Building an Independent Verification Oracle
A checker that shares its subject's assumptions cannot falsify anything. How to build a second implementation that is independent enough to disagree, and how to prove it is.
Learn / topic
Reliable AI Systems
5 published pieces on this topic.
Equal Counts Do Not Prove Two Stores Agree
Why a matching row count is the weakest possible integrity check, and what a ladder of comparisons on identity, content and provenance actually rules out.
Evidence Provenance for Knowledge Systems
What a stored generation has to carry so a claim about it can be re-checked later: captured bytes, ruleset digests, and the difference between recorded and installed.
Production AI Systems Need Refusal Paths
Every stage that can degrade should be able to decline. A synthesis across ingestion, retrieval, extraction and inference of where refusal belongs and what it costs to leave it out.
Re-ingest, Verify, Reconcile, Withdraw
Four operator actions that make a knowledge system maintainable: what each one is allowed to mutate, why the boundaries matter, and how to make every outcome machine-readable.