Safe & Trusted AI Pillar under IndiaAI Mission Strengthens Citizen Trust through Secure, Fair and Responsible AI Solutions

Key Highlights:

• The IndiaAI Mission, approved in March 2024 with an outlay of ₹10,371 crore over five years, aims to build a comprehensive AI ecosystem.
• Its Safe & Trusted AI pillar promotes responsible AI development through indigenous governance frameworks, standards, and evaluation mechanisms.
• The AI Safety Institute has been established to advance science-based research on AI safety and governance.
• India AI Governance Guidelines provide a risk-based framework addressing algorithmic bias, misinformation, and deepfakes.
• India is supporting 20 indigenous sovereign AI model proposals, including Sarvam AI's large language models.
• Over 26 lakh individuals have completed the YUVA AI for All programme, and 686 fellowships have been awarded.
• India actively engages with global processes like GPAI, G20, and UN to shape international AI governance norms.

Detailed Insights:

• The IndiaAI Mission comprises seven pillars: IndiaAI Compute, Foundation Models, AIKosh, IndiaAI Application Development Initiative, FutureSkills, Startup Financing, and Safe & Trusted AI.
• Progress includes 93 lakh GPU hours sanctioned for compute, 62 AI prototypes developed, and 27 India Data and AI Labs established.
• 13 Responsible AI projects have been approved in educational institutions, focusing on bias mitigation, machine unlearning, and privacy-preserving AI.
• Key projects include Saakshya for deepfake detection and AI Vishleshak for audio-visual forgery detection.
• NIT Raipur is developing responsible AI algorithms for healthcare, while IIT Delhi is working on privacy-preserving machine learning using federated learning.
• IIT Jodhpur is developing techniques for machine unlearning to remove unwanted information from generative models.
• The India AI Governance Guidelines propose institutional mechanisms like the AI Governance and Economic Group (AIGEG) and Technology and Policy Expert Committee (TPEC).
• India's "AI for All" vision aims for inclusive development while ensuring AI remains safe, trusted, and aligned with socio-economic priorities.
• The India AI Impact Summit 2026 led to New Delhi Frontier AI Impact Commitments and a Guidance Note on AI Governance.

Key Concepts Involved:

• IndiaAI Mission: A comprehensive national program to build an AI ecosystem in India.
• Safe & Trusted AI: A pillar of the IndiaAI Mission focusing on ethical and secure AI development and deployment.
• Responsible AI: The principle of developing and using AI in a way that is fair, transparent, accountable, and beneficial to society.
• AI Safety Institute: An institution established to conduct research and promote safe, secure, and trustworthy AI development.
• India AI Governance Guidelines: A policy framework for managing AI risks, including algorithmic bias and deepfakes.
• Deepfake Detection: Technology used to identify manipulated or synthetic media, particularly audio and video.