Each programme in the lab is an answer to a research question. We publish a verifier and test vectors under licence, share the specification under NDA, and label every result with the evidence behind it.
How do you move a live estate from classical to post-quantum keys without breaking the evidence it already produced?
What is the smallest record that lets anyone check, later and offline, what an agent did and under which rule?
How does a machine keep proving what it did when it cannot reach anyone for minutes, or months?
Can a record stay verifiable for decades, after the algorithm that signed it is retired?
How do you bring post-quantum trust to devices that cannot be changed and stay in service for twenty years?
How do you know a security claim is true when much of the code is written with AI?
ML-DSA for the post-quantum leg, Ed25519 for the classical leg. Hash-based signatures (FIPS 205, LMS / XMSS) for archives and firmware.
ML-KEM with X25519 for transport. The key exchange protects the link; it never appears inside a receipt.
Canonical JSON bytes so every verifier signs and checks the same thing; transparency-log proofs for inclusion.
Want to collaborate on research? Architecture, live demonstrations and detailed status are shared in a private technical briefing, under NDA. Write to hello@dcslabs.ai.
Request a technical briefing