The Artificial Intelligence is no longer an option but an essential component of working in every sphere of life: says Union Minister Dr. Jitendra Singh

1. At a Glance

• Union MoS (IC) S&T Dr. Jitendra Singh, at a BharatGen session (17 Feb 2026), declared AI is no longer optional but essential across every sphere, while lauding India's first government-owned Sovereign LLM / Multilingual AI stack. [1]
• Marks India's transition from being an AI consumer to a sovereign foundational-model producer — directly examinable under GS-III (Sci-Tech) and GS-II (Governance/Policy).
• Anchored on two funding pillars: ₹235 cr under NM-ICPS (DST) for BharatGen + ₹10,372 cr IndiaAI Mission (MeitY). [1][2][3]

2. Why in the News

• 17 Feb 2026: PIB release on Dr. Jitendra Singh's BharatGen session highlighting "whole-of-government" model. [1]
• PM Modi launched BharatGen's Param-2 text foundation model — 17 billion parameters, covering 22 scheduled Indian languages. [1][3]
• Param-2 to be formally showcased at the India AI Impact Summit 2026. [3]

3. Background & Evolution

• 2018: NITI Aayog's National Strategy for Artificial Intelligence ("#AIforAll").
• 2018-19: NM-ICPS launched by DST with ~₹3,660 cr outlay for Technology Innovation Hubs (TIHs); TIH at IIT Bombay hosts BharatGen. [3]
• Mar 2024: Union Cabinet approved the IndiaAI Mission with ₹10,371.92 cr for 5 years under MeitY. [2]
• 2024-25: BharatGen consortium operationalised — IIT Bombay-led, includes IIT Madras, Kanpur, Mandi, Hyderabad, Kharagpur, IIIT Hyderabad, IIIT Delhi, IIM Indore. [3]
• 2025: Release of PARAM-1 (2.9B params); domain models Ayur Param, Agri Param, Legal Param. [3]
• Feb 2026: PARAM-2 (17B, MoE architecture) launched. [1][3]

4. Core Static Facts

• Programme: BharatGen — India's first government-funded, sovereign multilingual & multimodal generative AI initiative. [3]
• Nodal Ministry for BharatGen: Department of Science & Technology (DST) via NM-ICPS. [3]
• Nodal Ministry for IndiaAI Mission: Ministry of Electronics & IT (MeitY). [2]
• Implementing TIH: Technology Innovation Hub, IIT Bombay. [3]
• BharatGen outlay: ₹235 crore under NM-ICPS. [1][3]
• IndiaAI Mission outlay: ₹10,371.92 crore / 5 years. [2]
• IndiaAI Mission's 7 pillars: Compute Capacity, Innovation Centre (IAIC), Datasets Platform, Application Development, FutureSkills, Startup Financing, Safe & Trusted AI. [2]
• GPU achievement: From 10,000 target → 38,000 GPUs empanelled. [2]
• Param-2: 17B parameters, Mixture-of-Experts (MoE), 22 scheduled languages, trained on Bharat Data Sagar datasets. [1][3]

5. Multi-Dimensional Analysis

Scientific / Technological

• Sovereign foundation models reduce dependence on US/China LLMs (GPT, Llama, DeepSeek). [1]
• MoE architecture in Param-2 cuts inference cost vs dense models. [3]

Economic

• ₹10,585 cr aggregate AI public investment seeds startup financing and compute markets. [1]
• Affordable GPU access (~₹67/GPU-hr subsidised under IndiaAI) lowers entry barriers. [2]

Social / Linguistic Inclusion

• Coverage of all 22 Eighth Schedule languages addresses digital divide for non-English users. [1][3]
• Domain Param models target Ayurveda, agriculture, legal aid — frontline citizen sectors. [3]

Ethical / Governance

• "Safe & Trusted AI" pillar institutionalises bias-audit, deepfake, watermarking tools. [2]
• "Whole-of-government" model signals federal-MeitY-DST coordination. [1]

Geopolitical / Strategic

• Sovereign LLM reduces data-extraterritoriality risk; aligns with Global Partnership on AI (GPAI) chairmanship trajectory.
• India AI Impact Summit 2026 positions India in post-Bletchley/Seoul/Paris AI summit series. [3]

6. Recent Developments (last 12-18 months)

• Mar 2024: Cabinet nod for IndiaAI Mission ₹10,371.92 cr. [2]
• 2025: PARAM-1 (2.9B) released; domain LLMs Ayur/Agri/Legal Param. [3]
• 17 Feb 2026: Dr. Jitendra Singh's BharatGen session; PM launches Param-2. [1]
• 2026: IndiaAI Compute crosses 38,000 GPUs; 12 teams shortlisted for indigenous foundational models. [2]

7. Prelims Hooks

• BharatGen is funded under NM-ICPS of DST — NOT directly under MeitY. [3]
• BharatGen's host TIH is at IIT Bombay. [3]
• BharatGen outlay = ₹235 crore. [1]
• IndiaAI Mission outlay = ₹10,371.92 crore over 5 years. [2]
• IndiaAI Mission approved by Cabinet in March 2024. [2]
• Param-2 has 17 billion parameters and is a Mixture-of-Experts model. [1][3]
• Param-2 supports 22 scheduled Indian languages (Eighth Schedule). [1]
• IndiaAI Mission has 7 pillars including "Safe & Trusted AI". [2]
• Indigenous GPU count under IndiaAI = ~38,000 (vs initial 10,000 target). [2]
• Training corpus called Bharat Data Sagar. [3]
• Domain-tuned variants: Ayur Param, Agri Param, Legal Param. [3]
• 12 teams shortlisted for indigenous foundational/large language models. [2]
• NM-ICPS administered by DST, launched in 2018-19.
• Param-2 to be unveiled at India AI Impact Summit 2026. [3]

8. Mains Relevance

• GS-III: Science & Technology — Indigenous tech development, IT & computers; Awareness of AI.
• GS-II: Government policies & interventions; Issues relating to development.
• GS-IV: Ethics — AI bias, accountability, privacy.

Likely question stems:

1. "Sovereign foundational AI models are no longer a luxury but a strategic necessity for India." Examine in the light of BharatGen and the IndiaAI Mission.
2. Discuss how the "whole-of-government" approach to AI — straddling DST's NM-ICPS and MeitY's IndiaAI Mission — addresses both research and deployment gaps.
3. Critically evaluate the ethical and linguistic-inclusion dimensions of indigenous LLMs in India.

9. Related Topics to Study Next

• IndiaAI Mission (7 pillars) — parent policy framework. [2]
• NM-ICPS & Technology Innovation Hubs — DST's CPS umbrella.
• National Strategy for AI, 2018 (NITI Aayog) — foundational policy doc.
• Digital Personal Data Protection Act, 2023 — data layer for AI training.
• GPAI (Global Partnership on AI) — India chaired 2024.
• Bhashini / National Language Translation Mission — complements multilingual LLMs.
• Semicon India Programme — compute hardware base for AI.
• AI Safety Summits (Bletchley 2023, Seoul 2024, Paris 2025, India 2026) — global context.

10. Common Errors / Trap Areas

• Wrong ministry: BharatGen sits under DST/NM-ICPS, not MeitY. The PIB release was issued by MeitY because Dr. Jitendra Singh holds S&T portfolio — easy confusion. [1]
• Outlay mix-up: ₹235 cr is BharatGen-specific; ₹10,372 cr is the broader IndiaAI Mission — do not conflate. [1][2]
• Parameter counts: PARAM-1 = 2.9B; PARAM-2 = 17B (MoE). Don't confuse with C-DAC's older "PARAM" supercomputer series. [3]
• Languages: 22 scheduled languages (Eighth Schedule), not "all Indian languages". [1]
• IndiaAI Mission approval year is 2024 (Cabinet), not 2023. [2]