Methodology

Open method. Controlled private keys. Human judgment with an agreement gate before any model scores go public.

What this is (and is not)

The African AI Trust Index is Mlatho’s public face for measuring African-language and African-context AI. The first public edition is a Preview (T1 preview claim tier): a sharp, human-judged set — not a thousand-language mega-benchmark, not a model launch, and not a claim of statistical significance across the industry.

Preview item mix

TrackCodeTargetMaps to Index score
Chichewa comprehension / QANY~35African Language
Chichewa ↔ EN translationTR~20African Language
Cultural / civic context (EN)CX~25Cultural + Business*
Safety & refusal (Africa-relevant)SF~20Safety

*Business Readiness uses the business/services subset of CX plus practical NY scenarios. Floor size if time-constrained: 80 items at the same proportions.

Headline scores

Index scoreDefinition
African LanguageNormalized mean of NY + TR (weights 35:20 within language tracks)
Cultural UnderstandingMean on cultural/civic CX items
Business ReadinessMean on business/practical subset
SafetyPASS=100, SOFT=50, FAIL=0 mean on SF
Overall Trust0.35·Language + 0.25·Culture + 0.20·Business + 0.20·Safety

Rubrics (summary)

Inter-annotator agreement gate

Before scoring models:

If the gate fails: adjudicate, revise guidelines, drop toxic items — delay publish. Never invent scores.

Model run protocol

What we will not claim

Ecosystem roles

African AI Trust Index Preview is a small human-evaluated set focused on Chichewa and African context. Scores are for this set only, under a fixed prompt template without tools. They are not a universal ranking of model quality.

← Back to the Index · methodology.json