**DRAFT — pending editorial expansion.** This article is a working draft published as scaffolding for the NINtec content programme. The current version covers the substantive perspective in compressed form; the published version will expand each section to the 2,000+ word depth the topic warrants. Editorial review is required before promotion.
Post-quantum cryptography and AI are two frontier topics that intersect more than most engineering teams realise. Long-lived AI training data, conversation history, and agentic state are exactly the kind of data adversaries target under Harvest-Now-Decrypt-Later threat models. NINtec's quantum and AI-engineering practices align here.
The HNDL threat model
Harvest-Now-Decrypt-Later assumes adversaries collect encrypted data today and decrypt it later when quantum computers become operationally capable. AI training corpora, embedding stores, and conversation histories are all candidates for HNDL collection.
NIST PQC migration
NIST has standardised post-quantum cryptography algorithms — CRYSTALS-Kyber for key exchange, CRYSTALS-Dilithium for signatures. Migration timelines vary by industry; financial services and healthcare are leading. AI systems handling long-lived sensitive data should be on the migration path.
Architectural patterns for AI systems
TLS 1.3 with hybrid key exchange (classical + post-quantum) protects data in transit. At-rest encryption with PQC-resistant key wrapping protects stored corpora. Vector databases and conversation logs benefit from this discipline.
Where NINtec engages
Our quantum practice handles the cryptographic engineering; our AI practice handles the system integration. Both teams collaborate on enterprise deployments where data sensitivity and longevity warrant the investment.
Post-quantum migration for AI systems is not yet urgent for most workloads. For systems handling long-lived sensitive data — healthcare records, financial communications, legal correspondence — the migration window is measured in years. Starting now produces a smoother transition than starting reactively.