“Harvest Now, Decrypt Later” Is Already in Production
Source: Dev.to
This is not a futurist piece. It is about why quantum risk has already become an operational, legal, and governance problem for developers, security teams, and engineering leadership—and why waiting for “real quantum computers” is already a failure mode.
The shift we are living through is subtle but decisive:
from future speculation to active liability.
By late 2025 / early 2026, that shift is no longer theoretical.
1. The HNDL Reality Check: Why Time Is Already Against You
The “Harvest Now, Decrypt Later” (HNDL) strategy has moved from theory to documented intelligence practice.
If the shelf‑life of your data (X), plus your migration time (Y), exceeds the time until a Cryptographically Relevant Quantum Computer exists (Z), your data is already compromised—just not yet decrypted.
- X (Data shelf‑life): 10–30 years (PII, genomic data, trade secrets)
- Y (Migration time): 5–10 years for large organizations
- Z (Threat horizon): Estimated 2030–2035
For many organizations today, X + Y > Z.
That means the breach already happened. The only thing missing is compute.
2. Standards Removed the Last Excuse
- August 2024: NIST finalized FIPS 203, 204, and 205.
- March 2025: HQC (Hamming Quasi‑Cyclic) was selected as a backup algorithm to ensure cryptographic diversity.
From that point forward, “there are no standards yet” stopped being a defensible position. By 2026, insurers, regulators, and courts increasingly treat the absence of a PQC migration plan the same way they treat unpatched known vulnerabilities: a failure to meet the standard of care.
3. Hardware Reality: The Shift to Logical Qubits
The conversation has moved beyond raw qubit counts. The real metric now is logical qubits—error‑corrected, stable computation.
| Platform | Status (2026) | Key Breakthrough |
|---|---|---|
| IBM | 120+ qubits, 300 mm fab scaling | 10× faster qLDPC decoding |
| 105‑qubit chip | Exponential error suppression | |
| Microsoft | 28 logical qubits | Topological hardware protection |
| QuEra | Targeting 100 logical qubits | Reconfigurable neutral atoms |
This is no longer science fiction. It is roadmap execution.
4. What This Means for Engineering Teams (Not Just Boards)
This is not a “wait‑and‑see” problem. It is an operations and governance problem. Three actions matter now:
Inventory your crypto like you inventory dependencies
Establish a Cryptographic Bill of Materials (CBOM). You cannot migrate—or defend—what you cannot see.
Wrap before you replace
Pilot hybrid key exchange (e.g., ML‑KEM alongside classical TLS). This immediately mitigates HNDL risk without ripping out proven systems.
Quantum risk is a supply‑chain problem
Audit third‑party dependencies and vendors. If your vendors cannot articulate a PQC roadmap, they are already a liability.
5. This Is Not a Migration Project
Quantum security is not a one‑time upgrade. It is an operational discipline. The real failure mode will not be “broken crypto.” It will be the inability to prove—technically and legally—that you acted responsibly after the risk was already known.
That is the line history keeps drawing. By 2026, that line is no longer theoretical.