Experts prefer ChatGPT's rice advice to extension agents' in Kano

Four blinded experts rated ChatGPT's answers to 32 rice farmers' questions from Kano State above those of extension agents and preferred them in 78% of cases. Some of its advice was wrong.

  • Not rated a person

    Farmers did not receive the chatbot's answers, so no change in their practices or yields is reported.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Reported and intended benefits.
Evidence

foundstudiedreportedclaimed

An independent peer-reviewed study supports the benefit.

Compared with

Answers by six Kano extension agents, three of them with extension materials, to the same 32 questions. Measured.

Without the system
7%
With it
40%
The system’s part
Blinded experts scored ChatGPT's answers against those of six extension agents to the same 32 questions, so the comparison isolates the chatbot's answers. The study measured answer quality only.
Limits
Not yet checked.
Costs and harms
Not yet checked.
Later
Not yet checked.

Download data Cite Suggest a correction

Benefit type
Improved Knowledge Discovery
System
ChatGPT
Made by
OpenAI
Used by
Africa Rice Center (AfricaRice) researchers, in a blinded test with extension agents in Kano State
Where
Kano State, Nigeria
Stage
Tested in a laboratory
Date
, when the source was published
Kind
Aggregate result
Broad benefit
Providing Consultation and Advice to Others, Interpreting the Meaning of Information for Others
Collections
Cross-border records
Last revised
31 October 2026

Sources

  1. Peer-reviewed study, by authors with no tie to the system.

    Ali Ibrahim, Kalimuthu Senthilkumar and Kazuki Saito. Evaluating responses by ChatGPT to farmers' questions on irrigated lowland rice cultivation in Nigeria. Scientific Reports 14: 3407, 10 February 2024.

    Overall, chatbot responses were rated significantly higher quality than EAs' responses. Chatbot responses received the best score nearly six times as often as the EAs' (40% vs. 7%). The evaluators preferred chatbot responses to EAs in 78% of cases.

Details

AI system

Workflow

The researchers put each farmer question into a fresh ChatGPT session with no earlier prompts. Four local rice experts scored all seven answers to each question without knowing who wrote them.

Follow-up

Conditions for replication

Advice on input rates and timing must be checked against local recommendations, where the chatbot erred. Farmers must be able to use long answers, which the study did not test with them.

Entry details

Sector

Agriculture / extension advice.

Purpose

OpenAI built ChatGPT to answer questions and write text in conversation with its users.

Not yet checked: model, release, provider, role, configuration, inputs and outputs, autonomy, human oversight, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). Experts prefer ChatGPT's rice advice to extension agents' in Kano (AIBD-0858). Artificial Intelligence Benefits Database, edition of 31 October 2026. https://petervartanian.xyz/aibd/records/AIBD-0858.html (archived at https://doi.org/10.7910/DVN/AADSII, version 4.0)

BibTeX Data Field definitions