IND-001

PM-KISAN Kisan e-Mitra AI Chatbot

Download PDF
India South Asia Lower middle income Full Production Deployment Confirmed

Department of Agriculture and Farmers Welfare (DA&FW), Government of India

At a Glance

What it does LLMs for content creation, transformation and modality conversion — User communication and interaction
Who runs it Department of Agriculture and Farmers Welfare (DA&FW), Government of India
Programme Pradhan Mantri Kisan Samman Nidhi (PM-KISAN)
Confidence Confirmed
Deployment Status Full Production Deployment
Key Risks Model-related risks
Key Outcomes 2.
Source Quality 5 sources — News article / media, Government website / press release, Other

Kisan e-Mitra is a voice-based AI-enabled chatbot developed to assist farmers with responses to their queries regarding the Pradhan Mantri Kisan Samman Nidhi (PM-KISAN) scheme, one of India's largest social assistance programmes, which provides income support of INR 6,000 per year to over 90 million farming families across India. The chatbot was launched on 21 September 2023 by the Union Minister of State for Agriculture and Farmers Welfare, Shri Kailash Choudhary, and is described as the first AI chatbot integrated with a major flagship scheme of the Government of India.

The primary purpose of Kisan e-Mitra is to provide farmers with timely and accurate information about their PM-KISAN application status, payment details, ineligibility status, and any updates or modifications to the scheme. It functions as a comprehensive grievance redressal tool, offering farmers a one-stop solution for resolving queries that were previously handled through manual processes. Users can interact with the chatbot through three input methods: text input, voice input in their preferred language, and selection from a list of frequently asked questions. The system processes 49 distinct query categories using predefined response templates that are tailored with beneficiary-specific data to generate personalised replies relevant to each farmer's individual circumstances.

The technical architecture of Kisan e-Mitra employs natural language processing to interpret the intent of farmer queries and retrieve relevant information from the PM-KISAN scheme database. Following a significant enhancement in February 2024 led by Wadhwani AI, the system was upgraded to incorporate a Large Language Model for generating responses, improving the quality and accuracy of its outputs. Multilingual support is provided through integration with Bhashini, the National Language Translation Mission, which uses the IndicTrans2 model for translation capabilities. This integration enables the chatbot to accept audio and text queries in the user's preferred language and to provide responses in the local language in both text and audio formats. During Phase I, the chatbot was launched in five languages: English, Hindi, Tamil, Bengali, and Odia. It was subsequently expanded to 11 languages, adding Malayalam, Gujarati, Punjabi, Telugu, Marathi, and Kannada, with plans to eventually support all 22 official Indian languages.

The development and enhancement of Kisan e-Mitra has involved multiple institutional partners. The initial development was supported by EkStep Foundation and Bhashini, operating under the Department of Agriculture and Farmers Welfare (DA&FW). Following the February 2024 enhancement, Wadhwani AI played a central role in developing the upgraded AI and ML capabilities, with support from Google.org, the National Informatics Centre (NIC), and Samagra Governance. The chatbot is accessible through the PM-KISAN official website and the PM-KISAN mobile application, which also features face authentication-based e-KYC technology.

In terms of scale and usage, Kisan e-Mitra has demonstrated substantial uptake since its launch. Following the February 2024 enhancement, the system answered 2.69 million queries in just 186 days, representing a 668 percent increase compared to manual grievance redressal processes. Government reporting as of April 2025 indicated the chatbot was handling over 20,000 farmer queries daily and had answered more than 92 lakh (9.2 million) queries in total. By July 2025, this figure had risen to over 95 lakh (9.5 million) queries resolved. The Wadhwani AI impact dashboard reports over 5 million grievances resolved, more than 290,000 unique farmers served, and over 400,000 farmers enrolled, with geographic coverage spanning 28 states and 8 Union Territories, and the service operates on a 24/7 basis.

The system operates as an informational and grievance support tool rather than a decision-making system. It does not make automated eligibility or payment decisions; its function is limited to providing information lookup and query resolution based on existing scheme records. The decision criticality is therefore assessed as low, as incorrect or incomplete responses would delay information access but would not directly affect benefit eligibility or payment amounts. No primary source has been identified describing escalation pathways to human agents or specific human oversight protocols governing the chatbot's operations.

From a risk perspective, the primary concerns relate to the accuracy and reliability of the chatbot's responses, particularly given its reliance on NLP and LLM components that could generate incorrect or misleading information. Transparency to users regarding the AI-generated nature of responses and the limitations of the system is a further consideration. Accessibility remains a concern given that language coverage, while expanding, does not yet encompass all 22 official Indian languages, potentially excluding farmers who speak less widely supported languages. The chatbot's dependence on the Bhashini translation infrastructure and the IndicTrans2 model introduces an upstream dependency on the quality and availability of these services. No public documentation has been identified regarding formal risk assessments, data protection impact assessments, bias audits, or independent evaluations of the system's performance or accuracy across different language and demographic groups.

Classifications follow the DCI AI Hub Taxonomy. Hover over field labels for definitions.

Social Protection Functions

Implementation/delivery chain
Accountability mechanisms primaryOutreach/communications/sensitisation
SP Pillar (Primary) The social protection branch: social assistance, social insurance, or labour market programmes. Social assistance
Programme Name Pradhan Mantri Kisan Samman Nidhi (PM-KISAN)
Programme Type The type of social protection programme, classified under social assistance, social insurance, or labour market programmes. View in glossary Poverty targeted Cash Transfers (conditional or unconditional)
System Level Where in the social protection system the AI is applied: policy level, programme design, or implementation/delivery chain. View in glossary Implementation/delivery chain
Programme Description PM-KISAN is a central government scheme providing income support of INR 6,000 per year in three equal instalments to over 90 million farming families across India, administered by the Department of Agriculture and Farmers Welfare (DA&FW). Kisan e-Mitra is the AI chatbot integrated into the PM-KISAN digital infrastructure to handle farmer queries and grievance redressal.
Implementation Type How the AI output is produced: Classical ML, Deep learning, Foundation model, or Hybrid. Affects validation, compute requirements, and governance profile. View in glossary Hybrid
Lifecycle Stage Current stage in the AI lifecycle, from problem identification through to monitoring, maintenance and decommissioning. View in glossary Monitoring, Maintenance and Decommissioning
Model Provenance Origin of the AI model: developed in-house, adapted from open-source, commercial/proprietary, or accessed via third-party API. View in glossary Not documented
Compute Environment Where the AI system runs: on-premise, government cloud, commercial cloud, or edge/device. View in glossary Not documented
Sovereignty Quadrant Classification of data and compute sovereignty: I (Sovereign), II (Federated/Hybrid), III (Cloud with safeguards), or IV (Shared Innovation Zone). View in glossary Not assessed
Hybrid Components NLP pipeline for query intent classification combined with a Large Language Model for response generation, integrated with Bhashini IndicTrans2 translation model for multilingual support across 11 languages.
Data Residency Where the data used by the AI system is stored: domestic, regional, or international. View in glossary Not documented
Cross-Border Transfer Whether data crosses national borders, and if so, whether documented safeguards are in place. View in glossary Not documented
Decision Criticality The rights impact of the decision the AI supports. High criticality requires HITL oversight; moderate requires HOTL; low may operate HOOTL. View in glossary Low
Human Oversight Type Level of human involvement: Human-in-the-Loop (active review), Human-on-the-Loop (monitoring), or Human-out-of-the-Loop (periodic audit). View in glossary HOTL
Development Process Whether the AI system was developed fully in-house, through a mix of in-house and third-party, or fully by an external provider. View in glossary Mix of in-house and third-party
Highest Risk Category The most significant structural risk source identified: data, model, operational, governance, or market/sovereignty risks. View in glossary Model-related risks
Risk Assessment Status Whether a formal risk assessment, informal assessment, or independent audit has been conducted for this system. Not assessed

Risk Dimensions

Market, sovereignty and industry structure risks
Operational and system integration risks

Impact Dimensions

Accountability, transparency and redress
Autonomy, human dignity and due process
Equality, non-discrimination, fairness and inclusion
Systemic and societal
  • Grievance mechanism
  • Human oversight protocol
CategorySensitivityCross-System LinkageAvailabilityKey Constraints
Beneficiary registries and MISPersonalSingle source (no linkage)Currently available and usedPM-KISAN beneficiary records including application status, payment details, and ineligibility status; used to generate personalised responses to 49 distinct query categories
Unstructured and text-based contentPersonalSingle source (no linkage)Currently available and usedFarmer text and voice queries submitted via chatbot interface in 11 languages; processed through NLP and Bhashini translation pipeline

DD News (2023) 'Union Minister Kailash Choudhary launches PM KISAN AI-Chatbot (Kisan e-Mitra) in New Delhi'. New Delhi: DD News. Available at: https://www.newsonair.gov.in/union-minister-kailash-choudhary-launches-pm-kisan-ai-chatbot-kisan-e-mitra-in-new-delhi/ (Accessed: 24 March 2026).

View source News article / media

Press Information Bureau (2023) 'Union Minister of State for Agriculture and Farmers Welfare, Shri Kailash Choudhary Launches AI Chatbot for PM-KISAN Scheme today'. New Delhi: Government of India. Available at: https://pib.gov.in/PressReleaseIframePage.aspx?PRID=1959461 (Accessed: 24 March 2026).

View source Government website / press release

Press Information Bureau (2025) 'KISAN E-MITRA and IoT enabled systems to improve crop productivity'. New Delhi: Government of India. Available at: https://www.pib.gov.in/PressReleasePage.aspx?PRID=2117392 (Accessed: 24 March 2026).

View source Government website / press release

Wadhwani AI (2024) 'Empowering Millions of Indian Farmers with Instant AI Chatbot Support'. Mumbai: Wadhwani AI. Available at: https://www.wadhwaniai.org/empowering-millions-of-indian-farmers-with-instant-ai-chatbot-support/ (Accessed: 24 March 2026).

View source Other

Wadhwani AI (n.d.) 'Kisan-e-Mitra'. Mumbai: Wadhwani AI. Available at: https://www.wadhwaniai.org/impact/agriculture-solutions/kisan-e-mitra/ (Accessed: 24 March 2026).

View source Other
Deployment Status How far the system has progressed into real-world operational use, from concept/exploration through to scaled and institutionalised. View in glossary Full Production Deployment
Year Initiated The year the AI system was first initiated or development began. 2023
Scale / Coverage The scale and geographic or population coverage of the deployment. Nationwide (28 states and 8 Union Territories); over 9.5 million queries resolved; over 290,000 unique farmers served; 20,000+ queries daily
Technical Partners External technology vendors, academic partners, or development partners involved. Wadhwani AI (AI/ML enhancement, February 2024); EkStep Foundation (initial development support); Bhashini / National Language Translation Mission (multilingual translation via IndicTrans2); National Informatics Centre (NIC); Samagra Governance; Google.org (support for Wadhwani AI enhancement).
Outcomes / Results 2.69 million queries answered in 186 days post-February 2024 enhancement (668% increase over manual processing). Over 20,000 queries handled daily as of April 2025. Over 95 lakh (9.5 million) queries resolved as of July 2025. Over 5 million grievances resolved, 290,000+ unique farmers served, 400,000+ farmers enrolled (Wadhwani AI dashboard). Operates 24/7 across 28 states and 8 Union Territories in 11 languages.

How to Cite

DCI AI Hub (2026). 'PM-KISAN Kisan e-Mitra AI Chatbot', AI Hub AI Tracker, case IND-001. Digital Convergence Initiative. Available at: https://socialprotectionai.org/use-case/IND-001 [Accessed: 1 April 2026].

Change History

Updated 1 Apr 2026, 08:11
by system (system)
Created 30 Mar 2026, 08:40
by v2-import (import)