found in one half, the other half would still stand.7
Hybrid key exchange is no longer exotic. In 2024, Google's Chrome team announced that the browser's hybrid key exchange would move to the standardized pairing of ML-KEM-768 and X25519.8 Signal has published a hybrid handshake for its messaging protocol, called PQXDH, which is designed to prevent harvest now, decrypt later attacks.9 Signatures can be paired in the same spirit, for example Ed25519 alongside ML-DSA-65.
The standards: FIPS 203, 204 and 205
In August 2024, the US National Institute of Standards and Technology published its first three post-quantum standards, the result of a public selection process that began in 2016:
- FIPS 203, ML-KEM: a lattice-based key encapsulation mechanism for agreeing on keys, derived from CRYSTALS-Kyber.1
- FIPS 204, ML-DSA: a lattice-based digital signature algorithm, derived from CRYSTALS-Dilithium.10
- FIPS 205, SLH-DSA: a hash-based digital signature algorithm, derived from SPHINCS+.2
NIST's transition plan, finalized in October 2026, also sets a timeline for the older methods: it will deprecate quantum-vulnerable public-key algorithms such as RSA and elliptic-curve methods after 2030 and disallow them under its standards after 2035.7
You can read about the program on the NIST post-quantum cryptography project page (opens in a new tab) and read the key encapsulation standard itself in FIPS 203 (opens in a new tab). One caution: an application that uses these algorithms is not thereby certified by NIST. Implementation quality, key handling and independent review still matter.
What to ask about post-quantum protection for your archive
If a service says it offers post-quantum protection, a few questions show what that means in practice:
- Which parts use post-quantum methods: sharing, signatures, or something else?
- Is the key exchange hybrid, combining a classical and a post-quantum algorithm?
- What happens to older data and older shares made before the upgrade?
- Does the service say clearly what post-quantum cryptography does not protect, such as metadata or a compromised device?
- Does it avoid absolute promises? Careful providers describe designs built to resist known attacks, not designs that can never fail.
For a broader look at how families keep archives readable across decades, our guide to archival file formats covers the file side of the same long view.
How Henedo uses post-quantum cryptography today
Henedo encrypts vault contents in your browser with AES-256-GCM, protected by a separate vault password run through Argon2id. Where sharing relies on each person's own public keys, such as files in a shared vault between partners or an answer a relative records through a link, the key is sealed with hybrid X25519 and ML-KEM-768 key exchange. Legacy contacts open what you assign them with a random access key you hand over yourself, which involves no public-key exchange at all. Current contact bundles and finalized wills carry hybrid Ed25519 and ML-DSA-65 signatures.
The limits are documented alongside the protections. Hybrid sharing depends on the recipient having post-quantum keys available, and older shares keep compatibility paths so they remain readable. New algorithms do not upgrade copies someone may have taken before. Account, storage and delivery metadata stays visible to the service, and using NIST algorithms does not mean NIST certifies Henedo. The details are on the security page. For the basics of keys and devices, start with end-to-end encryption, explained, and for the long view of family archives, see what digital continuity means.
Sources
Each number in the article links to the published work it draws on.