Key provenance records, giving every key a birth certificate
A key provenance record states where a key came from: which entropy source, inside which boundary, with what health evidence at the moment of generation. Today, for almost every key in production, the honest answer to how you know the entropy behind it was good is that nobody knows.
The question nobody asks yet
Pick a certificate in your infrastructure. Its key was generated at some point by some system drawing on some entropy source. Can you say which source? Whether it was healthy at that moment? Whether the same material seeded any other key?
For essentially every key in production the answer is no, and the reason is not negligence. The information was never captured, because there was nothing to capture it from.
Why it is becoming askable
Supply-chain assurance has moved from an idea to a requirement in a few years. Software has bills of materials. Firmware has signed provenance. Build systems attest to what produced an artefact.
Cryptographic key material is a conspicuous gap in that trend. It is the most security-critical artefact an organisation holds, and it typically carries no provenance at all.
What a record would contain
- Which entropy source instance produced the material, identified rather than described.
- The health measurements at the time of generation, not a certificate from evaluation.
- The boundary the generation happened inside.
- Enough to demonstrate the material was not reused, without recording the material.
The last constraint is the interesting engineering problem: the record must be inspectable without weakening the secret it describes.
Why only a locally verified source can answer
Provenance requires knowing which source, and that it was healthy then. A pooled operating-system entropy pool cannot say which source contributed — mixing is the point. A delivered service can attest to its own operation but not to what happened inside your boundary.
A local source that measures its own model and reports it is the only configuration that can produce the record, which is why LOVE and provenance are the same argument at two levels of abstraction.
Where this goes
Today it is a question we are teaching the market to ask. The organisations that will ask it first are the ones already answering it for software and firmware, and finding that the cryptographic layer underneath has no answer at all. Auditors are the leading indicator.
Keep reading
- Locally Verified Entropy
Entropy monitoring for SOC and NOC teams
Entropy monitoring means treating the health of a cryptographic entropy source as an operational signal: telemetry with thresholds, alarms with defined responses, and retention long enough to answer a question about a key generated months ago. It makes assurance an operations surface rather than a product claim.
- Locally Verified Entropy
Local or delivered entropy, an argument about trust boundaries
Entropy delivered over a network and entropy generated locally differ on four axes: availability, verifiability, sovereignty and attack surface. A shared service is a reasonable engineering choice in some settings, but it moves the trust boundary — and moving a trust boundary is an architectural decision, not a deployment detail.
- Locally Verified Entropy
What is Locally Verified Entropy?
Locally Verified Entropy means the randomness is generated inside the trust boundary that consumes it, and the source reports evidence about its own health while it runs. Quantum physics is how the entropy is made; local and verified are the two properties a customer actually buys.