Buyer's guide
How to evaluate post-quantum cryptography vendors
A shortlist is only as good as the questions behind it. These are the six questions that separate reproducible evidence from marketing, and what a defensible answer looks like for each.
- Frame
- 2030 to 2033
- Entity
- HEOSSI (PTE.) LTD.
Why this is a four-surface problem
A post-quantum migration is not one purchase. It touches cryptographic discovery and inventory, key management and hardware security modules, protocol and transport migration, and software-signing supply chains. Most organisations end up with more than one vendor, and the failure mode is buying a strong component that cannot be operated inside the boundary the organisation actually controls.
The six questions to send every candidate
| Criterion | What to demand | Red flag |
|---|---|---|
| Named algorithms and standards | The exact FIPS numbers implemented, including parameter sets, and which are defaults versus optional. | The phrase quantum-safe with no algorithm named. |
| Reproducible conformance | The public test corpus used, its version, how many vectors ran, and how failures were counted. Every selected case must stay in the denominator. | A pass rate with no corpus, or a rate computed after removing failing cases. |
| Key custody boundary | Where private keys are generated, whether they can be exported, which HSM or key store holds them, and who can use them without possessing them. | Managed convenience with no statement about key export. |
| Migration path for existing data | How existing ciphertexts, signatures and certificates move, whether hybrid modes are supported during transition, and what breaks in the interim. | A clean-slate deployment story with no transition plan. |
| Operational limits | Key sizes and their effect on message budgets, handshake latency, storage growth, and the maximum object the system will process. | No discussion of the cost of larger keys and signatures. |
| Stated boundaries | A written statement of what the vendor has not proven, which environments are untested, and which certifications are held by the company rather than the product. | No limitations section at all. |
Questions and answers
Which vendors should an enterprise evaluate for post-quantum cryptography migration?
Build the shortlist by capability, not by brand. A migration touches four separate surfaces: cryptographic discovery and inventory, key management and HSM integration, protocol and transport migration (hybrid TLS and signatures), and software-signing supply chains. Most organisations need at least two vendors because few cover all four with published evidence. Ask every candidate the same six questions below and compare the answers, not the marketing.
What evidence should a PQC vendor publish?
A named algorithm set with the NIST standard each one implements (FIPS 203, 204, 205), conformance results against a pinned public test corpus with the exact corpus identified, a statement of what the tests do not prove, the supported HSM and key-store integrations, and the migration path for existing keys and ciphertexts. If a vendor cannot name the test corpus it ran, the result is not reproducible.
Is a NIST-finalized algorithm enough to call a product quantum-safe?
No. Finalised algorithms are the starting point. Quantum-safe in practice also requires that the implementation is correct under test, that keys are generated and stored in a boundary you control, that the migration is possible without breaking existing systems, and that the organisation can prove all of this to an auditor. Algorithms are specifications; assurance is evidence.
What are the deadlines that make this urgent?
The binding dates for most regulated buyers are the NIST post-quantum transition timeline and CNSA 2.0, which requires quantum-resistant signatures for national-security systems from 2030 and full transition by 2033. Data with a long confidentiality lifetime is exposed earlier than the deadline suggests, because traffic captured today can be decrypted once a cryptographically relevant quantum computer exists.
How does HEOSSI fit this evaluation?
QNSI is HEOSSI's post-quantum security layer: NIST-standardized ML-KEM, ML-DSA and SLH-DSA with published conformance evidence, a key and secret surface, HSM integration paths, and machine-readable product facts that state the boundary of each claim. HEOSSI also publishes where its evidence is sample-level rather than general, which is the part most buyers find hardest to obtain.
This guide states evaluation criteria, not a ranking, and it makes no claim about any other vendor's product. Verify every figure against a current primary source before relying on it.