What “Harvest Now, Decrypt Later” really means

Most security threats follow a familiar shape: an attacker finds a weakness, exploits it, and the damage is discovered — sometimes quickly, sometimes not. Harvest now, decrypt later (HNDL) breaks that shape, and that is exactly what makes it dangerous to reason about.

In an HNDL campaign, the adversary does not break your encryption at all. They simply record it. Encrypted traffic crossing the public internet, intercepted VPN tunnels, exfiltrated database backups — all of it is stored, indexed, and held. The attacker is betting that a cryptographically relevant quantum computer will eventually run Shor's algorithm against the RSA and elliptic-curve keys protecting that data. When that day comes, everything in the archive becomes readable at once.

Why executives should treat this as a present-tense problem

The instinctive response — “we'll migrate when quantum computers are real” — contains a logical error. The relevant question is not when a quantum computer arrives. It is whether your data's required confidentiality lifetime, plus the years your migration will take, extends past that arrival.

This is often written as Mosca's inequality: if X (how long data must stay secret) plus Y (how long migration takes) is greater than Z (years until a cryptographically relevant quantum computer), you are already exposed. Patient records, financial contracts, government archives, and intellectual property routinely carry confidentiality requirements of ten to fifty years. Enterprise cryptographic migrations have historically taken five to fifteen. Under almost any serious estimate of Z, the inequality already fails for long-lived data.

That is the precise meaning of “the deadline already passed”: not panic, arithmetic.

What the regulators have already concluded

You do not need to take a vendor's word for the timeline. NIST finalized the first post-quantum standards — FIPS 203 (ML-KEM), FIPS 204 (ML-DSA), and FIPS 205 (SLH-DSA) — in August 2024. NSA's CNSA 2.0 guidance sets post-quantum expectations for national security systems phasing in well before 2030. NIST's transition guidance points toward deprecating RSA and ECC around 2030 and disallowing them by 2035.

Regulated industries should read those dates the way they read every compliance horizon: as the end of the runway, not the start of the taxi.

The practical first step

HNDL cannot be mitigated retroactively — data harvested last year is gone. What an organization controls is how quickly the window closes. That starts with a question most enterprises cannot currently answer: where, exactly, is your vulnerable cryptography?

A Cryptographic Bill of Materials (CBOM) — a complete inventory of algorithms, keys, certificates, and protocols across code and infrastructure — turns the migration from a debate into a project plan. It is the single artifact every subsequent decision depends on, and it is the first thing AutoPQC builds.