AI Regulation in India: Policy Framework and Global Stance

Introduction

Artificial Intelligence (AI) has emerged as the defining general-purpose technology of the 21st century, driving structural shifts across global economies, governance models, and security paradigms. The rapid proliferation of Generative AI (GenAI), Large Language Models (LLMs), and autonomous systems has introduced unprecedented opportunities alongside significant systemic risks. For India—the world’s most populous nation and a major hub for global software services—the governance of AI represents a delicate balancing act.

India's strategy toward AI regulation has evolved from an initial "light-touch," market-led approach aimed at fostering innovation to a more structured, risk-aware, and sector-specific framework. Unlike the European Union’s heavily precautionary regime or the United States’ market-driven, self-regulatory framework, India is carving out a distinct "Global South" approach. This framework prioritizes democratizing AI access, leveraging technology for socio-economic inclusion, and mitigating systemic harms such as deepfakes, algorithmic bias, and data privacy violations.

Historical Background / Context

The trajectory of AI policy and governance in India can be traced through several foundational initiatives, whitepapers, and regulatory interventions led by NITI Aayog, the Ministry of Electronics and Information Technology (MeitY), and legislative bodies.

+-----------------------------------------------------------------------------------+
|                            EVOLUTION OF AI POLICY IN INDIA                        |
+-----------------------------------------------------------------------------------+
| 2018: NITI Aayog launches 'National Strategy for Artificial Intelligence' (#AIforAll) |
| 2021: NITI Aayog releases 'Approach Documents for Responsible AI'                  |
| 2023: Enactment of Digital Personal Data Protection (DPDP) Act                    |
| 2023: India assumes Presidency of Global Partnership on Artificial Intelligence   |
| 2024: Cabinet approves Cabinet Approval for 'India AI Mission' (₹10,372 Crore)     |
| 2024+: Transition toward sector-specific regulation and proposed Digital India Act|
+-----------------------------------------------------------------------------------+

1. The 2018 NITI Aayog National Strategy (#AIforAll)

In June 2018, NITI Aayog published the landmark discussion paper titled "National Strategy for Artificial Intelligence: #AIforAll". It identified five core sectors where AI could yield maximum social impact:

• Healthcare: Increasing access and affordability of quality healthcare.
• Agriculture: Enhancing crop yields, pest management, and precision farming.
• Education: Improving access and quality of education through personalized learning.
• Smart Cities and Infrastructure: Optimizing mobility and urban planning.
• Smart Mobility and Transportation: Improving traffic management and autonomous transit solutions.

2. Principles of Responsible AI (2021)

In 2021, NITI Aayog issued a two-part approach paper on "Responsible AI for All". It established foundational principles for ethical AI deployment, emphasizing:

• Non-discrimination and Fairness
• Privacy and Security
• Inclusivity and Equality
• Transparency and Explainability
• Accountability and Human Oversight

3. Shift from Voluntary Guidelines to Regulatory Oversight (2023–2024)

The boom in generative AI tools (such as ChatGPT, Midjourney, and open-source foundation models) exposed vulnerabilities related to synthetic media, copyright infringement, and deepfakes. Consequently, MeitY issued advisories under the Information Technology Act, 2000 and IT Intermediary Guidelines and Digital Media Ethics Code Rules, 2021, signaling a shift toward mandatory compliance for platforms deploying untested or volatile AI models.

Key Features / Objectives of India's Policy Framework

India’s AI policy framework operates on a multi-pronged structure blending public investment, ethical guidelines, data protection legislation, and strategic international diplomacy.

1. The India AI Mission

Approved by the Union Cabinet with an allocation exceeding ₹10,372 Crore, the India AI Mission serves as the operational backbone for indigenous AI capacity building. Its primary pillars include:

• IndiaAI Compute Capacity: Establishing a public-private partnership (PPP) model to deploy over 10,000 High-Performance Computing (HPC) GPUs to provide affordable compute access to startups and researchers.
• IndiaAI Innovation Centre: Facilitating the development of indigenous Large Multimodal Models (LMMs) and domain-specific foundation models.
• IndiaAI Datasets Platform: Constructing a unified, non-personal data repository to simplify access for domestic AI developers.
• IndiaAI FutureSkills: Expanding specialized undergraduate, postgraduate, and PhD programs in AI fields to address skill shortages.
• IndiaAI Startup Financing: Streamlining capital flow to deep-tech AI startups.
• Safe & Trusted AI Pillar: Funding research into ethical AI, bias mitigation tools, and governance protocols.

2. Legislative and Regulatory Anchor Points

While India does not currently possess a single standalone "AI Act," regulation is enforced through a web of existing and upcoming laws:

• Digital Personal Data Protection (DPDP) Act, 2023: Governs the processing of personal data used for training AI models, mandating explicit consent, purpose limitation, and strict penalties for data breaches.
• Information Technology Act, 2000 & IT Rules, 2021: Mandates that intermediaries remove deepfakes, child sexual abuse material (CSAM), and harmful synthetic content within strict timeframes (e.g., 24 to 36 hours).
• Proposed Digital India Act (DIA): Expected to replace the aging IT Act 2000, the DIA aims to establish explicit legal frameworks for high-risk AI applications, algorithmic accountability, and user safety.

3. India's Global Stance: The "Global South" Perspective

India advocates for an inclusive, open-source, and human-centric global AI governance model. Key components of its international stance include:

• GPAI Leadership: As Chair of the Global Partnership on Artificial Intelligence (GPAI) in 2023–2024, India hosted the New Delhi GPAI Summit, promoting democratic principles, sovereign AI capability, and equitable technology sharing.
• Digital Public Infrastructure (DPI) Integration: India champions the integration of AI with DPIs (like Aadhaar, UPI, and ONDC), promoting projects like Bhashini (AI-powered language translation) to eliminate linguistic barriers in governance.
• Rejection of Monopoly Control: India consistently opposes the concentration of advanced AI technologies within a handful of Western tech conglomerates or authoritarian states, pushing for multilateral governance models under United Nations frameworks.

Comparative Analysis: Global Regulatory Approaches

To understand India's position, it is vital to contrast it with other major global regulatory regimes:

Significance for India

1. Socio-Economic Empowerment and Public Service Delivery

By leveraging AI through initiatives like Bhashini, India can deliver government services in 22 official languages, enabling digital inclusion for hundreds of millions of non-English speakers. In healthcare, AI diagnostics bridges the doctor-patient ratio gap in rural areas.

2. Economic Growth and Productivity

According to industry estimates (NASSCOM/McKinsey), AI has the potential to add $450–$500 billion to India's GDP by 2030, transforming traditional sectors such as agriculture, logistics, financial technology, and manufacturing.

3. Establishing Strategic Autonomy ("Sovereign AI")

Dependence on foreign foundational models exposes India to geopolitical risks, data sovereignty vulnerabilities, and cultural/linguistic biases. Building domestic AI compute infrastructure and national LLMs ensures strategic autonomy in critical digital infrastructure.

4. Leadership in Global Governance

By articulating a balanced middle path between over-regulation (EU) and under-regulation (US), India positions itself as a natural leader for developing nations seeking to harness AI without sacrificing domestic security or sovereignty.

Challenges and Concerns

+-----------------------------------------------------------------------------------+
|                             MAJOR CHALLENGES IN AI GOVERNANCE                     |
+-----------------------------------------------------------------------------------+
|  Hardware & Compute Bottlenecks    --> Heavy dependence on foreign hardware (GPUs)|
|  Data Quality & Representation     --> Lack of curated, high-quality local datasets|
|  Regulatory Dilemma                --> Stifling early-stage innovation vs. safety |
|  Deepfakes & Disinformation        --> Threats to social cohesion & elections     |
|  Intellectual Property & Copyright --> Unclear ownership of AI-generated content  |
|  Job Displacement Risks            --> Impact on routine IT/ITeS service jobs     |
+-----------------------------------------------------------------------------------+

1. Compute and Hardware Dependency

India faces a severe deficit in domestic semiconductor manufacturing and advanced GPU infrastructure. The global AI ecosystem is constrained by supply chains heavily reliant on a few entities (e.g., TSMC, NVIDIA). Importing advanced chips remains capital-intensive.

2. Algorithmic Bias and Data Inequities

Most global LLMs are trained on Western, English-centric data sets, leading to systemic bias when applied to the complex socio-cultural fabric of India. Conversely, building indigenous datasets requires balancing training needs with data privacy constraints under the DPDP Act 2023.

3. Risks of Deepfakes and Synthetic Misinformation

India's vibrant democracy faces threats from deepfake videos, manipulated audio, and automated disinformation campaigns used during elections. Regulatory enforcement remains reactive due to the rapid technological evolution of Generative Media.

4. Intellectual Property (IP) and Copyright Ambiguities

Existing Indian IP law (Copyright Act, 1957) does not explicitly address whether training AI models on copyrighted material constitutes "fair deal" or infringement. Furthermore, the legal status of content created solely by autonomous AI systems remains unsettled.

5. Labor Market Disruptions

India’s $250+ billion IT-BPM (Business Process Management) sector employs millions in software maintenance, customer support, and code testing. The automation of routine coding and support functions via GenAI poses medium-to-long-term risks to employment generation.

Conclusion and Way Forward

India’s AI policy journey reflects a realistic understanding that static legal codes cannot keep pace with exponential technologies. The optimal strategy requires dynamic governance built on adaptability, public-private partnerships, and robust technical infrastructure.

Strategic Recommendations:

1. Enact a Risk-Based Digital India Act: India should adopt an agile, risk-tiered classification for AI applications—distinguishing low-risk applications (e.g., spam filters) from high-risk deployment (e.g., biometric identification, critical infrastructure, credit scoring).
2. Establish Regulatory Sandboxes: MeitY and sectoral regulators (RBI, SEBI, TRAI) should expand live regulatory sandboxes, allowing startups to test innovative AI applications under safe, monitored conditions without immediate threat of punitive action.
3. Strengthen "Sovereign AI" Infrastructure: Accelerate the deployment of the India AI Mission's compute pool and incentivize private enterprise investments through Production Linked Incentive (PLI) schemes for AI servers and edge devices.
4. Promote Open Source and Local Language Models: Standardize and open public datasets through the National Data Governance Framework Policy, ensuring domestic LLMs accurately represent India's linguistic and cultural diversity.
5. Global Harmonization: Continue leading multilateral efforts at the UN, G20, and GPAI to establish common safety standards, provenance tagging (watermarking synthetic media), and shared protocols for red-teaming advanced models.

UPSC Prelims Fact File