AI is reshaping work, public services, and civic life—yet many students still lack access to basic AI literacy. This is not only a technology gap; it is a capability gap.

In 2025, 2.2 billion people remain offline, and digital divides are increasingly defined by speed, affordability, reliability, and skills—conditions that directly constrain what schools can deliver. Meanwhile, foundational learning remains fragile: the World Bank estimates 70% of 10-year-olds in low- and middle-income countries cannot read and understand a simple text.

Policy question: What does workable AI literacy look like under low-resource, low-connectivity constraints—and how can systems scale it without worsening inequity?

This time to walked into two local schools in India: MR High school in Gadhinglaj, Maharashtra and Kolhapur High School ( English medium), near Gandhi Maidan, Kolhapur

Key messages

AI literacy must be “low-bandwidth by design.” If delivery assumes stable connectivity and 1:1 devices, it will systematically exclude those most at risk.

Teacher-mediated scale is the fastest equitable pathway. Short recorded lessons + teacher facilitation + offline activities can reach classrooms that cannot support device-heavy models.

Competence, not dependency. OECD warns that general-purpose GenAI can improve task performance without learning gains and may foster “metacognitive laziness” when pedagogy is absent.

Case study: recorded AI fundamentals lesson used in a government school classroom

To respond to the capability gap, a recorded AI fundamentals training video was produced for underprivileged students and is now being used in a small government school classroom in Kolhapur.

Local delivery leadership: two young founders—Omkar Shinde and Ramkrishna Sawant—have spent the past two years bringing learning opportunities into local classrooms and advancing education equity.

What the lesson teaches (classroom-ready scope)

The lesson introduces:

what AI is (and what it is not)

real-world applications

how models are trained (beginner-level intuition)

an introduction to deep learning (conceptual, age-appropriate)

simple learning games inspired by Code.org activities (low-cost, engaging pedagogy)

self-directed learning pathways using free platforms such as Elements of AI

Why it matters: This is a minimum-viable model that can function where expert teachers, devices, and internet access are limited—without reducing AI literacy to hype or a one-off talk.

Policy implications

1) Treat constrained classrooms as the default design target

With billions still offline, equitable AI literacy must be deliverable with intermittent connectivity, shared devices, and limited teacher preparation time.

2) Make the classroom the “governance layer”

Many AI risks become real in learning settings: misinformation, over-reliance, shallow “copy-paste competence.” OECD’s evidence-based warning about performance without learning gains makes a clear point: AI literacy must build verification habits and self-regulated learning, not tool dependence.

3) Build on open, free learning infrastructure

Open resources reduce cost barriers and enable rapid diffusion. Code.org publicly commits to keeping authored curriculum resources free and openly licensed. Elements of AI positions itself as a series of free online courses designed to demystify AI for broad audiences.

Recommendations for education systems and partners

A. Standardize a “micro-curriculum” package (offline-first)

4–6 sessions: AI foundations + everyday applications

4–6 sessions: practice activities + reflection + simple games

downloadable video + printable activities + facilitator prompts

B. Provide a lightweight teacher facilitation guide

pacing, misconceptions, inclusion strategies

safe-use boundaries and verification routines

“what to do when students rely on AI answers”

C. Use simple, credible indicators (low burden)

pre/post concept checks (AI definitions, training intuition, limitations)

student explanation task: one use case + one limitation

teacher confidence checklist + brief observation notes