Named
The business or category is directly identified in the answer in a way that a reader can recognise.
Kenyan AI evidence
Kenyan AI visibility can be measured before it can be improved.
In the lab’s founding comparison, a Nairobi tour operator appeared in several AI answers while a similar coastal operator stayed unnamed, even with a working site, reviews and licence references. Kivuli Index Lab studies those gaps across Kenyan sectors, counties, languages and business forms. The lab records when businesses are named, skipped, blurred or displaced, then turns repeated prompt observations into benchmark frames that businesses and public bodies can actually use.
Anchor classification
The four visibility states of a Kenyan business in AI answers
Every observation begins as an answer state — a qualitative reading of what the answer did before the lab asks why. It is a typology, not a score.
Skipped
The business or category is absent even though the prompt makes it relevant to the comparison.
Blurred
The answer compresses a specific Kenyan business, cooperative, county or enterprise form into a generic label.
Displaced
A different reference takes the place that the tested Kenyan business or category could reasonably have occupied.
Method
How the lab reads visibility
The lab treats every answer as an observed state: a business, sector, county or business form is named, omitted, described inaccurately or replaced by another reference. Samples are built descriptively across sectors, counties, English and Swahili wording, formal and informal businesses, and online evidence levels. Repeatability matters because another reader should be able to reconstruct the prompt type, engine, language, date and classification logic.
Full methodologyStatus
In focus across Kenya now
The lab is studying Nairobi skew, county-level omissions, English-Swahili divergence and the weaker AI visibility of informal, cooperative and mobile-first enterprises. Current work also tracks when seasonal operations, licence wording and review scarcity change how Kenyan businesses are described.
01
Nairobi skew
02
County omissions
03
English ↔ Swahili
04
Informal & mobile-first
From the research bench
Studies in the corpus
The research index gathers notes on under-representation, sector differences, omission versus inaccuracy, formal-informal gaps, engine consistency and county-linked data. Materials are organised around practical visibility states, so a reader can move from a single answer oddity to the wider benchmark frame behind it.
Direction 1 · Representationoverview synthesis
Are Kenyan businesses underrepresented in AI answers?
A benchmark note on when Kenyan businesses vanish, blur or lose place in AI answers despite being relevant to the query.
Read note
Direction 1 · Representationrepeatable run
Does AI visibility pull Kenyan business toward Nairobi?
A county-skew benchmark on whether AI answers name Nairobi businesses more readily than comparable firms in Mombasa, Kisumu, Nakuru and rural counties.
Read note
Direction 2 · Language & formrepeatable run
Does Swahili wording change Kenyan sector visibility?
A language-variance study of how English and Swahili prompts can change Kenyan sector representation in AI answers.
Read note
Direction 3 · Freshness & benchmarkobservation
How do AI answers represent Kenyan cooperatives and SACCOs?
A close look at how Kenyan cooperatives, SACCOs and group enterprises become named, skipped, blurred or displaced in AI answers.
Read note
Contact
Measure the pattern before trying to fix the page.
Kivuli Index Lab helps readers see where Kenyan business evidence is present, thin, distorted or missing.
Contact the lab