evidoria

← Back to browse

Good practice

Jugalbandi — A Multilingual Generative-AI Chatbot Connecting Rural Indians to Government Welfare Schemes

India · Nuh · See the India profile

AI4Bharat, OpenNyAI and Microsoft Research built Jugalbandi, a WhatsApp/Telegram chatbot using Bhashini speech AI and GPT-4 to explain scheme eligibility in 10 of India's 22 official languages, piloted since April 2023 in rural Haryana.

Jugalbandi — A Multilingual Generative-AI Chatbot Connecting Rural Indians to Government Welfare Schemes

Details

Promoter
AI4Bharat (IIT Madras) and OpenNyAI, with Microsoft Research India
Period
April 2023–present (pilot reach as most recently reported)
Keywords
generative AI, multilingual NLP, e-government, rural digital inclusion, government scheme access

Description

Jugalbandi is a generative-AI chatbot built to help rural Indians navigate the roughly 20,000 government welfare schemes that exist across India's central and state governments — most of which are poorly known and documented only in English or Hindi. Developed by AI4Bharat (a government-backed language-AI initiative hosted at IIT Madras) and the open-source legal-tech collective OpenNyAI, with reasoning models supplied by Microsoft Research India through Azure OpenAI Service, the assistant was first introduced to villagers in Biwan, in Haryana's Nuh district, in April 2023.

The system layers Microsoft's GPT-4-based models on top of Bhashini — India's national public speech-and-translation AI mission — so that a resident can ask a question by voice or text in their own language and receive a spoken or written answer in that same language, even when the underlying scheme information exists only in English. Access runs through WhatsApp and Telegram, chosen because both are already in wide use in rural India, and the underlying 'Jugalbandi Manager' platform has been released as open-source software so other teams can adapt it to new languages or channels.

As of the initiative's most detailed public reporting, Jugalbandi covered 10 of India's 22 scheduled official languages and 171 of the country's approximately 20,000 government schemes — helping users understand which programmes they may be eligible for, what benefits they offer, and what documents are required. Coverage has been described by its developers and by independent technology press as expanding steadily from that initial village pilot, but no party has published usage volumes, resolution accuracy, or user-satisfaction figures for the deployment, so its measured impact on citizens' actual access to benefits remains undocumented.

The project's evidence base is therefore one of demonstrated technical reach rather than proven outcomes: real, verifiable facts about language and scheme coverage sit alongside an acknowledged gap in independent evaluation. Its open-source release and reliance on India's public Bhashini infrastructure — rather than a proprietary, closed system — are notable transparency strengths even without impact data.

Read the full analysis: https://www.forbes.com/sites/saibala/2023/05/31/microsoft-has-launched-jugalbandi-a-new-generative-ai-app-for-india/

Implementation

Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful