What India’s AI Safety Institute could do

In October 2024, the Ministry of Electronics and Information Technology (MeitY) convened meetings with experts to discuss the establishment of an AI Safety Institute under the IndiaAI Mission. This initiative comes on the heels of global discussions on AI safety, including Prime Minister Modi's U.S. visit and the United Nations’ Summit of the Future. The rising global focus on AI governance presents India with an opportunity to lead in shaping AI safety practices on the global stage.

India’s Strategic Role in AI Safety

• Building on Global Leadership: India’s leadership at the G20 and the Global Partnership on Artificial Intelligence (GPAI) positions it as a key player in AI governance. India has the potential to unify the global majority’s concerns, especially regarding the human-centric safety of AI, and contribute significantly to global discussions on AI regulation and oversight.
• AI Safety Institute’s Design and Purpose: The proposed AI Safety Institute should focus on building domestic capacity, leveraging India's advantages in AI, and connecting with international initiatives such as the Bletchley Process. It should serve as a non-regulatory technical institution, concentrating on AI risk assessments, safety research, and global collaboration without imposing restrictive regulations on AI deployment.
• Multistakeholder Collaboration in Governance: The United Nations’ Global Digital Compact emphasizes collaboration among governments, businesses, and civil society for AI governance. India’s AI Safety Institute could play a pivotal role in championing these principles, promoting inclusive participation from developing countries, and focusing on the global majority's interests in AI safety.

Institutional Challenges and Considerations

• Learning from Existing Models: India should learn from the AI safety frameworks of the U.S. and U.K., which have already established AI Safety Institutes. These institutes focus on proactive information sharing, risk assessment, and safety testing without taking on regulatory enforcement roles, allowing for open collaboration across stakeholders while ensuring public safety.
• Avoiding Over-Regulation: India should be cautious in its approach and avoid prescriptive regulatory controls similar to the EU and China. Over-regulation stifles innovation, as seen in the EU’s stringent AI regulations. India’s AI Safety Institute should focus on technical research, testing, and standardisation rather than enforcement, thus encouraging proactive industry collaboration.
• International Collaboration: The Bletchley network of AI Safety Institutes, led by countries like the U.S. and U.K., offers an international platform for collaboration. India’s AI Safety Institute should engage with this network, facilitating knowledge exchange and providing a technical backbone for AI governance. The exchange of insights and resources will help India keep pace with the rapid advancements in AI technology.

Charting India’s Approach to AI Safety

• Focus on Technical Research and Testing: The AI Safety Institute should be designed as an independent, non-regulatory body focused on AI safety research and testing. By doing so, India can avoid unnecessary bureaucratic control and ensure that AI technologies are evaluated based on scientific evidence and international best practices.
• Championing Human-Centric Risks: India’s AI Safety Institute could focus on addressing AI risks related to bias, discrimination, social exclusion, privacy, and labor markets. This would allow India to bring a global perspective on the socio-economic and gendered impacts of AI to the forefront of international dialogues.
• Global Stewardship of AI Governance: If successful, India could position itself as a global leader in AI governance, advocating for evidence-based, inclusive, and proportionate policy solutions. The AI Safety Institute could demonstrate India’s commitment to scientific temper and responsible AI deployment, establishing India as a steward of forward-thinking global AI governance.

Practice Question:

Q. Discuss the role of AI Safety Institutes in global AI governance and assess how India can leverage its comparative advantages in AI to contribute to international AI safety discussions. How should India design its AI Safety Institute to balance innovation and safety?