The Long Shelf Life of Secrets: Board Governance of Quantum-Safe Cryptography

Board papers rarely mention quantum computing without a reassuring caveat: not yet, not soon, not our problem. That framing is comfortable and increasingly wrong. The governance issue is not the arrival date of a cryptographically relevant quantum computer. It is the shelf life of the information an organisation must protect, measured against the time it will take to replace the cryptography that protects it today.

Why this belongs on the board agenda

A determined adversary does not need to decrypt a message now. Encrypted traffic and stored data can be copied today and held until the tools exist to break it. This is the harvest-now-decrypt-later problem, and it turns a future technology into a present-day exposure for any organisation holding information that must remain confidential for years or decades. Pension records, health data, merger and acquisition material, legal advice, intellectual property, customer databases and long-term research all have a shelf life that may outlast the cryptography protecting them.

The UK's National Cyber Security Centre has set a clear direction of travel: organisations should identify their cryptographic assets by 2028 and complete migration of highest-priority systems by 2035. Those dates are not distant when set against procurement cycles, embedded systems and the time needed to test and decommission legacy platforms. Regulated sectors add their own expectations. Firms subject to the FCA and PRA operational resilience regime must think in terms of important business services. Organisations within scope of the NIS Regulations, the Telecoms Security Act, the NHS data security standards or the Cyber Assessment Framework will increasingly be asked to evidence cryptographic resilience rather than assert it.

The board's role is not to choose algorithms. It is to ensure the organisation knows what it is protecting, who owns the transition, what it will cost and how progress will be assured.

Start with discovery, not with technology

Most organisations cannot yet answer a simple question: where is cryptography used across the estate, and which uses protect information that matters? The answer is rarely confined to the security team. It sits in applications, payment systems, virtual private networks, code-signing processes, backups, archives, email gateways, mobile device management, industrial control systems, building management platforms and the firmware of devices nobody has patched since installation.

A cryptographic inventory, sometimes called a cryptographic bill of materials, is the foundation. It should be tied to data classification and retention schedules, not treated as a technical asset list. The board should expect a named executive owner, a funded discovery phase and a reporting line that reaches the risk committee. Without that, the migration programme will be built on guesswork.

Directors should also ask where keys are held, how they are rotated and who can access them. Cryptographic migration is as much a key management and identity problem as an algorithm problem. An organisation that cannot locate its certificates or hardware security modules is not ready to migrate them.

The dependency problem: suppliers and embedded systems

The most difficult exposures will not be in systems the organisation controls directly. They will be in supplier products, outsourced services and embedded devices with long operational lives. Medical equipment, building controls, industrial sensors, payment terminals and telecommunications infrastructure are rarely replaced on a board-friendly timetable. Many were designed without crypto-agility, meaning the ability to swap algorithms without replacing the device.

Boards should require procurement and contract teams to ask suppliers for their post-quantum roadmap, their support lifecycle and their ability to update cryptography in the field. Where a supplier cannot answer, that is a risk to record, not a reason to move on. For critical dependencies, the board may need to consider contractual assurances, escrow arrangements or a planned exit. The same discipline applied to cloud concentration risk should be applied to cryptographic concentration risk.

This is also where regulated-sector expectations bite. An insurer, bank or utility cannot simply wait for a vendor. It must show that it understands the dependency and has a credible plan to manage it.

Budgeting a transition nobody can see

Post-quantum migration is not a single project with a single finish line. It is a multi-year programme of discovery, prioritisation, testing, replacement and decommissioning. The board should expect costs to appear in several places: specialist discovery tools and expertise, application remediation, hardware security module and certificate authority refresh, supplier change requests, testing and assurance, and the parallel running of legacy and new systems.

A useful board discipline is to fund discovery before funding migration. Discovery produces the evidence needed to prioritise. It also exposes the uncomfortable truths that make later budgeting realistic. Directors should resist the temptation to approve a token pilot and call the risk managed. The question is not whether a pilot works, but whether the organisation can move its most sensitive long-lived data onto quantum-resistant cryptography before the protection window closes.

Crypto-agility deserves specific board attention. Systems that can swap algorithms quickly will be cheaper to migrate and easier to keep compliant. Systems that cannot may need replacement. That is a capital planning issue, not a technical preference.

Questions directors should ask

• Which of our information assets must remain confidential for more than ten years, and where are they encrypted today?
• Who owns the cryptographic inventory, and when will the board see a first complete version?
• Which suppliers and embedded devices represent the longest migration lead times, and what have they committed to in writing?
• How will we fund discovery, remediation and replacement across the next three budget cycles?
• What are our internal milestones against the NCSC's 2028 and 2035 dates, and which business services are in the first wave?
• How will we assure progress through internal audit or independent review rather than self-reporting?
• What happens to our incident response and breach notification obligations if encrypted data copied today becomes readable in the future?
• Are our contracts and procurement criteria embedding crypto-agility and post-quantum roadmaps as requirements?

A realistic agenda for 2026

The board does not need a quantum strategy in the abstract. It needs a governed programme with an executive owner, a risk register entry, a discovery budget and a reporting rhythm. It should set internal deadlines ahead of the external ones, require evidence from suppliers and treat cryptographic failure as an operational resilience issue rather than a specialist curiosity.

The deadline that matters is not the day a quantum computer arrives. It is the day the organisation's most sensitive information must still be secret after that arrival. Boards that begin with inventory, ownership and dependency mapping will make informed trade-offs. Those that wait will be handed a panic programme with no time left to prioritise.