PIB TrackerBuilding guardrails for AI: IndiaAI's 'safe and trusted' pillar
The government detailed the Safe & Trusted AI pillar of the ₹10,371-crore IndiaAI Mission — 13 responsible-AI projects, 58 AI Centres of Excellence and 27 data-and-AI labs — as India builds a home-grown AI stack and pushes to shape global AI governance.
What happened
- The government detailed the Safe & Trusted AI pillar of the IndiaAI Mission (approved March 2024).
- The Mission has a ₹10,371-crore outlay over five years across seven pillars.
- 13 responsible-AI projects, 58 AI Centres of Excellence and 27 India Data & AI Labs are running.
- 20 indigenous 'sovereign' model proposals (12 LLMs, 8 SLMs) are being supported.
- The goal is secure, fair, accountable AI and a role in global AI governance.
For Prelims
- IndiaAI Mission: Approved March 2024, outlay ₹10,371 crore over five years to build a full AI ecosystem.
- Seven pillars: Compute, Foundation Models, AIKosh, Application Development, FutureSkills, Startup Financing, Safe & Trusted AI.
- Safe & Trusted AI: 13 responsible-AI projects, 58 AI Centres of Excellence, 27 India Data & AI Labs.
- Sovereign models: 20 indigenous model proposals — 12 LLMs and 8 SLMs.
- DPDP Act, 2023: India's data-protection law, the legal backdrop to trustworthy AI.
- Goal: democratise AI (jobs, opportunity) while managing bias, safety and misuse risks.
For UPSC: A science-tech-and-governance item on AI. Use it for the IndiaAI Mission and sovereign compute/models, responsible and safe AI, the DPDP Act, and India's bid to shape global AI governance.
What it is NOT: This is progress on a policy-and-funding pillar for responsible AI — not a binding AI law or a specific regulation of AI systems.
For Mains
Syllabus: GS3.13 · GS3.12 · Linkage L2
Anchor
Innovation with guardrails — India funds an AI stack and the safeguards to trust it.
Substantiation (data)
The IndiaAI Mission (₹10,371 cr, seven pillars) with 13 responsible-AI projects, 58 CoEs, 27 data-AI labs and 20 sovereign model proposals.
Exemplification
Backing indigenous LLMs/SLMs and safe-AI tooling while positioning India in global AI-governance debates.
Problematisation
Bias, safety, deepfakes and misuse, compute dependence and the pace of technology outstripping regulation are live risks.
Way-forward
Pair sovereign compute and models with enforceable safe-AI standards, DPDP implementation and international coordination.
Position
India should democratise AI's benefits while building credible guardrails — and help write the global rules.
Deploys into: New tech + indigenisation (GS3.13, GS3.12) · the IndiaAI Mission, sovereign compute and models, responsible AI and the DPDP Act.
Ministry of Electronics & IT · 2026-07-27 · PRID 2289946 · PIB source ↗