IAS/UPSC Coaching Institute  

 Editorial 2: The approach to regulating AI in India

Context

There are different ways to manage and regulate AI, and we can learn a lot from how countries around the world have dealt with data regulations and policies.

 

Introduction

Over the past year, the governance and regulation of AI have gained a lot of global attention. While the focus has recently shifted from social safetyinclusivity, and human rights to prioritizing innovation and economic growth, only a few countries or regions have introduced laws to regulate AI. These include China, the European UnionCanadaSouth KoreaPeru, and the U.S. (although President Donald Trump has now reversed the AI executive order from former President Joe Biden). Several countries, such as the U.K.JapanBrazilCosta RicaColombia, and Pakistan, have draft bills waiting for approval from their legislative bodies.

  • A common approach globally is publishing a policy or strategy document.
  • These documents outline intentions, plans, budgets, and a road map for using AI to boost socio-economic development.
  • They aim to ensure growth is inclusiveethical, and sustainable.
  • Around 85 countries and the African Union have released official National AI Strategy documents.

 

India's Approach to AI Governance

  • Lack of Official National AI Strategy: India does not have an officially approved National AI Strategy document or a specific AI regulation law.
  • Focus on Government Mission: Instead, India has focused on a government mission designed to support AI development and adoption.
  • NITI Aayog's 2018 Document: The 2018 NITI Aayog document titled ‘National Strategy for Artificial Intelligence’ provides comprehensive suggestions but remains a recommendation without formal government approval, implementation plan, or budget.

 

IndiaAI Mission and Its Goals

  • Seven Pillars of IndiaAI Mission:
    • Innovative ecosystem
    • Skilled workforce
    • Safe AI development
    • Trustworthy AI systems
  • Foundational AI Model: Several initiatives, including a foundational AI model, are in development.
  • Advisory Group: An expert advisory group is working to develop governance framework recommendations. However, there is uncertainty about whether these recommendations will become official policies.

 

Benefits and Gaps in India’s Approach

Benefits

  • Flexibility: Ability to adapt plans based on evolving technology, geopolitics, economics, and citizen sentiment.

 

Gaps

  • Lack of Comprehensive Vision: No clear roadmap of India’s visionprioritiescapacity, or accountability mechanisms.
    • Initiatives remain reactive and may not follow a planned trajectory.
    • Potential dependence on individual leadership.

 

AI Adoption in India

  • Global AI Development: AI development is mainly concentrated in U.S.EUU.K., and China.
    However, India is witnessing a rapid rise in AI adoption.
  • Potential Risks: As AI usage grows, concerns about discriminationexclusionunfair outcomescybersecurity risksprivacy breaches, and unequal opportunities are emerging.

 

Current AI Governance and Issues

  • Voluntary Guidelines: Current AI governance in India is largely voluntary and lacks clarity.
  • Lack of Public Awareness: There is little public awareness of algorithmic useefficacy, or evaluation metrics in sectors like bankinginsuranceeducationhealthcare, and public administration.
  • Absence of Civic Discourse: There is limited discussion on issues like algorithmic alignmentmodel evaluationdata provenancelabour market disruptions, and cybersecurity and privacy risks.
  • Concerns from AI-generated Content: India has already seen violence and social harm linked to AI-generated content on social media platforms in recent years.

 

Approaches to AI Governance and Regulation

  • Global Lessons from Data Regulation and Policies:
    • Different countries have adopted varying approaches to AI governance and data protection, offering valuable lessons.

 

AI Governance Approaches by Country

Country

Approach

Characteristics

India

Digital Personal Data Protection (DPDP) Act, 2023

Cross-sectoral, centralised, and comprehensive. Similar to EU's GDPR and China's Personal Information Protection Law.

U.S.

Decentralised and Sector-Specific

Decentralised regulations with sector-specific laws for data protection and privacy.

China

Focused Laws for Specific AI Types and Use Cases

Laws tailored for specific AI technologies (e.g., generative AI) and use cases (e.g., deep synthesis).

  • India's Potential Approach:
    • India could:
      • Adopt one of these existing models.
      • Develop a hybrid model based on the framework established by the DPDP Act, 2023.

 

Short-term Goal: AI Policy for India

  • AI Policy for India:
    • short-term goal for India to:
      • Pilot enforcement tools before formal legislation.
  • Key Areas for AI Policy:

 

Essential Elements for India’s AI Policy

Area

Key Consideration

Possible Approaches

Vision for AI

Defining India’s strategic goals and aspirations for AI.

Clear national goals and targets for AI in India's development.

Capacity Building

Creating infrastructure and resources to support AI development.

Government-led initiatives to build AI ecosystem and workforce.

Policy Implementation

Designating the government authority responsible for AI policy.

Centralised or coordinated agency overseeing AI governance.

Ethical Guidelines

Establishing frameworks for responsible and ethical AI use.

Code of ethics emphasizing fairness, transparency, and accountability.

Priority Sectors

Identifying sectors where AI can drive socio-economic growth.

Focus on healthcare, agriculture, education, and manufacturing.

  • Public Engagement: Urgent need for the government to initiate public discussions on AI use, ensuring broader participation and feedback.

 

Insights from 85 Global AI Policies

Area

Key Consideration

Vision for AI

Defining clear long-term goals for AI development.

Capacity Building

Building infrastructure, resources, and training for AI growth.

Policy Implementation

Identifying and empowering the government body responsible for policy execution.

Ethical Guidelines

Establishing standards for fair, transparent, and accountable AI use.

Priority Sectors

Focusing on economic and social sectors where AI can make a significant impact.

 

Conclusion

In conclusion, India's AI governance approach lacks a formal national strategy and faces gaps in visionaccountability, and public engagement. Drawing lessons from global examples like the EUChina, and the U.S., India must establish a clear AI policy, implement pilot initiatives, and encourage public discourse to ensure responsible AI development and adoption.