---
title: 'Edo State schools / World Bank study: Teacher-facilitated GPT-4 tutoring'
slug: nigeria-ai-tutoring
stable_id: ae70ae8b90d164ae
company: Edo State schools / World Bank study
function_code: enrollment
pattern_codes:
  - creator_to_judge
evidence_strength: verified
publication_tier: showcase
freshness: current
reviewed_at: '2026-08-23'
updated_at: '2026-08-22'
source_quality_summary: 2 independent; publication outcomes are verified.
caveat_summary: >-
  World Bank Policy Research Working Paper, not formal peer review; 654 observations in
  the main outcome tables; short, supervised intervention among interested students.
collections: []
bundle_version: 1.0.0
bundle_fingerprint: sha256:c23c6cc2b88153d008ea8fda928f632ce0011fc2d4c5036672a16e5d895bab93
canonical_url: https://brianletort.ai/transformations/nigeria-ai-tutoring
---

# Edo State schools / World Bank study: Teacher-facilitated GPT-4 tutoring

A general model becomes a tutor only inside structured pedagogy.

Function: Enrollment. Patterns: Creator to judge. Evidence: verified.

Freshness: current. Reviewed: 2026-08-23. Updated: 2026-08-22.


Source quality: 2 independent; publication outcomes are verified.

## Before

1. **Public-school student** — Continues business-as-usual classes and study without the six-week after-school intervention. (control: Randomized control condition; the paper does not claim a separate prior tutoring queue.)
2. **Classroom teacher** — Delivers ordinary curriculum and grades school assessments. (control: Human pedagogy and assessment authority.)

## After

1. **Student pair using Microsoft Copilot/GPT-4** — Completes curriculum-aligned English, AI, and digital-skills activities in twelve 90-minute lab sessions. (control: Carefully crafted prompts; pair work; intermittent connectivity/electricity are documented failure modes.)
2. **Teacher facilitator** — Introduces activities, monitors model use, and redirects shortcutting or unsuitable responses. (control: Teacher retains pedagogy; outcomes measured independently after six weeks.)

## Decision rights

Teachers own pedagogy and monitoring.

## Exception path

Internet, electricity or unsuitable output is handled by teacher.

## Outcomes

- **Endline assessment** (verified): Control business-as-usual → 0.310 SD weighted total score; 0.238 SD English skills (abstract reports 0.23; narrative rounds to 0.24). World Bank Policy Research Working Paper, not formal peer review; 654 observations in the main outcome tables; short, supervised intervention among interested students.

## Executive lesson

The model made supervised practice abundant; it did not replace the teacher.

## Anti-pattern

Do not generalize to unsupervised access.

## Questions for leaders

- Where is the operating threshold set and who can override it?
- What measured result would trigger rollback or retraining?
- Which residual decisions must remain human-owned?

## Sources

- [From Chalkboards to Chatbots](https://documents1.worldbank.org/curated/en/099548105192529324/pdf/IDU-c09f40d8-9ff8-42dc-b315-591157499be7.pdf) — World Bank
- [How AI tutors improved learning in Nigeria](https://voxdev.org/topic/education/how-ai-tutors-improved-learning-nigeria) — VoxDev