Opinion
AI regulation: India can look to Japan for a balanced framework
Summary
- The Japanese model of AI regulation places the tech industry at the centre, with the government acting an enabler and Generative AI treated as a public good. The Eastern idea of collective privacy may serve India well.
In 1991, the Indian government articulated a ‘Look East’ policy to build extensive economic and strategic relations with its eastern neighbours and Southeast Asia. This was different from the traditional ‘West facing’ outlook, in which we looked to learn from Europe and North America.
With the age of AI upon us, India is considering how it should both foster and regulate this powerful new technology which can reshape industry, society and geopolitics. AI promises huge economic and social benefits, but there are concerns around copyright, human rights violations, privacy and fairness.
For instance, the issue of deepfakes clouded the recent elections, with fears of democracy being subverted. For regulating AI, should India look West towards the EU and its AI Act, or at the US and UK, which are considering more Big Tech-friendly regimes?
I believe that India should bring back its Look East philosophy and learn from another large democracy, Japan, which has been quietly building an AI regulatory regime that balances the need for integrating AI safely in society with building an environment for innovation. It sees AI as a lever with which it can leapfrog decades of stagnation in its innovation and tech ecosystem. Similarly, AI can offer India a strong impetus to become Viksit Bharat by 2047.
Also read: ‘Social order could collapse’ in AI era, two top Japanese companies say
Japan effectively leveraged the 2023 G7 summit, where the Hiroshima AI process gave a clarion call for exploring various frameworks to regulate AI. It was no coincidence that the summit was in Hiroshima, with its stark reminder on what could happen if technology is misused.
Two basic approaches emerged: a ‘hard law-based’ approach with strict obligations, as evidenced by the EU’s AI Act and Chinese regulation, and a ‘soft law-based’ one, which emphasizes non-binding guidance and principles. Japan leans towards the latter.
Inge Odendaal of Stellenbosch University describes the principles very well in her paper (bit.ly/3Volng1):
The government acts as a facilitator: AI is a complex ecosystem of technologies, and the Japanese government prefers to be a facilitator rather than sole creator of innovation. It acknowledges the private sector as the primary driver and activates its governance institutions to help them.
Japan’s foreign affairs ministry coordinated the Hiroshima process, for example, and its cultural affairs ministry is looking at copyright issues. This is a perfect template for India to adopt. The government should place the private technology sector at the centre of Indian efforts, with enabling roles to be played by its ministries of electronics and information technology, law, and information and broadcasting.
Focused funding and investment: Japan has created a large fund for Generative AI, domestic data centres and local chip production. It offers incentives and subsidies in partnership with cloud computing providers and tech companies.
India has a similar fund, set up as part of the India AI Mission. But the difference is that much of the Indian funding is being used to procure processing units centrally rather than decentralize innovation (like in Japan).
Also read: Japan embraces softened AI regulations to drive economic growth and chip leadership: Report
Effective government-academia-industry collaboration: The AI Japan R&D Network works across universities, research institutes, private sector and global tech companies to recommend policies on AI research and companies. India is doing something similar; we already have the constituents, with our IITs and IISc, a robust private sector and a strong global tech presence.
Building domestic competence: Japan is looking to build its own large language model (LLM); India has a raft of them being built and they would benefit greatly from government support. We also have expertise in building citizen-centric technology with initiatives like Aadhaar and UPI. A similar model can help build GenAI as a public good.
Philosophical argument: Western and Eastern societies differ considerably in their approach to fundamental societal issues like privacy and data usage. In my research at Cambridge University, I realized that the Western idea of privacy as being the ‘right to be left alone’ was quite different from the Eastern notion of ‘collective or societal privacy.’
AI is both transformational and inevitable. As it shapes industry, society, humanity and the global order, the world’s most populous country cannot afford to be left behind. Other countries are racing to set up frameworks, and Indian policymakers need to step up the urgency. As they do so, they would be well served by looking at Japan, which has similar cultural traits, and taking a hard-nosed approach that balances innovation and regulation.
