ANKASecure© migrates existing encrypted data and signatures to NIST post-quantum algorithms (ML-KEM, ML-DSA, SLH-DSA) without exposing plaintext: the payload is decrypted and re-encrypted under a fresh content key inside the control plane, in streaming with bounded memory, without the plaintext ever being written to disk, logged or returned to the caller, and a companion operation re-signs. A policy template aligned with NIST, NSA CNSA 2.0, BSI, ANSSI, ETSI or CRYPTREC enforces the target across every application in under a second, PKCS#12 keystores are imported through a seven-stage validation pipeline, and composite hybrid keys require both the classical and the post-quantum component to decrypt. Every step is recorded in a signed, tamper-evident audit trail. Discovery and inventory of the existing estate are delivered by ANKATech channel partners.

The data you encrypted last year is already collected.

Harvest now, decrypt later is not a forecast — it is an accounting problem with a date attached. Migrating to post-quantum algorithms means re-encrypting what already exists, and the usual answer is to decrypt it in every application that wrote it. ANKASecure© does it inside the platform, without exposing plaintext.

CISOHead of cryptographySecurity architect

Is this you

Three ways this lands on your desk

A regulator set a date

NIS2, DORA, a central-bank circular, or an executive order that directs a federal procurement rule on FIPS 203, 204 and 205. You need a plan with evidence, not a statement of intent.

A customer sent a questionnaire

It asks which algorithms protect their data and when you will move. Nobody in the building can answer it across all the applications.

Someone did the HNDL maths

Your retention period is longer than the window anyone trusts. The exposure already exists, and it grows every day you keep writing with the current algorithm.

Who signs this: A CISO or head of cryptography with a dated obligation, a budget line, and an application estate they do not personally control. If cryptography still lives in each team’s code, this is the conversation.

Why existing approaches fall short

A TLS upgrade protects the next session. Not the data you already stored.

Transport cryptography negotiates its algorithms per session, so upgrading it changes what happens from now on. Data at rest is different: it was encrypted under long-lived keys, by applications that chose the algorithm in their own code, and it stays exposed for as long as it is kept. Upgrading the channel and swapping the library are both necessary. Neither moves what already exists.

  • What it protects

    Upgrade TLS
    Sessions negotiated after the upgrade, on every hop that supports it
    Swap the library in each application
    Data that application writes after its next release
    Govern centrally and re-encrypt
    New and existing data, under the keys the policy allows
  • Data already encrypted under RSA or elliptic-curve keys

    Upgrade TLS
    Untouched
    Swap the library in each application
    Moved only if each application reads, decrypts and rewrites its own data
    Govern centrally and re-encrypt
    Re-encrypted on the server without the application that wrote it; the plaintext never leaves the platform
  • Signatures already issued

    Upgrade TLS
    Untouched
    Swap the library in each application
    Re-signed only where the application still holds the key and the logic
    Govern centrally and re-encrypt
    Re-signed after every existing signature is verified, keeping co-signatures intact
  • Where the algorithm decision lives

    Upgrade TLS
    In each endpoint’s configuration
    Swap the library in each application
    In each application’s code
    Govern centrally and re-encrypt
    In policy, enforced where the operation executes
  • The next transition, from hybrid to post-quantum only

    Upgrade TLS
    Another rollout across the endpoints
    Swap the library in each application
    The same projects, again
    Govern centrally and re-encrypt
    A policy change, a successor key and the same re-encryption

The backup that outlives the upgrade

An organization moves its edge to hybrid key exchange in TLS 1.3, and every new session is protected. The nightly database backup it has written for years is not. Its data keys are wrapped under RSA, it is retained for ten years because a regulation requires it, and copies sit at a recovery site and with a cloud provider. The EU roadmap calls data that must stay confidential for ten years or more a high-risk use case. The TLS upgrade changed none of it. Moving it means opening every application that wrote it, or moving the data under a new key without them.

Four primary sources. One point in common: the data already stored.

  • OMB M-26-15

    Issued by
    White House Office of Management and Budget, June 2026
    What it says about data already encrypted
    “Agencies must prioritize identifying and re-encrypting long-lived sensitive data using keys protected by PQC mechanisms.”
  • NIST IR 8547

    Issued by
    NIST, final, October 2026
    What it says about data already encrypted
    Encrypted data remains at risk from harvest now, decrypt later, and because sensitive data often keeps its value for many years, starting the transition now is critical.
  • BSI TR-02102-1

    Issued by
    German Federal Office for Information Security, version 2026-01
    What it says about data already encrypted
    For information requiring long-term protection, storing data now and decrypting it later is a relevant threat; quantum-safe key agreement is recommended in hybrid mode.
  • EU coordinated implementation roadmap

    Issued by
    EU Member States, supported by the European Commission, June 2025
    What it says about data already encrypted
    Data that must remain confidential for at least ten years should be protected from quantum attack no later than the end of 2030.

How it works

Four moves, none of them a rewrite

The applications keep calling the same operation. What changes is what the control plane decides to do underneath.

  1. Declare the target, once

    Pick the policy template your obligation names — NIST, NSA CNSA 2.0, BSI, ANSSI, ETSI, CRYPTREC and more. ANKASecure© starts refusing key material that does not satisfy it, everywhere, in under a second. No application ships to make that true.

  2. Bring the existing keys in

    ANKASecure© imports PKCS#12 keystores through a seven-stage validation pipeline that reports each certificate’s real status and its migration path. A key past its expiry is refused for re-encryption, so the move starts while the keys are still valid.

  3. Patent pending

    Re-encrypt without exposing plaintext

    The payload is decrypted inside the control plane and re-encrypted under a fresh content key, without the plaintext ever being written to disk, logged or returned to the caller; the caller needs a re-encryption grant and no decrypt grant. It streams with bounded memory, so a terabyte archive is the same operation as a 4 KB record, and a companion operation re-signs, so signatures migrate too.

    • REENCRYPT
    • RESIGN
    • streaming, bounded memory
    • no plaintext egress
  4. Hedge while the standards settle

    Composite keys pair a classical and a post-quantum component and require both to decrypt — the AND-decrypt mode. If one component is broken, the payload is not. ANKASecure© deliberately does not implement the OR-decrypt variant.

Hybrid done right

Hybrid is a hedge. A hedge you will have to unwind.

The most credible critics of hybrid cryptography are the standards bodies themselves. NIST IR 8547 says hybrid solutions add complexity that can increase security risk and cost, and expects them to be temporary measures leading to a second transition to post-quantum algorithms only. OMB M-26-15 calls hybrid an intricate and resource-intensive stopgap. BSI and the EU roadmap recommend it anyway. Both positions hold: hybrid is worth having when it is built to a standard and built to be left.

AND, never OR

A composite encryption key requires both components to decrypt, and a composite signature is valid only when both components verify. ANKASecure© implements exactly these two modes. A mode in which either component alone decrypts is deliberately absent: it would turn the hedge into a second way in.

Standard combiners, not ad hoc ones

The two shared secrets are combined with the key combiner of NIST SP 800-227, instantiated with HKDF, or with the CatKDF and CasKDF of ETSI TS 103 744. Composite signatures follow the IETF composite ML-DSA construction.

The weakest component sets the level

A composite key is only as strong as its weakest component. Ed25519 with ML-DSA-44 is a quantum-safe hybrid at level 1, so a rotation policy that forbids downgrades refuses it as the successor to an ML-DSA-87 key.

The exit is planned, not improvised

Within hybrid, a key rotates to another composite under the same identifier. Leaving hybrid for post-quantum only uses a successor key: the old one is revoked as ALGORITHM_DEPRECATED, so it still decrypts and verifies, and the data moves by the same re-encryption that brought it into hybrid.

Re-encryption does not hand anyone the plaintext. The caller holds a re-encryption grant from the source key to the target key and needs no decrypt grant on either. The payload is decrypted and re-encrypted inside the platform under a fresh content-encryption key, and the plaintext is never returned to the caller.

Mosca’s inequality

Nobody controls when a quantum computer arrives. Everyone controls how long they keep writing exposure.

Mosca’s inequality, which NIST IR 8547 cites, sets the frame. If X is how long data must stay secure, Y is how long the transition takes and Z is when a cryptographically relevant quantum computer exists, data is exposed whenever X plus Y exceeds Z. Z is outside anyone’s control. X is set by retention rules and contracts. Y is the only variable an organization owns, and NIST notes that past transitions took ten to twenty years from standardization to full integration. For data at rest, Y has two parts: the time until the organization stops writing under quantum-vulnerable algorithms, and the time to move what it has already written.

  • Data in transit

    How the exposure arises
    Sessions captured today and decrypted later.
    What makes Y long
    Every hop in the chain must support post-quantum key exchange, and a hop that does not negotiates a classical one.
    What shortens it
    The payload is encrypted under a composite key before it enters the channel, so TLS becomes the second layer rather than the only one.
  • New data at rest

    How the exposure arises
    Every day of writing under a quantum-vulnerable key adds to the stock.
    What makes Y long
    Each application must change and ship before its writes change.
    What shortens it
    A policy refuses quantum-vulnerable key generation and rotation targets centrally, without a release.
  • Existing data at rest

    How the exposure arises
    Backups and archives under RSA or elliptic-curve keys, already copied and stored elsewhere.
    What makes Y long
    Re-encryption depends on applications that may have no owner left.
    What shortens it
    Server-side re-encryption without the originating application, streamed for large objects.
  • Long-lived signatures

    How the exposure arises
    Not harvest now, decrypt later, but forgery once a quantum computer exists.
    What makes Y long
    Signatures sit inside documents, chains and evidence that other parties verify.
    What shortens it
    Re-signing under ML-DSA or a composite key, after every existing signature is verified.
  • Old key material

    How the exposure arises
    It must stay readable for the data it protected.
    What makes Y long
    Destroying it loses the data; leaving it active keeps it writing.
    What shortens it
    A retired version keeps decrypting and verifying while new encryption and signing are denied.

The order matters. First stop adding to the exposure, which is a policy decision; then move what exists, starting with the data whose X is longest. Discovery tells you where the exposure is. Re-encryption is what reduces it.

Crypto-agility →Digital signatures and long-term evidence →

Where the migration actually stands

Post-quantum adoption measured over a real estate, not estimated — the classical, post-quantum and hybrid split, and every algorithm in use ranked by how many keys hold it. The number you start from is usually the uncomfortable one.

ANKASecure© · Analytics · Algorithm usage
The ANKASecure© algorithm usage report: 4,143 keys, post-quantum adoption at 18.2 per cent, 2,147 classical, 547 post-quantum and 208 hybrid keys, with the algorithms in use ranked beneath.
Real deployment · non-production data.

Fifteen minutes

What you will actually watch

Live, on your data shape, against a real deployment. Nothing in this list is a mock-up or a roadmap item.

  1. A file encrypted under RSA. The envelope, and the key it points at.

  2. The policy is tightened to a post-quantum minimum. The next key generation is refused, with the reason.

  3. Re-encryption to ML-KEM. Memory stays flat while the file streams. No plaintext anywhere you can point at.

  4. The old envelope still opens, because the old key was retired and not destroyed. The lifecycle state that makes that a rule rather than a favour.

  5. The audit trail for every step, signed, with the algorithm, key and outcome of each operation.

  6. Your question, answered on the deployment rather than on a slide.

Same destination, different routes

The regulators agree on the destination. They do not agree on the route.

The deadlines are on the post-quantum page, and they converge on 2035. What differs is the route: whether to pass through hybrid, for how long, and what the architecture must be able to do afterwards. An organization with operations in more than one jurisdiction may answer to several of these at once.

  • NIST IR 8547 (final, October 2026)

    Position on hybrid
    A hedge against a flaw in one component, typically temporary and leading to a second transition to post-quantum only. NIST accommodates hybrid key establishment and dual signatures in FIPS 140 validation.
    What it asks of the architecture
    Systems that cannot support post-quantum cryptography should be identified and prioritized for replacement or decommissioning.
  • OMB M-26-15 (June 2026)

    Position on hybrid
    A useful tool for managing risk, and an intricate, resource-intensive stopgap; agencies choose on their own risk assessment.
    What it asks of the architecture
    All systems cryptographically agile, with the algorithm or hybrid scheme set in external configuration rather than compiled into the application.
  • BSI TR-02102-1 (version 2026-01)

    Position on hybrid
    Quantum-safe mechanisms should be used in hybrid form. Classical key agreement alone is recommended only until the end of 2031, and classical signatures until the end of 2035.
    What it asks of the architecture
    Products with an expected lifetime beyond 2030 should be upgradeable to quantum-safe methods.
  • EU coordinated implementation roadmap (June 2025)

    Position on hybrid
    Standardized and tested hybrid solutions whenever feasible. For high-risk use cases, quantum-vulnerable public-key mechanisms are not to be used stand-alone after the end of 2030.
    What it asks of the architecture
    Support cryptographic agility and a quantum-safe upgrade path.

Two transitions, not one

Under NIST’s own reading, hybrid leads to a second transition. An estate that can re-encrypt once can do it twice. One that rewrote its applications to reach hybrid will rewrite them again to leave it.

One estate, several routes

The algorithm availability policy is set deployment-wide and per tenant: a BSI hybrid-enforced template for one subsidiary and a NIST-aligned one for another, both enforced on every operation of the same installation.

Progress as evidence

OMB asks for a risk-based prioritization and a plan for a cryptographically agile architecture. Every re-encryption is recorded in the signed audit trail with its source and target algorithm, and the classical, post-quantum and hybrid split of live keys is a report, not a survey.

What it rests on

The capabilities underneath, with their state

  • Shipping

    Re-encryption and re-signing without plaintext exposure

    Compact and streaming, bounded memory at any file size.

  • Shipping

    120+ algorithms

    The NIST post-quantum set (ML-KEM, ML-DSA, SLH-DSA), FALCON and HQC alongside the classical and regional families, plus 41 composite hybrid pairings.

  • Shipping

    37 policy templates

    29 of them algorithm-availability templates, 6 rotation and 2 lifecycle; 24 of the availability templates tied to a named standard or jurisdiction.

  • Shipping

    PKCS#12 import

    Seven-stage validation pipeline with certificate status and migration-path analysis.

  • Shipping

    Signed audit trail

    Tamper-evident chain, queryable per tenant, forwardable to your SIEM with the signature preserved.

  • Per channel

    Discovery and inventory

    What your estate uses today, delivered by ANKATech channel partners. Discovery produces visibility; the control plane produces control.

All of the platform →

The framework

Where this sits in CAPA

The Cryptographic Control Plane reference architecture describes each pillar through the scenarios it must handle and one practical test. This solution is two of those scenarios, and it is measured against their tests.

01

Primary pillar

Crypto-Agility

Scenario realised

Transitioning to post-quantum cryptography

Practical test

“Can the cryptographic strategy change without requiring an application change or release?”

03

Primary pillar

Frictionless Modernization

Scenario realised

Modernizing historical encrypted data

Practical test

“Can the protection of data encrypted years ago be changed today, without the originating application and without plaintext at rest?”

04

Supporting pillar

Cryptographic Governance & Compliance

Scenario realised

Enforcing a new enterprise security requirement

Practical test

“Can the organization prove, by querying the control plane, which policy applied to a given operation, and change that policy centrally?”

L1–L2L3–L4

Maturity move

From algorithms hardcoded in application code to a Control Plane where an algorithm change is a policy update. The migration itself is the step from level 2 to level 3; keeping it that way is level 4.

The six maturity levels →Reference architecture: cryptographiccontrolplane.org

What this does not do

  • It does not find your cryptography for you. Inventory across an estate is a discovery engagement, and ANKATech partners run it.

  • It does not re-encrypt data it cannot reach. Anything held by a system that will not delegate the operation stays where it is.

  • It does not certify you. ANKASecure© maps its controls to NIST CSWP 39, the GSA PQC Buyer’s Guide and OWASP and hands you the evidence; the determination is still your auditor’s.

Where to start

Three tracks, started on the same day

Programs stall when a complete inventory becomes the precondition for any action: the estate keeps writing under quantum-vulnerable algorithms while it is being measured. Discovery is necessary, and it runs in parallel. It does not have to come first.

  1. 1

    Greenfield: no new exposure

    New applications stop embedding cryptography. They encrypt and sign against stable key identifiers under a post-quantum or hybrid policy from their first release, so nothing new joins the stock.

  2. 2

    Critical brownfield: the longest-lived data first

    The handful of systems whose data must stay confidential longest, which a CISO can usually name without a scanner. Their keys come in, their data is re-encrypted and their signatures re-signed, without changing the applications that wrote them.

  3. 3

    General brownfield: discovery in parallel

    The inventory of the rest of the estate, delivered by ANKATech partners, onboarding systems by risk as it progresses. It informs the other two tracks and blocks neither.

Six questions for the CISO

  1. Which of your data must remain confidential for ten years or more, and under which keys is it protected today?
  2. If every TLS endpoint were post-quantum tomorrow, what would still be decryptable from a backup copied last year?
  3. How many applications would have to change to re-encrypt the data they wrote, and how many of them still have an owner?
  4. Can you stop new keys from being created under quantum-vulnerable algorithms this week, without a release?
  5. If your first transition is to hybrid, what is your plan for the second one?
  6. Can you show an auditor, key by key, which data has moved and which has not?

Bring your worst file.

A terabyte archive, an expired certificate, a format nobody owns. Fifteen minutes on a live deployment.

Book the demo