evidoria

← Back to browse

Good practice

aprendIA — IRC's AI Chatbot Bringing Learning Support to Crisis-Affected Northeast Nigeria

Nigeria · Maiduguri · See the Nigeria profile

The International Rescue Committee's aprendIA chatbot uses OpenAI's models on WhatsApp and SMS to coach teachers and caregivers of crisis-affected children in Borno, Adamawa and Yobe. Initial testing found 78% of educators found it intuitive, used 3-4 times weekly.

aprendIA — IRC's AI Chatbot Bringing Learning Support to Crisis-Affected Northeast Nigeria

Details

Promoter
International Rescue Committee (IRC) — Airbel Impact Lab, with OpenAI
Period
2024–ongoing (Nigeria testing 2025–2026)
Keywords
AI chatbot, refugee and crisis-affected education, non-formal learning, teacher and caregiver coaching

Description

The International Rescue Committee's (IRC) Airbel Impact Lab built aprendIA in response to a global crisis: 224 million crisis-affected school-age children, 72 million of them out of school entirely. aprendIA runs on RapidPro (UNICEF's open-source messaging engine) integrated with OpenAI's ChatGPT, delivering AI-driven pedagogical support over WhatsApp and SMS — channels that need no smartphone or reliable internet, which matters in the conflict-affected states of Borno, Adamawa and Yobe in northeast Nigeria.

The chatbot delivers IRC's own pedagogical frameworks (Healing Classrooms, Learning through Play, Teaching at the Right Level) through chat, and is being tested in both formal and non-formal learning centres. It is designed to be adapted to a new acute-crisis context within 14 days; the Bangladesh version was built and launched in under four weeks.

In initial testing, educators used the tool three to four times a week for 30-40 minute sessions, completing two to three courses; 78% found it intuitive, and users reported higher learner engagement and reduced workload. Backed in part by a $250,000 OpenAI award, IRC aims to reach 10,000 teachers and caregivers, indirectly affecting roughly 500,000 students, en route to a longer-term target of one million learners by the end of 2026 across Nigeria, Bangladesh and the Democratic Republic of Congo.

These figures are IRC's and OpenAI's own usage and satisfaction metrics from early testing, not independently peer-reviewed learning-outcome data — no test-score or literacy/numeracy gain evidence has yet been published, and the claimed cost advantage ("less than half the cost of traditional teacher training") is IRC's internal estimate.

Read the full analysis: https://airbel.rescue.org/projects/aprendia/

Implementation

Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.

Data sources

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

Attachments

Similar practices you may find useful