evidoria

← Back to browse

Good practice Imported

Farmer.Chat in Ethiopia — Digital Green's Government-Backed AI Advisory Pilot Scales to a 200,000-Farmer Impact Evaluation

Ethiopia · Addis Ababa · See the Ethiopia profile · See the Addis Ababa profile

Evidence: Descriptive / self-reported Top 62% 53/100 · Ask Evidence Copilot about this practice

Digital Green's Farmer.Chat AI advisory tool, run with Ethiopia's Ministry of Agriculture, scaled from a 3-district pilot to a 15-district, 200,000-farmer evaluation with Bordeaux School of Economics and Addis Ababa University measuring yields, income and cost-effectiveness.

3
Pilot districts (woredas)
32
Pilot communities
100
Government development agents trained (pilot)
15
Impact-evaluation districts
~200,000
Farmers covered by evaluation
3,000
Extension agents in evaluation
Farmer.Chat in Ethiopia — Digital Green's Government-Backed AI Advisory Pilot Scales to a 200,000-Farmer Impact Evaluation

Details

Maturity
Scaling
Promoter
Digital Green, in close collaboration with Ethiopia's Ministry of Agriculture
Period
April 2025 – December 2027 (pilot + impact evaluation)
Keywords
agriculture, rural extension services, digital public goods

Context

Since April 2025, Digital Green has piloted Farmer.Chat, a GPT-4-based conversational AI agricultural advisory tool in Afaan Oromo, in close collaboration with Ethiopia's Ministry of Agriculture.

Activities

The pilot ran across 3 woredas of Oromia region and 32 communities, training 100 government development agents; funded by the Fund for Innovation in Development, it has since scaled into a formal impact evaluation covering 15 districts, roughly 200,000 farmers and 3,000 extension agents.

Conclusions

The evaluation is designed to measure effect on adoption of farming practices, crop yields, incomes and cost-effectiveness versus Ethiopia's existing agent-only extension model; results were not yet public as of this review.

Implementation
Indicative cost
Medium (€50k–€500k)
Time to results
Long (> 3 years)
Staffing & skills
Digital Green team, 3,000 government extension agents, Bordeaux School of Economics / Addis Ababa University / RIDA researchers

Conditions for success

  • Integration with existing government extension network
  • Multi-language advisory support

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful