The Lawnise Trust Index methodology is being prepared for public release. Until then, the working draft, its review history, and the evidence-anchored adjudication approach are documented here for procurement review.
The methodology describes how Lawnise scans public AI engines, deduplicates observations against context snapshots, and adjudicates findings against evidence anchored to public, dated source material. It documents PII redaction with Malaysian recognizers, the separation between customer configuration and benchmark reference data, and the right-to-reply path for institutions named in published findings.
The draft is reviewed on a fixed cadence and revised against platform reality. Material revisions are versioned, dated, and attributed. The published methodology will carry the same audit metadata as this draft entry.