Which US Federal Post-Quantum Deadline Binds You

For four years, US federal post-quantum policy told agencies to inventory their cryptography, name a migration lead, and report annually. It never told them when the migration itself had to be finished. Five policy documents published across three weeks in June 2026 supplied the missing dates. Post-quantum cryptography, the family of algorithms designed to survive attack by a future quantum computer, now has a federal calendar attached to it.

Those documents did not supply one date. They supplied a set of them, sorted by who owns the system and by what the cryptography is doing inside it. For a governance, risk, and compliance team, the question is no longer whether post-quantum cryptography applies but which deadline binds your systems and what you’ll be asked to show against it.

We wrote this reference for that question.

The three layers, built in order

The June documents were assembled in sequence, and reading them in that order explains why the pieces fit.

Governance came first. National Security Presidential Memorandum 12, issued June 12, 2026, rebuilt the cybersecurity governance framework for National Security Systems. A National Security System, or NSS, is one that handles classified material or supports military and intelligence functions directly. NSPM-12 rescinded a 1990 directive and its 2022 successor, re-established the Committee on National Security Systems with authority to issue binding directives, and named the Director of the NSA as National Manager for NSS. None of that is about post-quantum cryptography as such. It’s the layer that lets one cryptographic standard be enforced across every agency that runs an NSS, not only the defense agencies.

The mandates came ten days later. Executive Order 14412 set dated migration deadlines for civilian high-value assets and high-impact systems, and explicitly carved out NSS. A companion order, EO 14413, covered federal investment in quantum computing itself. The Department of Defense post-quantum strategy, published June 23, set department-wide migration gates for all DoD systems. The carve-out in EO 14412 is a jurisdictional boundary rather than an exemption, and it works only because the governance layer was already in place.

Implementation guidance followed two days after that. OMB Memorandum M-26-15 arrived on June 24, well inside the ninety days the executive order allowed. It carries the five-phase civilian timeline, the plan submission requirements, and the most detailed technical direction OMB has published on this subject.

The legal floor that predates June

Three authorities were already in force before any of this, and they remain the foundation.

The Quantum Computing Cybersecurity Preparedness Act, Public Law 117-260, was signed in December 2022. It requires OMB to direct agencies to prioritize migration and to report to Congress on progress. Both the 2022 and 2026 OMB memoranda cite it as their statutory basis. An executive order cannot revoke it.

National Security Memorandum 10, from May 2022, set the planning framework and a goal of mitigating as much quantum risk as feasible by 2035. The June 2026 order converts that goal into enforceable dates for the highest-priority systems.

NIST published the algorithms in August 2024. FIPS 203 specifies ML-KEM, which establishes a shared secret between two parties and replaces the key exchange in protocols like TLS. FIPS 204 specifies ML-DSA and FIPS 205 specifies SLH-DSA, both of which produce digital signatures and replace RSA and ECDSA in certificates, code signing, and authentication. FN-DSA is expected as a further signature standard. Every June document points back to these three.

Alongside them sits NIST IR 8547, which proposed deprecating quantum-vulnerable public-key cryptography after 2030 and disallowing it after 2035. That draft is the reason the federal schedule has two horizons rather than one.

Which deadline binds your systems

Civilian high-value assets and high-impact systems. A high-value asset, or HVA, is a CISA designation for a system whose compromise would cause serious harm to an agency mission or to national interests. A high-impact system is one rated “high” under FIPS 199, the federal standard for categorizing system impact. For these, EO 14412 sets December 31, 2030 for key establishment and December 31, 2031 for digital signatures.

Remaining civilian systems. Everything else in the non-NSS civilian estate falls into the final phase of the OMB timeline, with 2035 as the outer limit and risk-based prioritization deciding the order. Agencies still have to account for these systems in the plan they submit this year.

Department of Defense systems. All DoD systems must support post-quantum cryptography or be phased out by December 31, 2030, and use it by December 31, 2031. The strategy attaches an “unless otherwise noted” qualifier, which defers the hardest categories to the NSS schedule below.

National Security Systems. These follow CNSA 2.0, the NSA’s second Commercial National Security Algorithm Suite. CNSA 2.0 is more granular than either the civilian order or the DoD strategy, because it sorts by product category rather than by cryptographic function. Software and firmware signing and traditional networking equipment reach exclusive use by 2030. Web and cloud services, operating systems, large public key infrastructures, and constrained or niche devices reach exclusive use by 2033. The suite is ML-KEM-1024, ML-DSA-87, LMS and XMSS for firmware signing, AES-256, and SHA-384 or SHA-512.

Anyone running mixed estates hits one divergence. The OMB memorandum accepts all three NIST signature and key-establishment standards, including SLH-DSA. CNSA 2.0 does not include SLH-DSA. A civilian agency can choose a hash-based signature scheme that avoids lattice assumptions entirely. An NSS operator cannot.

Federal contractors. EO 14412 directs the Federal Acquisition Regulation (FAR) Council to publish a proposed rule requiring covered contractors to comply with the NIST post-quantum standards by December 31, 2030. The DoD strategy adds direction to update CMMC, the Cybersecurity Maturity Model Certification used across the defense industrial base, to include post-quantum requirements.

Critical infrastructure. No direct migration deadline. Sector Risk Management Agencies are directed to help owners and operators develop plans. Sectoral regulators, insurers, and customers are the more likely source of a binding date here.

Why key establishment comes first

The one-year gap between the 2030 and 2031 civilian dates is deliberate, and it follows the threat rather than the difficulty.

Harvest Now, Decrypt Later describes an adversary capturing encrypted traffic today and storing it until a quantum computer can decrypt it. The exposure is already running for any data with a long confidentiality life. Swapping ML-KEM into a key exchange is also the lighter operation, because it touches protocol negotiation rather than an entire trust hierarchy.

Signatures get the extra year for the opposite reason. Migrating to ML-DSA touches certificate authorities, cross-certification, issuance, revocation, and every device that validates a chain. It is a public key infrastructure project, not a protocol change.

Two things follow for a GRC function. Authentication is not deferrable, only sequenced, and the DoD strategy is explicit that a confidentiality-only migration does not count as migrated at all. And the signature work should start while the key-establishment work is running, because the discovery effort is the same effort.

What a compliant plan has to contain

Agency migration plans are due to OMB and the Office of the National Cyber Director (ONCD) roughly 120 days after the memorandum, which puts submission in late October 2026. The memorandum names what a plan must include, and the list doubles as a self-assessment for any organization outside the federal perimeter that expects to be asked the same questions by a customer.

• Risk-based prioritization of systems, with the reasoning recorded
• A phased timeline aligned to the applicable deadline
• An automated inventory methodology, not a one-time manual survey
• A crypto-agility architecture plan
• Third-party and supplier coordination
• Resource estimates
• Named governance roles and accountable officials

The shift from the 2022 guidance is in the last four items. The earlier expectation was a prioritized inventory. The current expectation is an executable plan with an owner attached to each part of it.

Agility is now a compliance property

Crypto-agility means a system can change cryptographic algorithms through configuration rather than redevelopment. The OMB guidance specifies provider-based cryptographic libraries, configuration-driven algorithm selection, negotiated cipher suites, and key management and hardware security modules that can hold more than one algorithm family.

Read plainly, a system hardcoded to ML-KEM is non-compliant in spirit even though ML-KEM is the correct choice today. NIST has already signalled further standards, and the 2035 horizon is long enough that the current suite will not be the last one. The practical move is to write agility into procurement specifications now, while contracts are being renewed anyway.

The inventory format that does not exist yet

A Cryptographic Bill of Materials, or CBOM, is a structured, machine-readable declaration of what cryptography a system contains and where. The June order directs CISA and NIST to publish minimum elements for one, on a clock that lands in early 2027.

That creates an awkward but manageable gap. Agencies are directed to build automated inventories now, using software composition analysis, static and dynamic testing for cryptographic calls, and network scanning for protocols in use. The standard format for reporting the result arrives later. Teams that start now should record findings in a structure they can map onto a published schema, rather than waiting for the schema and losing a year of discovery time.

For contractors, the CBOM is the one piece of genuinely good news in the package. Three overlapping compliance regimes cannot be tracked in a spreadsheet, and a common taxonomy is what makes it possible to demand structured cryptographic data from your own suppliers rather than reading their PDFs.

The bottleneck nobody has widened

Both tracks depend on validated products, and validation is the constraint.

On the commercial side, the Cryptographic Module Validation Program administers FIPS 140-3 testing, and a module that has not cleared it cannot be sold into most federal use. The queue has been the well-known limiting factor on how fast new cryptography reaches production. On the high-assurance side, NSA certification gates every encryption device.

EO 14412 directs NIST to accelerate the validation process, and the DoD strategy calls for simpler certification. Both documents identify the problem correctly. Neither has yet added capacity. An instruction to accelerate a queue is a statement of intent, and intent has never validated a module. Plan procurement on the assumption that validated products arrive later than the roadmaps say.

Dates ahead

Late October 2026. Agency migration plans due to OMB and ONCD.

December 2026. FAR Council proposed rule on contractor compliance due. NIST begins its validation programme revision.

March 2027. CISA and NIST minimum elements for a CBOM due.

December 31, 2027. NIST post-quantum pilot project completion.

January 2, 2030. Federal TLS 1.3 support deadline.

December 31, 2030. Civilian key-establishment deadline for HVAs and high-impact systems. DoD support-or-phase-out gate. Contractor compliance date under the proposed FAR rule. CNSA 2.0 exclusive use for signing and traditional networking.

December 31, 2031. Civilian digital-signature deadline. DoD use gate.

1. CNSA 2.0 exclusive use for web and cloud, operating systems, large PKI, and constrained devices.

1. Full migration of remaining federal civilian systems.

What to do in the next two quarters

Start the inventory with automated tooling and accept that the first pass will be incomplete. Discovery is the long pole in every migration programme we see, and a partial inventory available in October is worth more than a complete one available in 2028.

Establish your binding date before you plan anything else. Organizations that operate across several classifications should assume the earliest applicable deadline governs the programme, because a mixed estate cannot run two migration schedules through one change window.

Put agility language into procurement now. Every contract renewed in the next eighteen months is a chance to make algorithm flexibility a supplier obligation rather than a later renegotiation.

Ask suppliers for a dated post-quantum roadmap in writing. The procurement cascade is what carries this mandate past the federal boundary. A prime that needs validated products by 2030 will push the requirement down to its subcontractors, and they will push it to theirs. That pattern is how FedRAMP became the default cloud security bar, and it is enforced at the point of sale rather than by an inspector.

Stop treating the arrival date of a cryptographically relevant quantum computer as the planning input. None of the June documents estimates one. Policy calendars, not qubit counts, now set the schedule.

How this reference is maintained

Federal post-quantum policy is on a published clock, so this page is reviewed against each milestone in the date list above, and after any Federal Register publication that changes a deadline or an algorithm choice. The CNSA 2.0 category schedule and the status of NIST IR 8547 are the two items most likely to move between reviews.

Where the capability comes from

Every requirement above resolves into people. Someone has to classify the systems, run the discovery tooling, read what it returns, decide the migration order, write the plan, and defend the reasoning to an auditor who will ask why one system was prioritized over another. Agencies are required to name accountable officials. Contractors are about to be asked the same questions by their customers.

Quantum Academy trains that capability directly: cryptographic discovery and inventory practice, system classification against federal impact ratings, migration sequencing, crypto-agility requirements in procurement, and the evidence a plan needs to survive review. Our programs are private professional credentials rather than accredited degrees or government licenses, and they are built for the people who have to produce the deliverables named in these documents.

Browse the certification programs at quantumacademy.com/. For the migration methodology in full, see pqcframework.org. For the underlying policy analysis document by document, see the federal mandate reference on PostQuantum.com. For where this work leads as a career, see QuantumCareers.com.