Capacity–Locality Certification (CLC)
Structural certification framework for predictive systems under information capacity and locality limits.
This repository contains:
- Theory — certified-limits note (docs/papers/certified-limits-note.md)
- Whitepaper — industry whitepaper (docs/whitepapers/clc-whitepaper.md)
- Specification — NDGate and certificate schemas (docs/spec/*.md)
- Examples — templates and sample certificate (docs/examples/*.md)
- Executive summary — plain-English intro (docs/EXECUTIVE.md)
This also serves as the basis for the public reference site:
👉 https://inaciovasquez2020.github.io/capacity-locality-certification/
What is CLC?
CLC defines when an AI predictive system is:
- Allowed to predict (liquid regime),
- Required to abstain (frozen regime),
- And how to produce a machine-readable certificate of that determination.
It is intended as a public reference standard, not tied to any architecture or vendor.
License
This work is published under the MIT License.