AI: Teaching’s New Batmobile
TL;DR.
Across Africa and India, AI is now being used to support teachers in their everyday classroom routine: to score student work in minutes, regroup learners in the same lesson, and get actionable insights without adding extra hours. The result: more teaching time where it matters most, with the students who need it most.
… But first, the news:
… But first, the news:
India and SSA upskill their teachers and entrepreneurs in AI. Kenyalaunched a National AI Skilling Alliance led by Microsoft Kenya and the Kenya Private Sector Alliance. In Ghana, the ASCD completed a nationwide AI training series to empower educators with classroom-ready tools. Google for Startups India launched Startup School: Prompt to Prototype, a two-week hands-on AI skilling program for early-stage founders and entrepreneurs.
Nigeria commits to digitizing all public schools by 2027. The initiative will include providing 60,000 tablets to students and teachers, digitizing school records, and introducing digital systems to track attendance and student performance in real time for better planning.
AI browsers take centre stage. OpenAI launched Atlas, a Mac-based browser that turns ChatGPT into a live co-pilot. Just days later, Microsoftannounced Co-Pilot mode in Edge, while Anthropic released a browser-based coding tool using Claude.
Open-source model sets new records. Moonshot AI’s Kimi K2 has overtaken GPT-5 and Claude Sonnet 4.5 on leading benchmarks, while trending number one on Hugging Face. The model is open-source for research and commercial use, requiring only light-touch attribution for large-scale deployments.
Teacher-AI Tandems in the wild
When we say Teacher–AI Tandems, we mean AI woven into teachers’ daily routines: grading, regrouping, fluency checks, feedback. Teachers remain in control; AI accelerates the loop between instruction → feedback → re-teaching. The result is more teaching time with the students who need it most. The teacher drives the instructional choices; AI speeds up the insight.
Here’s a short of a teacher in Rwanda grading a formative assessment using VisionAI on WhatsApp for the first time.
And here you can see Smart Paper in action, an AI app by EdOptimize to score paper worksheets.
Why this matters now
Teachers are adopting AI where it lightens the load: In India, over 70% of teachers reported using AI tools this year, mainly for lesson prep and assessment. Similar momentum is visible in early pilots across Ghana, Kenya, and Sierra Leone.
Governments are moving from bans to blueprints: Ministries are crafting competency frameworks and adopting national AI policies for responsible, teacher-in-the-loop AI applications in classrooms.
Low-connectivity classrooms are leading innovation: Teachers in low connectivity schools are already using AI for grading, regrouping, fluency checks, and feedback loops - tasks that once consumed their evenings.
Bottom line: The change isn’t conceptual anymore. It’s happening in everyday classrooms.
Three snapshots of the Teacher + AI Tandem
India: Authentic assessment with Smart Paper by EdOptimize
In Rajasthan’s Jodhpur district, 300 teachers used Smart Paper AI to scan and grade 10,000+ handwritten exams from 2,500 students across 55 schools - using only their phones. The tool’s handwriting recognition surfaced common misconceptions and auto-grouped students for reteaching. Over three years, the platform has logged 1 billion data points from 5 million students, showing how grading automation can restore authentic, open-response assessment at scale (Patel, LinkedIn).
India & Africa: ASR for oral reading fluency
In Gujarat and Delhi, Wadhwani AI’s mobile ASR listens to children reading, counts words-per-minute, and flags hesitation or error patterns in multiple Indian languages. These low-cost, phone-based fluency checks now complement EGRA and TaRL, enabling teachers to run short reading checks every week instead of once a term. The approach has already logged millions of ORF samples and is being evaluated for accuracy against human raters.
Rising Academies is testing Amira’s ASR-powered reading app in Ghana, where students read aloud and Amira identifies their reading skill gaps. Teachers can track student usage, mastery, risk, skills gaps, progress over time. This offers a high-quality and consistent loop between student activity and teacher oversight, but constant connectivity can be a challenge.
Sierra Leone: Teacher support chatbots
A new study from Sierra Leone analyzed over 40,000 teacher AI chatbot conversations and found that most queries centered on understanding facts and concepts, lesson planning, assessment, and classroom management. The study showed clear evidence that AI teacher support bots can act as meaningful co-planners and PD companions. The system ran at just a few dollars per teacher per year, proving that high-quality, conversational support for teaching can scale affordably even in low-connectivity settings.
Teacher support tools: Coaching, planning, and insight
Together, these tools help teachers reflect and adjust continuously, informed by real classroom data, not just occasional observations.
Design principles emerging from the field
Paper-first, phone-native workflows scale fastest in low-connectivity settings.
Short, frequent checks (8–12 items) outperform long assessment cycles and enable rapid regrouping.
Teacher-in-control loops: AI proposes; teachers review, confirm, change and decide.
Signals to watch in 2025
Voice-based learning data becomes mainstream: ASR pilots for reading spread to Kenya, Tanzania, Morocco under the EGRA-AI umbrella.
Local-language support accelerates: chatbots supporting conversation in Wolof, Kiswahili, and Hausa will show measurable engagement boosts.
AI-powered coaching pilots deepen in practice: teacher apps begin integrating AI summaries of lesson observations to feed teacher PD and communities of practice (TeachFX pilot in Sierra Leone).
Closing thought
The most meaningful AI in classrooms isn’t replacing teachers. When teachers spend less time marking and more time responding, that time becomes learning: regrouping, reteaching, and supporting the students who need it most. The big opportunity is in faster feedback loops that turn time saved into learning gained.
This edition of the Learning Futures Briefing was led and written by Shabnam Aggarwal, with contributions from Dr. Robin Horn, Ayesha Khan, Sara Cohen, and Faizan Ul Haq.