The Pragati: AI for Impact initiative, a collaborative effort by Meta and The/Nudge Institute, is harnessing artificial intelligence to bridge educational gaps in India's most underserved regions, empowering entrepreneurs and organisations to deliver context-aware learning solutions that prioritise foundational skills, teacher support, and livelihood preparation. Launched amid the government's ambitious IndiaAI Mission, which allocates ₹10,000 crore to democratise AI across sectors, Pragati spotlights bottom-up innovations from groups like the OpenLinks Foundation, Saarthi Education, and The Apprentice Project (TAP). These entities are reimagining education at the "last mile"—rural and low-resource classrooms where over 260 million students grapple with disparities highlighted by the Annual Status of Education Report (ASER) 2024.

The report, surveying 600,000+ children across 18,000 villages, revealed that only 37% of Grade 3 students can read a Grade 2 text, and fewer than 30% handle basic subtraction, despite high enrolment rates exceeding 95% for ages 6-14. By integrating AI with local insights, Pragati aims to foster responsive ecosystems that evolve through real-world interactions, aligning with the National Education Policy (NEP) 2020's emphasis on equitable, technology-driven learning.

At the heart of this initiative is the OpenLinks Foundation's Vinoba platform, an adaptive AI tool designed to augment teachers in diverse Indian classrooms by learning from their on-ground adaptations rather than imposing rigid templates. Vinoba assists with lesson planning by analysing how educators reorder activities, simplify concepts, or tweak engagement strategies based on student responses, gradually refining its algorithms to offer hyper-localised recommendations. This bottom-up approach addresses a core limitation of top-down digital platforms like DIKSHA and the National Digital Education Architecture (NDEAR), which provide scalable infrastructure but often overlook regional nuances such as multilingual needs or varying pedagogical styles. In states like Haryana, where AI literacy is now compulsory from Class 9, tools like Vinoba could amplify teacher training by incorporating behavioural data from thousands of sessions, ensuring that AI evolves as a collaborative partner rather than a one-size-fits-all directive. Early pilots in Uttar Pradesh and Bihar have shown a 25% improvement in lesson efficacy, as measured by student engagement metrics, demonstrating how such systems can personalise education without requiring advanced digital infrastructure.

Saarthi Education's AI-driven numeracy model tackles another persistent bottleneck: the feedback loop in resource-constrained government schools, where paper-based assessments overwhelm teachers and delay remedial interventions. The offline-capable tool scans handwritten worksheets using optical character recognition, identifies conceptual errors in real time, and auto-generates tailored exercises—transforming static evaluations into dynamic learning aids. This innovation is particularly vital in rural areas, where internet penetration hovers at 30-40%, allowing AI to operate via simple mobile devices and reducing teacher workload by up to 40%, per internal evaluations. By focusing on foundational numeracy—an NEP priority linked to broader economic outcomes like reduced dropout rates—Saarthi not only boosts learning continuity but also generates anonymised datasets on common misconceptions, which can inform national curricula. In tandem with initiatives like the National Mission on Foundational Literacy and Numeracy, this model exemplifies Pragati's philosophy of "co-evolutionary AI", where technology learns from human errors and iterations to create inclusive pathways for marginalised learners, including girls and first-generation students.

Also Read: Dustin Moskovitz Steps Down as Asana CEO After 13 Years Citing Exhaustion

The Apprentice Project (TAP) extends Pragati's impact beyond classrooms into livelihood skilling, deploying an AI-powered WhatsApp chatbot to deliver bite-sized, personalised modules on creativity, communication, and financial literacy—skills essential for India's youth entering a gig-driven economy projected to add 90 million jobs by 2030. Unlike traditional e-learning apps, TAP's system employs behavioural analytics to monitor engagement patterns, such as drop-off triggers or interest spikes, dynamically adjusting content delivery to sustain motivation among underserved 18-25-year-olds. Operating in vernacular languages across 10+ states, it has reached over 50,000 users since 2024, with completion rates 35% higher than generic platforms due to its observatory-like function that treats users as co-creators of knowledge. This forward-looking approach aligns with the government's Skill India Mission, providing policymakers with granular insights on skill acquisition barriers, from urban migrants to rural artisans, and fostering entrepreneurial ecosystems where AI enables scalable, hyper-local training.

Pragati's holistic framework—spanning classroom augmentation, remedial diagnostics, and vocational upskilling—illustrates a paradigm shift from AI as a deployment tool to one of mutual adaptation, where diverse stakeholders co-build intelligence that resonates with India's socio-cultural mosaic. As the initiative scales through partnerships with state education departments and philanthropies, it holds promise for narrowing the 42% foundational learning gap identified in ASER, ultimately cultivating a generation equipped for the AI-augmented workforce. By blending policy ambition with grassroots ingenuity, Pragati not only accelerates India's "AI for All" vision but also redefines impact measurement: success lies not in metrics of reach but in the depth of dialogue between humans and machines, ensuring technology serves as an enabler of equity rather than an echo of uniformity.

Also Read: Apple’s Siri AI Chief Jumps to Meta, Casting Doubt on Future of Smart Assistant