Cryptography
Cryptography & Privacy Technologies
The mathematics that decides who can read data, who can change it without being noticed, and who can work with it without seeing it. When your messages are encrypted, they become scrambled nonsense to anyone without the key. That means the real question is not where data is stored but who holds the keys. More than half of human web traffic through one large network already uses new locks designed to resist future quantum computers.
Maturity
Where it feeds in
Where you have already met this
A sovereignty audit that counts data centres and ignores key custody is counting the wrong thing. Where the data physically sits matters far less than who holds the keys and who can compel their production.
If this moved
Privacy-preserving computation reaching general-purpose performance would dissolve a trade-off both sides of the public argument treat as permanent - you could compute on data you are not allowed to read. That is a genuine structural change rather than an efficiency gain.
What to watch for
Post-quantum migration with procurement force behind it, since encrypted traffic captured today can be decrypted later. Homomorphic encryption moving out of narrow high-value cases.
Not to be confused with
The lawful-access debate. That argument is about reading messages; the consequential development is computing on data nobody can read.
What it is
The mathematics and engineering that determine who can read data, who can alter it undetected, and who can compute on it without seeing it.
How it works
- 01Two machines agree a secret in the openA key-encapsulation mechanism lets one side lock a fresh random secret with the other side's public key. Only the holder of the matching private key can unlock it. Nothing secret crosses the wire.
- 02The secret drives fast symmetric encryptionBoth sides use that shared secret to encrypt the actual data with a symmetric cipher such as AES. This part is cheap, and a quantum computer does not break it.
- 03Signatures prove who is talkingThe sender signs with a private key; anyone with the public key can check the signature. Certificates chain that public key back to an authority the browser already trusts.
- 04Post-quantum schemes swap the hard problemRSA and elliptic curves rest on problems a large quantum computer could solve. ML-KEM and ML-DSA rest on lattice problems no known quantum algorithm cracks. The price is size: signatures are dozens of times larger.
- 05Hybrids run old and new togetherToday's deployments combine a classical exchange with ML-KEM, so the connection stays safe if either one fails. This is what browsers switched on first.
- 06Privacy tech computes without lookingHomomorphic encryption and secure multi-party computation let a party work on data it cannot read. They work - slowly, and so far only where the data is worth the cost.
State of play
- NIST approved its first three post-quantum standards - FIPS 203 for key exchange, FIPS 204 and FIPS 205 for signatures - on 13 August 2024, at the end of an eight-year process (NIST, 13 Aug 2024).
- An ML-DSA-65 post-quantum signature is 3,309 bytes; a classical ECDSA P-256 signature is 64 bytes (NIST FIPS 204, 13 Aug 2024; NIST FIPS 186-5, Feb 2023).
- The share of human HTTPS traffic through Cloudflare protected by post-quantum key exchange rose from 29% at the start of 2025 to 52% in early December (Cloudflare Radar Year in Review, 15 Dec 2025 - INDUSTRY-BELIEF).
- In October 2025 not a single public web certificate used a post-quantum signature, although all major browsers had switched on post-quantum key exchange by default (Cloudflare, 28 Oct 2025 - INDUSTRY-BELIEF).
- NIST's transition guidance, IR 8547, was published as a draft on 12 November 2024; comments closed on 10 January 2025, and it was still a draft in September 2026 (NIST CSRC, checked 16 Sep 2026).
- NIST published SP 800-227, its recommendations for using key-encapsulation mechanisms securely, on 18 September 2025 (NIST CSRC, 18 Sep 2025).
Why it matters
It decides where jurisdiction actually falls. Encryption is what makes the difference between data being physically somewhere and being legally reachable - which is the entire substance of the sovereignty argument.
- It decides where jurisdiction actually falls. Data held abroad under keys you control is a different thing from data held abroad under someone else's.
- It feeds digital sovereignty and contested commons, both at 2 and held there by geopolitical alignment. At 3 it holds none of the megatrends it feeds.
- It is the clearest split on the map between two halves of one migration: key exchange moved in about a year, signatures and certificates have not moved at all.
What holds it back
Signatures do not fit
A post-quantum signature is about fifty times the size of the elliptic-curve one it replaces. Certificate chains, smart cards and small devices were sized for the old one.
Hardware lives for years
Hardware security modules, identity cards and embedded chips were certified for classical algorithms and are replaced on procurement cycles, not software cycles.
Guidance still in draft
Organisations wait for retirement dates before budgeting. NIST's transition document has sat in draft since November 2024.
Privacy tech is slow
Computing on encrypted data costs far more than computing on plain data, so it stays in narrow, high-value cases.
Terms
- Key-encapsulation mechanism (KEM)
- A way for two parties to agree a shared secret key over an open network using public-key cryptography.
- ML-KEM / ML-DSA
- NIST's standard post-quantum key-exchange (FIPS 203) and signature (FIPS 204) schemes, built on lattice problems.
- Hybrid key exchange
- Running a classical and a post-quantum exchange together, so the result is safe if either holds.
- Harvest now, decrypt later
- Recording encrypted traffic today in order to decrypt it once a capable quantum computer exists.
- Homomorphic encryption
- Encryption that allows calculations on the encrypted data without decrypting it first.
- Zero-knowledge proof
- A way to prove a statement is true without revealing the data behind it.
Steepens
| Megatrend | Weight | Role |
|---|---|---|
| Digital Sovereignty | large | |
| Contested Commons and Chokepoints | moderate | |
| Digital Identity and the Virtual Self | moderate | |
| Institutional Trust Decline | moderate | |
| Financialisation | moderate |
The gradient behaves like a sticky rubber band: it is pulled hardest toward the weakest enabler. Marginal links are shown but do not pull.
If this moved one step
Computed from the map as it stands: this score raised from 3 to 4, every other score unchanged. It is arithmetic, not a forecast.
Raised one step on its own, it lifts no megatrend's weakest link; each gradient it feeds still rises a little.
The gradient also rises in
- Contested Commons and Chokepoints3.1/5 , 3.2/5
- Digital Identity and the Virtual Self1.9/5 , 1.9/5
- Digital Sovereignty2.9/5 , 2.9/5
- Financialisation2.3/5 , 2.4/5
- Institutional Trust Decline2.8/5 , 2.9/5
What would change the gradient
Conditional, not predictive. No dates, no probabilities.
- 01Post-quantum migration deadlines with procurement force behind them
- 02Homomorphic encryption performance reaching general-purpose viability
- 03Key custody rules - the actual determinant of jurisdiction
Common misconception
That the debate is encryption versus lawful access. The more consequential development is that computing on data without decrypting it is becoming practical, which dissolves parts of the trade-off both camps assume is fixed.
Talin, B. (2026). Cryptography & Privacy Technologies. The Future Explorer, dataset v0.4.0 (2026-11-16). https://explorer.talin.digital/e/cryptography-and-privacy-technologies/