IIT Madras and Tech Partners Roll Out Indic AI Tutoring Models and School Training Initiatives

Expanding AI in Indian Classrooms Beyond English

Educational institutions across India are accelerating the development and deployment of artificial intelligence tools designed specifically around Indian languages and domestic curricula. In classrooms where instruction has traditionally depended on standard textbooks and English-dominated digital platforms, new sovereign technology initiatives are opening pathways for students and educators to interact in regional tongues. The primary objective is ensuring that learners can seek academic guidance, ask conceptual questions, and receive feedback directly in their native languages without first having to translate their queries into English ^[3].

At the forefront of this shift is Bodhan AI, the Centre of Excellence in AI for Education incubated at the Indian Institute of Technology (IIT) Madras. Working in partnership with AI4Bharat, Bodhan AI has introduced an integrated ecosystem comprising open foundational models, a dedicated Student Tutor Bot, and an administrative Teacher Assistant Bot. This infrastructure is envisioned as part of the broader Bharat EduAI Stack, serving as sovereign digital infrastructure designed specifically to cater to the diverse linguistic and structural needs of the Indian schooling system ^[3].

Open Foundational Indic Models and the Bharat EduAI Stack

To power multilingual learning at scale, Bodhan AI has launched four foundational AI models covering automatic speech recognition, optical character recognition (OCR), text-to-speech, and machine translation. These models are structured to process diverse Indian languages and scripts, serving as open-weight Digital Public Goods that are also accessible through hosted application programming interfaces (APIs) built on sovereign infrastructure ^[3].

The speech recognition model accommodates 27 Indian languages, while both the optical character recognition and text-to-speech engines support 23 languages each. The translation engine spans 22 Indian languages. By providing these open-weight resources, the initiative establishes a practical technological backbone for school environments. These tools allow digital educational software to parse handwritten student assignments, facilitate seamless translation of learning materials across Indian languages, and support voice-driven interactions so that learners are not restricted exclusively to keyboard input ^[3].

Curriculum-Aligned Student Tutor Bot for Classes 6 to 12

Direct support for secondary school students is structured through Bodhan AI's newly launched Student Tutor Bot, specifically engineered for learners in Classes 6 through 12. Rather than providing unverified web answers, the bot is directly aligned with curricula established by the National Council of Educational Research and Training (NCERT) and various State Councils of Educational Research and Training (SCERT) ^[3].

Learners can converse with the bot through voice or text in 22 Indian languages. The system provides clear explanations, relevant examples, guided problem-solving sequences, assessments, and specialized interactive features designed for calculations and diagrams. By combining vernacular speech and text understanding with official school curricula, the platform seeks to democratize personalized academic tutoring across diverse economic and geographic settings ^[3].

Teacher Assistant Bots and Classroom Workflows

Parallel to student-facing applications, technology developers are emphasizing the alleviation of administrative burdens carried by school educators. Bodhan AI's launch includes a dedicated Teacher Assistant Bot capable of generating tailored lesson plans, classroom worksheets, revision guides, practice quizzes, and homework modules structured around specific subjects, topics, grade levels, and difficulty parameters ^[3].

The software also allows teachers to upload student submissions for structured evaluation against customized assessment criteria. To preserve pedagogical authority, developers emphasize that teachers maintain ultimate discretion over classroom delivery. Educators can inspect, edit, regenerate, or discard any AI-suggested content, ensuring that automated tools function strictly as instructional aids rather than autonomous replacements for human mentorship ^[3].

State Implementations and Institutional Collaborations in Assam

Complementing software releases, state education authorities are moving quickly to pilot these emerging systems inside government schools. In northeastern India, Samagra Shiksha Assam recently signed official agreements with IIT Madras alongside six corporate entities to train teachers in utilizing AI platforms for structured lesson planning ^[6].

According to Assam Education Minister Ranoj Pegu, the state program aims to integrate technological capabilities into public education and equip teachers with contemporary tools for instructional planning. IIT Madras will supply a dedicated AI educational platform, with the initial rollout focused on preparing educators to formulate lesson frameworks. Additionally, the Assam government entered into a memorandum of understanding with Google AI to support automated lesson design and classroom preparation. The initial phase will operate as a pilot project spanning 70 schools across Assam, establishing practical benchmarks for school-level AI adoption ^[6].

Nationwide Educator Training and Classroom Guardrails

Coinciding with state-level deployments, broader professional development programs are taking shape across the country. Ahead of Teachers' Day, Google announced its nationwide "Google AI Educator Series," delivering free virtual and hands-on workshops aligned with the National Education Policy (NEP) 2020 to build AI literacy among K–12 and higher education faculty ^[4]. Conducted in multiple Indian languages—including Hindi, English, Assamese, Telugu, Punjabi, and Marathi—the curriculum explores tools like Gemini and Guided Learning to streamline administrative routines and enhance assessment feedback while prioritizing educational science and human connection ^[4].

Meanwhile, teachers in urban centers are actively integrating generative AI into everyday classroom practices while introducing necessary guardrails. Educators are leveraging tools to produce differentiated worksheets targeting varying difficulty levels and creative visual aids, while simultaneously implementing strict assessment policies such as conducting in-class assessments and enforcing prompt-engineering guidelines to prevent passive student dependence on automated text generators ^[8]. Together, the release of sovereign open-weight models from IIT Madras, localized state pilot deployments, and national upskilling initiatives highlight a concerted push to embed artificial intelligence into India's public and private educational fabric responsibly ^[3][4][6][8].

Sources

1. IIT Madras’ Bodhan AI launches AI tutor for Classes 6-12 in 22 Indian languages - India Today www.indiatoday.in
2. Teachers’ Day 2026: Google rolls out nationwide AI training sessions for Indian educators - EducationTimes.com www.educationtimes.com
3. Assam to introduce AI tools in school education, pilot project to cover 70 schools - Assam to introduce AI tools in school education, pilot project to cover 70 schools - www.indiatodayne.in
4. From worksheets to videos: Teachers are using AI tools to teach and upskill students and themselves - The Hindu www.thehindu.com